publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-10-03T17:57:24Z | http://arxiv.org/pdf/2310.02255v3 | http://arxiv.org/abs/2310.02255v3 | Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, Jianfeng Gao | 2024-01-21T03:47:06Z | cs.CV, cs.AI, cs.CL, cs.LG | Large Language Models (LLMs) and Large Multimodal Models (LMMs) exhibit
impressive problem-solving skills in many tasks and domains, but their ability
in mathematical reasoning in visual contexts has not been systematically
studied. To bridge this gap, we present MathVista, a benchmark designed to
combine challenges from diverse mathematical and visual tasks. It consists of
6,141 examples, derived from 28 existing multimodal datasets involving
mathematics and 3 newly created datasets (i.e., IQTest, FunctionQA, and
PaperQA). Completing these tasks requires fine-grained, deep visual
understanding and compositional reasoning, which all state-of-the-art
foundation models find challenging. With MathVista, we have conducted a
comprehensive, quantitative evaluation of 12 prominent foundation models. The
best-performing GPT-4V model achieves an overall accuracy of 49.9%,
substantially outperforming Bard, the second-best performer, by 15.1%. Our
in-depth analysis reveals that the superiority of GPT-4V is mainly attributed
to its enhanced visual perception and mathematical reasoning. However, GPT-4V
still falls short of human performance by 10.4%, as it often struggles to
understand complex figures and perform rigorous reasoning. This significant gap
underscores the critical role that MathVista will play in the development of
general-purpose AI agents capable of tackling mathematically intensive and
visually rich real-world tasks. We further explore the new ability of
self-verification, the application of self-consistency, and the interactive
chatbot capabilities of GPT-4V, highlighting its promising potential for future
research. The project is available at https://mathvista.github.io/. | cs.CV | MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts |
2023-10-03T17:53:09Z | http://arxiv.org/pdf/2310.02249v1 | http://arxiv.org/abs/2310.02249v1 | Ananya Joshi, Raviraj Joshi | 2023-10-03T17:53:09Z | cs.CL, cs.LG | In our increasingly interconnected digital world, social media platforms have
emerged as powerful channels for the dissemination of hate speech and offensive
content. This work delves into the domain of hate speech detection, placing
specific emphasis on three low-resource Indian languages: Bengali, Assamese,
and Gujarati. The challenge is framed as a text classification task, aimed at
discerning whether a tweet contains offensive or non-offensive content.
Leveraging the HASOC 2023 datasets, we fine-tuned pre-trained BERT and SBERT
models to evaluate their effectiveness in identifying hate speech. Our findings
underscore the superiority of monolingual sentence-BERT models, particularly in
the Bengali language, where we achieved the highest ranking. However, the
performance in Assamese and Gujarati languages signifies ongoing opportunities
for enhancement. Our goal is to foster inclusive online spaces by countering
hate speech proliferation. | cs.CL | Harnessing Pre-Trained Sentence Transformers for Offensive Language Detection in Indian Languages |
2023-10-03T17:48:14Z | http://arxiv.org/pdf/2310.02238v2 | http://arxiv.org/abs/2310.02238v2 | Ronen Eldan, Mark Russinovich | 2023-10-04T05:20:19Z | cs.CL, cs.AI | Large language models (LLMs) are trained on massive internet corpora that
often contain copyrighted content. This poses legal and ethical challenges for
the developers and users of these models, as well as the original authors and
publishers. In this paper, we propose a novel technique for unlearning a subset
of the training data from a LLM, without having to retrain it from scratch.
We evaluate our technique on the task of unlearning the Harry Potter books
from the Llama2-7b model (a generative language model recently open-sourced by
Meta). While the model took over 184K GPU-hours to pretrain, we show that in
about 1 GPU hour of finetuning, we effectively erase the model's ability to
generate or recall Harry Potter-related content, while its performance on
common benchmarks (such as Winogrande, Hellaswag, arc, boolq and piqa) remains
almost unaffected. We make our fine-tuned model publicly available on
HuggingFace for community evaluation. To the best of our knowledge, this is the
first paper to present an effective technique for unlearning in generative
language models.
Our technique consists of three main components: First, we use a reinforced
model that is further trained on the target data to identify the tokens that
are most related to the unlearning target, by comparing its logits with those
of a baseline model. Second, we replace idiosyncratic expressions in the target
data with generic counterparts, and leverage the model's own predictions to
generate alternative labels for every token. These labels aim to approximate
the next-token predictions of a model that has not been trained on the target
data. Third, we finetune the model on these alternative labels, which
effectively erases the original text from the model's memory whenever it is
prompted with its context. | cs.CL | Who's Harry Potter? Approximate Unlearning in LLMs |
2023-10-03T17:45:39Z | http://arxiv.org/pdf/2310.02235v1 | http://arxiv.org/abs/2310.02235v1 | Pedro Miguel Moás, Carla Teixeira Lopes | 2023-10-03T17:45:39Z | cs.CL, cs.AI, cs.CY, cs.LG, cs.SI | Wikipedia is the world's largest online encyclopedia, but maintaining article
quality through collaboration is challenging. Wikipedia designed a quality
scale, but with such a manual assessment process, many articles remain
unassessed. We review existing methods for automatically measuring the quality
of Wikipedia articles, identifying and comparing machine learning algorithms,
article features, quality metrics, and used datasets, examining 149 distinct
studies, and exploring commonalities and gaps in them. The literature is
extensive, and the approaches follow past technological trends. However,
machine learning is still not widely used by Wikipedia, and we hope that our
analysis helps future researchers change that reality. | cs.CL | Automatic Quality Assessment of Wikipedia Articles -- A Systematic Literature Review |
2023-10-03T17:37:22Z | http://arxiv.org/pdf/2310.02229v2 | http://arxiv.org/abs/2310.02229v2 | Hangyu Tu, Lifeng Han, Goran Nenadic | 2023-10-08T21:17:54Z | cs.CL, cs.AI | Clinical texts, represented in electronic medical records (EMRs), contain
rich medical information and are essential for disease prediction, personalised
information recommendation, clinical decision support, and medication pattern
mining and measurement. Relation extractions between medication mentions and
temporal information can further help clinicians better understand the
patients' treatment history. To evaluate the performances of deep learning (DL)
and large language models (LLMs) in medication extraction and temporal
relations classification, we carry out an empirical investigation of
\textbf{MedTem} project using several advanced learning structures including
BiLSTM-CRF and CNN-BiLSTM for a clinical domain named entity recognition (NER),
and BERT-CNN for temporal relation extraction (RE), in addition to the
exploration of different word embedding techniques. Furthermore, we also
designed a set of post-processing roles to generate structured output on
medications and the temporal relation. Our experiments show that CNN-BiLSTM
slightly wins the BiLSTM-CRF model on the i2b2-2009 clinical NER task yielding
75.67, 77.83, and 78.17 for precision, recall, and F1 scores using Macro
Average. BERT-CNN model also produced reasonable evaluation scores 64.48,
67.17, and 65.03 for P/R/F1 using Macro Avg on the temporal relation extraction
test set from i2b2-2012 challenges. Code and Tools from MedTem will be hosted
at \url{https://github.com/HECTA-UoM/MedTem} | cs.CL | Extraction of Medication and Temporal Relation from Clinical Text using Neural Language Models |
2023-10-03T17:32:41Z | http://arxiv.org/pdf/2310.02226v3 | http://arxiv.org/abs/2310.02226v3 | Sachin Goyal, Ziwei Ji, Ankit Singh Rawat, Aditya Krishna Menon, Sanjiv Kumar, Vaishnavh Nagarajan | 2024-04-21T03:39:21Z | cs.CL, cs.AI, cs.LG | Language models generate responses by producing a series of tokens in
immediate succession: the $(K+1)^{th}$ token is an outcome of manipulating $K$
hidden vectors per layer, one vector per preceding token. What if instead we
were to let the model manipulate say, $K+10$ hidden vectors, before it outputs
the $(K+1)^{th}$ token? We operationalize this idea by performing training and
inference on language models with a (learnable) $\textit{pause}$ token, a
sequence of which is appended to the input prefix. We then delay extracting the
model's outputs until the last pause token is seen, thereby allowing the model
to process extra computation before committing to an answer. We empirically
evaluate $\textit{pause-training}$ on decoder-only models of 1B and 130M
parameters with causal pretraining on C4, and on downstream tasks covering
reasoning, question-answering, general understanding and fact recall. Our main
finding is that inference-time delays show gains when the model is both
pre-trained and finetuned with delays. For the 1B model, we witness gains on 8
of 9 tasks, most prominently, a gain of $18\%$ EM score on the QA task of
SQuAD, $8\%$ on CommonSenseQA and $1\%$ accuracy on the reasoning task of
GSM8k. Our work raises a range of conceptual and practical future research
questions on making delayed next-token prediction a widely applicable new
paradigm. | cs.CL | Think before you speak: Training Language Models With Pause Tokens |
2023-10-03T17:30:33Z | http://arxiv.org/pdf/2310.02224v1 | http://arxiv.org/abs/2310.02224v1 | Yang Chen, Ethan Mendes, Sauvik Das, Wei Xu, Alan Ritter | 2023-10-03T17:30:33Z | cs.CL | Large multimodal language models have proven transformative in numerous
applications. However, these models have been shown to memorize and leak
pre-training data, raising serious user privacy and information security
concerns. While data leaks should be prevented, it is also crucial to examine
the trade-off between the privacy protection and model utility of proposed
approaches. In this paper, we introduce PrivQA -- a multimodal benchmark to
assess this privacy/utility trade-off when a model is instructed to protect
specific categories of personal information in a simulated scenario. We also
propose a technique to iteratively self-moderate responses, which significantly
improves privacy. However, through a series of red-teaming experiments, we find
that adversaries can also easily circumvent these protections with simple
jailbreaking methods through textual and/or image inputs. We believe PrivQA has
the potential to support the development of new models with improved privacy
protections, as well as the adversarial robustness of these protections. We
release the entire PrivQA dataset at https://llm-access-control.github.io/. | cs.CL | Can Language Models be Instructed to Protect Personal Information? |
2023-10-03T17:06:52Z | http://arxiv.org/pdf/2310.02207v3 | http://arxiv.org/abs/2310.02207v3 | Wes Gurnee, Max Tegmark | 2024-03-04T18:25:29Z | cs.LG, cs.AI, cs.CL | The capabilities of large language models (LLMs) have sparked debate over
whether such systems just learn an enormous collection of superficial
statistics or a set of more coherent and grounded representations that reflect
the real world. We find evidence for the latter by analyzing the learned
representations of three spatial datasets (world, US, NYC places) and three
temporal datasets (historical figures, artworks, news headlines) in the Llama-2
family of models. We discover that LLMs learn linear representations of space
and time across multiple scales. These representations are robust to prompting
variations and unified across different entity types (e.g. cities and
landmarks). In addition, we identify individual "space neurons" and "time
neurons" that reliably encode spatial and temporal coordinates. While further
investigation is needed, our results suggest modern LLMs learn rich
spatiotemporal representations of the real world and possess basic ingredients
of a world model. | cs.LG | Language Models Represent Space and Time |
2023-10-03T16:39:20Z | http://arxiv.org/pdf/2310.18322v1 | http://arxiv.org/abs/2310.18322v1 | Nutchanon Yongsatianchot, Tobias Thejll-Madsen, Stacy Marsella | 2023-10-03T16:39:20Z | cs.CL, cs.AI | Large Language Models (LLM) have recently been shown to perform well at
various tasks from language understanding, reasoning, storytelling, and
information search to theory of mind. In an extension of this work, we explore
the ability of GPT-4 to solve tasks related to emotion prediction. GPT-4
performs well across multiple emotion tasks; it can distinguish emotion
theories and come up with emotional stories. We show that by prompting GPT-4 to
identify key factors of an emotional experience, it is able to manipulate the
emotional intensity of its own stories. Furthermore, we explore GPT-4's ability
on reverse appraisals by asking it to predict either the goal, belief, or
emotion of a person using the other two. In general, GPT-4 can make the correct
inferences. We suggest that LLMs could play an important role in affective
modeling; however, they will not fully replace works that attempt to model the
mechanisms underlying emotion-related processes. | cs.CL | What's Next in Affective Modeling? Large Language Models |
2023-10-03T16:34:47Z | http://arxiv.org/pdf/2310.04450v1 | http://arxiv.org/abs/2310.04450v1 | Nutchanon Yongsatianchot, Parisa Ghanad Torshizi, Stacy Marsella | 2023-10-03T16:34:47Z | cs.CL, cs.AI | Large Language Models (LLM) like ChatGPT have significantly advanced in
recent years and are now being used by the general public. As more people
interact with these systems, improving our understanding of these black box
models is crucial, especially regarding their understanding of human
psychological aspects. In this work, we investigate their emotion perception
through the lens of appraisal and coping theory using the Stress and Coping
Process Questionaire (SCPQ). SCPQ is a validated clinical instrument consisting
of multiple stories that evolve over time and differ in key appraisal variables
such as controllability and changeability. We applied SCPQ to three recent LLMs
from OpenAI, davinci-003, ChatGPT, and GPT-4 and compared the results with
predictions from the appraisal theory and human data. The results show that
LLMs' responses are similar to humans in terms of dynamics of appraisal and
coping, but their responses did not differ along key appraisal dimensions as
predicted by the theory and data. The magnitude of their responses is also
quite different from humans in several variables. We also found that GPTs can
be quite sensitive to instruction and how questions are asked. This work adds
to the growing literature evaluating the psychological aspects of LLMs and
helps enrich our understanding of the current models. | cs.CL | Investigating Large Language Models' Perception of Emotion Using Appraisal Theory |
2023-10-03T16:08:41Z | http://arxiv.org/pdf/2310.02174v5 | http://arxiv.org/abs/2310.02174v5 | Qiming Xie, Zengzhi Wang, Yi Feng, Rui Xia | 2024-06-11T15:22:07Z | cs.CL, cs.AI, cs.LG | We observe that current conversational language models often waver in their
judgments when faced with follow-up questions, even if the original judgment
was correct. This wavering presents a significant challenge for generating
reliable responses and building user trust. To comprehensively assess this
issue, we introduce a \textsc{Follow-up Questioning Mechanism} along with two
metrics to quantify this inconsistency, confirming its widespread presence in
current language models. To mitigate this issue, we explore various prompting
strategies for closed-source models; moreover, we develop a training-based
framework \textsc{Unwavering-FQ} that teaches language models to maintain their
originally correct judgments through synthesized high-quality preference data.
Our experimental results confirm the effectiveness of our framework and its
ability to enhance the general capabilities of models. | cs.CL | Ask Again, Then Fail: Large Language Models' Vacillations in Judgment |
2023-10-03T16:05:48Z | http://arxiv.org/pdf/2310.02170v1 | http://arxiv.org/abs/2310.02170v1 | Zijun Liu, Yanzhe Zhang, Peng Li, Yang Liu, Diyi Yang | 2023-10-03T16:05:48Z | cs.CL, cs.AI, cs.MA | Large language model (LLM) agents have been shown effective on a wide range
of tasks, and by ensembling multiple LLM agents, their performances could be
further improved. Existing approaches employ a fixed set of agents to interact
with each other in a static architecture, which limits their generalizability
to various tasks and requires strong human prior in designing these agents. In
this work, we propose to construct a strategic team of agents communicating in
a dynamic interaction architecture based on the task query. Specifically, we
build a framework named Dynamic LLM-Agent Network ($\textbf{DyLAN}$) for
LLM-agent collaboration on complicated tasks like reasoning and code
generation. DyLAN enables agents to interact for multiple rounds in a dynamic
architecture with inference-time agent selection and an early-stopping
mechanism to improve performance and efficiency. We further design an automatic
agent team optimization algorithm based on an unsupervised metric termed
$\textit{Agent Importance Score}$, enabling the selection of best agents based
on the contribution each agent makes. Empirically, we demonstrate that DyLAN
performs well in both reasoning and code generation tasks with reasonable
computational cost. DyLAN achieves 13.0% and 13.3% improvement on MATH and
HumanEval, respectively, compared to a single execution on GPT-35-turbo. On
specific subjects of MMLU, agent team optimization in DyLAN increases accuracy
by up to 25.0%. | cs.CL | Dynamic LLM-Agent Network: An LLM-agent Collaboration Framework with Agent Team Optimization |
2023-10-03T16:02:36Z | http://arxiv.org/pdf/2310.02168v3 | http://arxiv.org/abs/2310.02168v3 | Shengyu Mao, Xiaohan Wang, Mengru Wang, Yong Jiang, Pengjun Xie, Fei Huang, Ningyu Zhang | 2024-04-06T14:32:40Z | cs.CL, cs.AI, cs.CY, cs.LG, cs.MA | This paper introduces an innovative task focused on editing the personality
traits of Large Language Models (LLMs). This task seeks to adjust the models'
responses to opinion-related questions on specified topics since an
individual's personality often manifests in the form of their expressed
opinions, thereby showcasing different personality traits. Specifically, we
construct a new benchmark dataset PersonalityEdit to address this task. Drawing
on the theory in Social Psychology, we isolate three representative traits,
namely Neuroticism, Extraversion, and Agreeableness, as the foundation for our
benchmark. We then gather data using GPT-4, generating responses that not only
align with a specified topic but also embody the targeted personality trait. We
conduct comprehensive experiments involving various baselines and discuss the
representation of personality behavior in LLMs. Our intriguing findings uncover
potential challenges of the proposed task, illustrating several remaining
issues. We anticipate that our work can provide the NLP community with
insights. Code and datasets are available at
https://github.com/zjunlp/EasyEdit. | cs.CL | Editing Personality for Large Language Models |
2023-10-03T15:57:00Z | http://arxiv.org/pdf/2310.02166v1 | http://arxiv.org/abs/2310.02166v1 | Mikhail Salnikov, Hai Le, Prateek Rajput, Irina Nikishina, Pavel Braslavski, Valentin Malykh, Alexander Panchenko | 2023-10-03T15:57:00Z | cs.CL | Recently, it has been shown that the incorporation of structured knowledge
into Large Language Models significantly improves the results for a variety of
NLP tasks. In this paper, we propose a method for exploring pre-trained
Text-to-Text Language Models enriched with additional information from
Knowledge Graphs for answering factoid questions. More specifically, we propose
an algorithm for subgraphs extraction from a Knowledge Graph based on question
entities and answer candidates. Then, we procure easily interpreted information
with Transformer-based models through the linearization of the extracted
subgraphs. Final re-ranking of the answer candidates with the extracted
information boosts Hits@1 scores of the pre-trained text-to-text language
models by 4-6%. | cs.CL | Large Language Models Meet Knowledge Graphs to Answer Factoid Questions |
2023-10-03T15:10:46Z | http://arxiv.org/pdf/2310.02129v5 | http://arxiv.org/abs/2310.02129v5 | Zhoubo Li, Ningyu Zhang, Yunzhi Yao, Mengru Wang, Xi Chen, Huajun Chen | 2024-05-10T18:35:48Z | cs.CL, cs.AI, cs.CV, cs.DB, cs.LG | As the cost associated with fine-tuning Large Language Models (LLMs)
continues to rise, recent research efforts have pivoted towards developing
methodologies to edit implicit knowledge embedded within LLMs. Yet, there's
still a dark cloud lingering overhead -- will knowledge editing trigger
butterfly effect? since it is still unclear whether knowledge editing might
introduce side effects that pose potential risks or not. This paper pioneers
the investigation into the potential pitfalls associated with knowledge editing
for LLMs. To achieve this, we introduce new benchmark datasets and propose
innovative evaluation metrics. Our results underline two pivotal concerns: (1)
Knowledge Conflict: Editing groups of facts that logically clash can magnify
the inherent inconsistencies in LLMs-a facet neglected by previous methods. (2)
Knowledge Distortion: Altering parameters with the aim of editing factual
knowledge can irrevocably warp the innate knowledge structure of LLMs.
Experimental results vividly demonstrate that knowledge editing might
inadvertently cast a shadow of unintended consequences on LLMs, which warrant
attention and efforts for future works. Code and data are available at
https://github.com/zjunlp/PitfallsKnowledgeEditing. | cs.CL | Unveiling the Pitfalls of Knowledge Editing for Large Language Models |
2023-10-03T15:05:52Z | http://arxiv.org/pdf/2310.02124v3 | http://arxiv.org/abs/2310.02124v3 | Jintian Zhang, Xin Xu, Ningyu Zhang, Ruibo Liu, Bryan Hooi, Shumin Deng | 2024-05-27T11:12:45Z | cs.CL, cs.AI, cs.CY, cs.LG, cs.MA | As Natural Language Processing (NLP) systems are increasingly employed in
intricate social environments, a pressing query emerges: Can these NLP systems
mirror human-esque collaborative intelligence, in a multi-agent society
consisting of multiple large language models (LLMs)? This paper probes the
collaboration mechanisms among contemporary NLP systems by melding practical
experiments with theoretical insights. We fabricate four unique `societies'
comprised of LLM agents, where each agent is characterized by a specific
`trait' (easy-going or overconfident) and engages in collaboration with a
distinct `thinking pattern' (debate or reflection). Through evaluating these
multi-agent societies on three benchmark datasets, we discern that certain
collaborative strategies not only outshine previous top-tier approaches, but
also optimize efficiency (using fewer API tokens). Moreover, our results
further illustrate that LLM agents manifest human-like social behaviors, such
as conformity and consensus reaching, mirroring foundational social psychology
theories. In conclusion, we integrate insights from social psychology to
contextualize the collaboration of LLM agents, inspiring further investigations
into the collaboration mechanism for LLMs. We commit to sharing our code and
datasets\footnote{\url{https://github.com/zjunlp/MachineSoM}.}, hoping to
catalyze further research in this promising avenue. | cs.CL | Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View |
2023-10-03T14:59:35Z | http://arxiv.org/pdf/2310.02118v2 | http://arxiv.org/abs/2310.02118v2 | Rafael Ferreira, Diogo Tavares, Diogo Silva, Rodrigo Valério, João Bordalo, Inês Simões, Vasco Ramos, David Semedo, João Magalhães | 2024-01-22T14:41:43Z | cs.CL, cs.AI, I.2.7 | In this report, we describe the vision, challenges, and scientific
contributions of the Task Wizard team, TWIZ, in the Alexa Prize TaskBot
Challenge 2022. Our vision, is to build TWIZ bot as an helpful, multimodal,
knowledgeable, and engaging assistant that can guide users towards the
successful completion of complex manual tasks. To achieve this, we focus our
efforts on three main research questions: (1) Humanly-Shaped Conversations, by
providing information in a knowledgeable way; (2) Multimodal Stimulus, making
use of various modalities including voice, images, and videos; and (3)
Zero-shot Conversational Flows, to improve the robustness of the interaction to
unseen scenarios. TWIZ is an assistant capable of supporting a wide range of
tasks, with several innovative features such as creative cooking, video
navigation through voice, and the robust TWIZ-LLM, a Large Language Model
trained for dialoguing about complex manual tasks. Given ratings and feedback
provided by users, we observed that TWIZ bot is an effective and robust system,
capable of guiding users through tasks while providing several multimodal
stimuli. | cs.CL | TWIZ-v2: The Wizard of Multimodal Conversational-Stimulus |
2023-10-03T14:51:34Z | http://arxiv.org/pdf/2310.02107v4 | http://arxiv.org/abs/2310.02107v4 | Saurabh Srivastava, Chengyue Huang, Weiguo Fan, Ziyu Yao | 2024-06-11T19:00:03Z | cs.CL | Large language models (LLMs) have revolutionized zero-shot task performance,
mitigating the need for task-specific annotations while enhancing task
generalizability. Despite its advancements, current methods using trigger
phrases such as "Let's think step by step" remain limited. This study
introduces PRomPTed, an approach that optimizes the zero-shot prompts for
individual task instances following an innovative manner of "LLMs in the loop".
Our comprehensive evaluation across 13 datasets and 10 task types based on
GPT-4 reveals that PRomPTed significantly outperforms both the naive zero-shot
approaches and a strong baseline (i.e., "Output Refinement") which refines the
task output instead of the input prompt. Our experimental results also
confirmed the generalization of this advantage to the relatively weaker
GPT-3.5. Even more intriguingly, we found that leveraging GPT-3.5 to rewrite
prompts for the stronger GPT-4 not only matches but occasionally exceeds the
efficacy of using GPT-4 as the prompt rewriter. Our research thus presents a
huge value in not only enhancing zero-shot LLM performance but also potentially
enabling supervising LLMs with their weaker counterparts, a capability
attracting much interest recently. Finally, our additional experiments confirm
the generalization of the advantages to open-source LLMs such as Mistral 7B and
Mixtral 8x7B. | cs.CL | Instances Need More Care: Rewriting Prompts for Instances with LLMs in the Loop Yields Better Zero-Shot Performance |
2023-10-03T14:35:27Z | http://arxiv.org/pdf/2310.05976v1 | http://arxiv.org/abs/2310.05976v1 | Reiji Suzuki, Takaya Arita | 2023-10-03T14:35:27Z | physics.soc-ph, cs.AI, cs.CL, cs.GT, cs.MA, cs.NE, q-bio.PE, 68W50, 91A22 | This paper aims to shed light on the evolutionary dynamics of diverse and
social populations by introducing the rich expressiveness of generative models
into the trait expression of social agent-based evolutionary models.
Specifically, we focus on the evolution of personality traits in the context of
a game-theoretic relationship as a situation in which inter-individual
interests exert strong selection pressures. We construct an agent model in
which linguistic descriptions of personality traits related to cooperative
behavior are used as genes. The deterministic strategies extracted from Large
Language Model (LLM) that make behavioral decisions based on these personality
traits are used as behavioral traits. The population is evolved according to
selection based on average payoff and mutation of genes by asking LLM to
slightly modify the parent gene toward cooperative or selfish. Through
preliminary experiments and analyses, we clarify that such a model can indeed
exhibit the evolution of cooperative behavior based on the diverse and
higher-order representation of personality traits. We also observed the
repeated intrusion of cooperative and selfish personality traits through
changes in the expression of personality traits, and found that the emerging
words in the evolved gene well reflected the behavioral tendency of its
personality in terms of their semantics. | physics.soc-ph | An evolutionary model of personality traits related to cooperative behavior using a large language model |
2023-10-03T14:13:36Z | http://arxiv.org/pdf/2310.02071v4 | http://arxiv.org/abs/2310.02071v4 | Liang Chen, Yichi Zhang, Shuhuai Ren, Haozhe Zhao, Zefan Cai, Yuchi Wang, Peiyi Wang, Tianyu Liu, Baobao Chang | 2023-11-28T11:23:14Z | cs.AI, cs.CL, cs.CV, cs.RO | In this study, we explore the potential of Multimodal Large Language Models
(MLLMs) in improving embodied decision-making processes for agents. While Large
Language Models (LLMs) have been widely used due to their advanced reasoning
skills and vast world knowledge, MLLMs like GPT4-Vision offer enhanced visual
understanding and reasoning capabilities. We investigate whether
state-of-the-art MLLMs can handle embodied decision-making in an end-to-end
manner and whether collaborations between LLMs and MLLMs can enhance
decision-making. To address these questions, we introduce a new benchmark
called PCA-EVAL, which evaluates embodied decision-making from the perspectives
of Perception, Cognition, and Action. Additionally, we propose HOLMES, a
multi-agent cooperation framework that allows LLMs to leverage MLLMs and APIs
to gather multimodal information for informed decision-making. We compare
end-to-end embodied decision-making and HOLMES on our benchmark and find that
the GPT4-Vision model demonstrates strong end-to-end embodied decision-making
abilities, outperforming GPT4-HOLMES in terms of average decision accuracy
(+3%). However, this performance is exclusive to the latest GPT4-Vision model,
surpassing the open-source state-of-the-art MLLM by 26%. Our results indicate
that powerful MLLMs like GPT4-Vision hold promise for decision-making in
embodied agents, offering new avenues for MLLM research. Code and data are open
at https://github.com/pkunlp-icler/PCA-EVAL/. | cs.AI | Towards End-to-End Embodied Decision Making via Multi-modal Large Language Model: Explorations with GPT4-Vision and Beyond |
2023-10-03T13:51:01Z | http://arxiv.org/pdf/2310.02053v1 | http://arxiv.org/abs/2310.02053v1 | Chunliu Wang, Rik van Noord, Johan Bos | 2023-10-03T13:51:01Z | cs.CL | A bare meaning representation can be expressed in various ways using natural
language, depending on how the information is structured on the surface level.
We are interested in finding ways to control topic-focus articulation when
generating text from meaning. We focus on distinguishing active and passive
voice for sentences with transitive verbs. The idea is to add pragmatic
information such as topic to the meaning representation, thereby forcing either
active or passive voice when given to a natural language generation system. We
use graph neural models because there is no explicit information about word
order in a meaning represented by a graph. We try three different methods for
topic-focus articulation (TFA) employing graph neural models for a
meaning-to-text generation task. We propose a novel encoding strategy about
node aggregation in graph neural models, which instead of traditional encoding
by aggregating adjacent node information, learns node representations by using
depth-first search. The results show our approach can get competitive
performance with state-of-art graph models on general text generation, and lead
to significant improvements on the task of active-passive conversion compared
to traditional adjacency-based aggregation strategies. Different types of TFA
can have a huge impact on the performance of the graph models. | cs.CL | Controlling Topic-Focus Articulation in Meaning-to-Text Generation using Graph Neural Networks |
2023-10-03T13:43:50Z | http://arxiv.org/pdf/2310.02050v1 | http://arxiv.org/abs/2310.02050v1 | Hao Zhang, Nianwen Si, Yaqi Chen, Wenlin Zhang, Xukui Yang, Dan Qu, Xiaolin Jiao | 2023-10-03T13:43:50Z | cs.CL, cs.CV | With the emergence of large language models (LLMs), multimodal models based
on LLMs have demonstrated significant potential. Models such as LLaSM, X-LLM,
and SpeechGPT exhibit an impressive ability to comprehend and generate human
instructions. However, their performance often falters when faced with complex
tasks like end-to-end speech translation (E2E-ST), a cross-language and
cross-modal translation task. In comparison to single-modal models, multimodal
models lag behind in these scenarios. This paper introduces LST, a Large
multimodal model designed to excel at the E2E-ST task. LST consists of a speech
frontend, an adapter, and a LLM backend. The training of LST consists of two
stages: (1) Modality adjustment, where the adapter is tuned to align speech
representation with text embedding space, and (2) Downstream task fine-tuning,
where both the adapter and LLM model are trained to optimize performance on the
E2EST task. Experimental results on the MuST-C speech translation benchmark
demonstrate that LST-13B achieves BLEU scores of 30.39/41.55/35.33 on
En-De/En-Fr/En-Es language pairs, surpassing previous models and establishing a
new state-of-the-art. Additionally, we conduct an in-depth analysis of
single-modal model selection and the impact of training strategies, which lays
the foundation for future research. We will open up our code and models after
review. | cs.CL | Tuning Large language model for End-to-end Speech Translation |
2023-10-03T13:31:28Z | http://arxiv.org/pdf/2310.02040v2 | http://arxiv.org/abs/2310.02040v2 | Devrim Cavusoglu, Secil Sen, Ulas Sert, Sinan Altinuc | 2024-05-20T15:40:07Z | cs.CL, cs.AI, I.2.7; D.1.3 | Evaluation plays a critical role in deep learning as a fundamental block of
any prediction-based system. However, the vast number of Natural Language
Processing (NLP) tasks and the development of various metrics have led to
challenges in evaluating different systems with different metrics. To address
these challenges, we introduce jury, a toolkit that provides a unified
evaluation framework with standardized structures for performing evaluation
across different tasks and metrics. The objective of jury is to standardize and
improve metric evaluation for all systems and aid the community in overcoming
the challenges in evaluation. Since its open-source release, jury has reached a
wide audience and is available at https://github.com/obss/jury. | cs.CL | Jury: A Comprehensive Evaluation Toolkit |
2023-10-03T13:17:35Z | http://arxiv.org/pdf/2310.02031v7 | http://arxiv.org/abs/2310.02031v7 | Zhen Bi, Ningyu Zhang, Yida Xue, Yixin Ou, Daxiong Ji, Guozhou Zheng, Huajun Chen | 2024-05-23T17:04:07Z | cs.CL, cs.AI, cs.CE, cs.LG, cs.RO | Ocean science, which delves into the oceans that are reservoirs of life and
biodiversity, is of great significance given that oceans cover over 70% of our
planet's surface. Recently, advances in Large Language Models (LLMs) have
transformed the paradigm in science. Despite the success in other domains,
current LLMs often fall short in catering to the needs of domain experts like
oceanographers, and the potential of LLMs for ocean science is under-explored.
The intrinsic reasons are the immense and intricate nature of ocean data as
well as the necessity for higher granularity and richness in knowledge. To
alleviate these issues, we introduce OceanGPT, the first-ever large language
model in the ocean domain, which is expert in various ocean science tasks. We
also propose OceanGPT, a novel framework to automatically obtain a large volume
of ocean domain instruction data, which generates instructions based on
multi-agent collaboration. Additionally, we construct the first oceanography
benchmark, OceanBench, to evaluate the capabilities of LLMs in the ocean
domain. Though comprehensive experiments, OceanGPT not only shows a higher
level of knowledge expertise for oceans science tasks but also gains
preliminary embodied intelligence capabilities in ocean technology. | cs.CL | OceanGPT: A Large Language Model for Ocean Science Tasks |
2023-10-03T12:03:06Z | http://arxiv.org/pdf/2310.01991v1 | http://arxiv.org/abs/2310.01991v1 | Aniruddha Deb, Neeva Oza, Sarthak Singla, Dinesh Khandelwal, Dinesh Garg, Parag Singla | 2023-10-03T12:03:06Z | cs.CL, cs.AI, cs.LG, I.2.3 | While forward reasoning (i.e. find the answer given the question) has been
explored extensively in the recent literature, backward reasoning is relatively
unexplored. We examine the backward reasoning capabilities of LLMs on Math Word
Problems (MWPs): given a mathematical question and its answer, with some
details omitted from the question, can LLMs effectively retrieve the missing
information?
In this paper, we formally define the backward reasoning task on math word
problems and modify three datasets to evaluate this task: GSM8k, SVAMP and
MultiArith. Our findings show a significant drop in the accuracy of models on
backward reasoning compared to forward reasoning across four SOTA LLMs (GPT4,
GPT3.5, PaLM-2, and LLaMa-2). Utilizing the specific format of this task, we
propose three novel techniques that improve performance: Rephrase reformulates
the given problem into a forward reasoning problem, PAL-Tools combines the idea
of Program-Aided LLMs to produce a set of equations that can be solved by an
external solver, and Check your Work exploits the availability of natural
verifier of high accuracy in the forward direction, interleaving solving and
verification steps. Finally, realizing that each of our base methods correctly
solves a different set of problems, we propose a novel Bayesian formulation for
creating an ensemble over these base methods aided by a verifier to further
boost the accuracy by a significant margin. Extensive experimentation
demonstrates that our techniques successively improve the performance of LLMs
on the backward reasoning task, with the final ensemble-based method resulting
in a substantial performance gain compared to the raw LLMs with standard
prompting techniques such as chain-of-thought. | cs.CL | Fill in the Blank: Exploring and Enhancing LLM Capabilities for Backward Reasoning in Math Word Problems |
2023-10-03T11:11:55Z | http://arxiv.org/pdf/2310.01960v1 | http://arxiv.org/abs/2310.01960v1 | Anastasia Kritharoula, Maria Lymperaiou, Giorgos Stamou | 2023-10-03T11:11:55Z | cs.CL, cs.AI | Visual Word Sense Disambiguation (VWSD) is a novel challenging task that lies
between linguistic sense disambiguation and fine-grained multimodal retrieval.
The recent advancements in the development of visiolinguistic (VL) transformers
suggest some off-the-self implementations with encouraging results, which
however we argue that can be further improved. To this end, we propose some
knowledge-enhancement techniques towards improving the retrieval performance of
VL transformers via the usage of Large Language Models (LLMs) as Knowledge
Bases. More specifically, knowledge stored in LLMs is retrieved with the help
of appropriate prompts in a zero-shot manner, achieving performance
advancements. Moreover, we convert VWSD to a purely textual question-answering
(QA) problem by considering generated image captions as multiple-choice
candidate answers. Zero-shot and few-shot prompting strategies are leveraged to
explore the potential of such a transformation, while Chain-of-Thought (CoT)
prompting in the zero-shot setting is able to reveal the internal reasoning
steps an LLM follows to select the appropriate candidate. In total, our
presented approach is the first one to analyze the merits of exploiting
knowledge stored in LLMs in different ways to solve WVSD. | cs.CL | Language Models as Knowledge Bases for Visual Word Sense Disambiguation |
2023-10-03T11:05:14Z | http://arxiv.org/pdf/2310.01957v2 | http://arxiv.org/abs/2310.01957v2 | Long Chen, Oleg Sinavski, Jan Hünermann, Alice Karnsund, Andrew James Willmott, Danny Birch, Daniel Maund, Jamie Shotton | 2023-10-13T21:59:27Z | cs.RO, cs.AI, cs.CL, cs.CV | Large Language Models (LLMs) have shown promise in the autonomous driving
sector, particularly in generalization and interpretability. We introduce a
unique object-level multimodal LLM architecture that merges vectorized numeric
modalities with a pre-trained LLM to improve context understanding in driving
situations. We also present a new dataset of 160k QA pairs derived from 10k
driving scenarios, paired with high quality control commands collected with RL
agent and question answer pairs generated by teacher LLM (GPT-3.5). A distinct
pretraining strategy is devised to align numeric vector modalities with static
LLM representations using vector captioning language data. We also introduce an
evaluation metric for Driving QA and demonstrate our LLM-driver's proficiency
in interpreting driving scenarios, answering questions, and decision-making.
Our findings highlight the potential of LLM-based driving action generation in
comparison to traditional behavioral cloning. We make our benchmark, datasets,
and model available for further exploration. | cs.RO | Driving with LLMs: Fusing Object-Level Vector Modality for Explainable Autonomous Driving |
2023-10-03T10:13:36Z | http://arxiv.org/pdf/2310.01929v2 | http://arxiv.org/abs/2310.01929v2 | Mor Ventura, Eyal Ben-David, Anna Korhonen, Roi Reichart | 2023-11-29T15:11:02Z | cs.CL, cs.AI, cs.LG | Text-To-Image (TTI) models, such as DALL-E and StableDiffusion, have
demonstrated remarkable prompt-based image generation capabilities.
Multilingual encoders may have a substantial impact on the cultural agency of
these models, as language is a conduit of culture. In this study, we explore
the cultural perception embedded in TTI models by characterizing culture across
three hierarchical tiers: cultural dimensions, cultural domains, and cultural
concepts. Based on this ontology, we derive prompt templates to unlock the
cultural knowledge in TTI models, and propose a comprehensive suite of
evaluation techniques, including intrinsic evaluations using the CLIP space,
extrinsic evaluations with a Visual-Question-Answer (VQA) model and human
assessments, to evaluate the cultural content of TTI-generated images. To
bolster our research, we introduce the CulText2I dataset, derived from four
diverse TTI models and spanning ten languages. Our experiments provide insights
regarding Do, What, Which and How research questions about the nature of
cultural encoding in TTI models, paving the way for cross-cultural applications
of these models. | cs.CL | Navigating Cultural Chasms: Exploring and Unlocking the Cultural POV of Text-To-Image Models |
2023-10-03T09:46:02Z | http://arxiv.org/pdf/2310.01917v2 | http://arxiv.org/abs/2310.01917v2 | Iva Bojic, Jessica Chen, Si Yuan Chang, Qi Chwen Ong, Shafiq Joty, Josip Car | 2023-10-12T07:59:56Z | cs.CL, cs.HC | Human evaluation plays a crucial role in Natural Language Processing (NLP) as
it assesses the quality and relevance of developed systems, thereby
facilitating their enhancement. However, the absence of widely accepted human
evaluation metrics in NLP hampers fair comparisons among different systems and
the establishment of universal assessment standards. Through an extensive
analysis of existing literature on human evaluation metrics, we identified
several gaps in NLP evaluation methodologies. These gaps served as motivation
for developing our own hierarchical evaluation framework. The proposed
framework offers notable advantages, particularly in providing a more
comprehensive representation of the NLP system's performance. We applied this
framework to evaluate the developed Machine Reading Comprehension system, which
was utilized within a human-AI symbiosis model. The results highlighted the
associations between the quality of inputs and outputs, underscoring the
necessity to evaluate both components rather than solely focusing on outputs.
In future work, we will investigate the potential time-saving benefits of our
proposed framework for evaluators assessing NLP systems. | cs.CL | Hierarchical Evaluation Framework: Best Practices for Human Evaluation |
2023-10-03T08:44:50Z | http://arxiv.org/pdf/2310.01889v4 | http://arxiv.org/abs/2310.01889v4 | Hao Liu, Matei Zaharia, Pieter Abbeel | 2023-11-27T06:38:47Z | cs.CL | Transformers have emerged as the architecture of choice for many
state-of-the-art AI models, showcasing exceptional performance across a wide
range of AI applications. However, the memory demands imposed by Transformers
limit their ability to handle long sequences, thereby posing challenges in
utilizing videos, actions, and other long-form sequences and modalities in
complex environments. We present a novel approach, Ring Attention with
Blockwise Transformers (Ring Attention), which leverages blockwise computation
of self-attention and feedforward to distribute long sequences across multiple
devices while fully overlapping the communication of key-value blocks with the
computation of blockwise attention. Our approach enables training and inference
of sequences that are up to device count times longer than those achievable by
prior memory-efficient Transformers, without resorting to approximations or
incurring additional communication and computation overheads. Extensive
experiments on language modeling and reinforcement learning tasks demonstrate
the effectiveness of our approach in allowing millions of tokens context size
and improving performance. | cs.CL | Ring Attention with Blockwise Transformers for Near-Infinite Context |
2023-10-03T08:39:33Z | http://arxiv.org/pdf/2310.01886v3 | http://arxiv.org/abs/2310.01886v3 | Weisen Jiang, Baijiong Lin, Han Shi, Yu Zhang, Zhenguo Li, James T. Kwok | 2024-02-03T15:22:33Z | cs.LG, cs.CL, cs.CV | Recently, various merging methods have been proposed to build a multi-task
model from task-specific finetuned models without retraining. However, existing
methods suffer from a large performance deterioration compared to using
multiple task-specific models. In this paper, we propose to inject
task-specific knowledge into the merged model and design two
parameter-efficient approaches (BYOM-FFT and BYOM-LoRA) to Build Your Own
Multi-task model. BYOM-FFT is for merging fully finetuned models, while
BYOM-LoRA is for LoRA-finetuned models. Both methods are data-free and
computation-efficient. Extensive experiments on computer vision and natural
language processing tasks show that the proposed BYOM methods outperform
existing merging methods by a large margin. Moreover, BYOM-FFT is general and
can be integrated into existing merging methods to further boost performance. | cs.LG | BYOM: Building Your Own Multi-Task Model For Free |
2023-10-03T07:34:30Z | http://arxiv.org/pdf/2310.01854v1 | http://arxiv.org/abs/2310.01854v1 | Jingyuan Sun, Marie-Francine Moens | 2023-10-03T07:34:30Z | cs.AI, cs.CL | To decipher the algorithm underlying the human brain's language
representation, previous work probed brain responses to language input with
pre-trained artificial neural network (ANN) models fine-tuned on NLU tasks.
However, full fine-tuning generally updates the entire parametric space and
distorts pre-trained features, cognitively inconsistent with the brain's robust
multi-task learning ability. Prompt-tuning, in contrast, protects pre-trained
weights and learns task-specific embeddings to fit a task. Could prompt-tuning
generate representations that better account for the brain's language
representations than fine-tuning? If so, what kind of NLU task leads a
pre-trained model to better decode the information represented in the human
brain? We investigate these questions by comparing prompt-tuned and fine-tuned
representations in neural decoding, that is predicting the linguistic stimulus
from the brain activities evoked by the stimulus. We find that on none of the
10 NLU tasks, full fine-tuning significantly outperforms prompt-tuning in
neural decoding, implicating that a more brain-consistent tuning method yields
representations that better correlate with brain data. Moreover, we identify
that tasks dealing with fine-grained concept meaning yield representations that
better decode brain activation patterns than other tasks, especially the
syntactic chunking task. This indicates that our brain encodes more
fine-grained concept information than shallow syntactic information when
representing languages. | cs.AI | Fine-tuned vs. Prompt-tuned Supervised Representations: Which Better Account for Brain Language Representations? |
2023-10-03T07:23:22Z | http://arxiv.org/pdf/2310.01846v1 | http://arxiv.org/abs/2310.01846v1 | Xiang Lisa Li, Vaishnavi Shrivastava, Siyan Li, Tatsunori Hashimoto, Percy Liang | 2023-10-03T07:23:22Z | cs.CL, cs.LG | As of September 2023, ChatGPT correctly answers "what is 7+8" with 15, but
when asked "7+8=15, True or False" it responds with "False". This inconsistency
between generating and validating an answer is prevalent in language models
(LMs) and erodes trust. In this paper, we propose a framework for measuring the
consistency between generation and validation (which we call
generator-validator consistency, or GV-consistency), finding that even GPT-4, a
state-of-the-art LM, is GV-consistent only 76% of the time. To improve the
consistency of LMs, we propose to finetune on the filtered generator and
validator responses that are GV-consistent, and call this approach consistency
fine-tuning. We find that this approach improves GV-consistency of Alpaca-30B
from 60% to 93%, and the improvement extrapolates to unseen tasks and domains
(e.g., GV-consistency for positive style transfers extrapolates to unseen
styles like humor). In addition to improving consistency, consistency
fine-tuning improves both generator quality and validator accuracy without
using any labeled data. Evaluated across 6 tasks, including math questions,
knowledge-intensive QA, and instruction following, our method improves the
generator quality by 16% and the validator accuracy by 6.3% across all tasks. | cs.CL | Benchmarking and Improving Generator-Validator Consistency of Language Models |
2023-10-03T07:19:59Z | http://arxiv.org/pdf/2310.01845v2 | http://arxiv.org/abs/2310.01845v2 | Ali Mayladan, Hasan Nasrallah, Hasan Moughnieh, Mustafa Shukor, Ali J. Ghandour | 2024-02-11T14:28:09Z | cs.CV, cs.AI, cs.CL, cs.LG | Foundation models have excelled in various tasks but are often evaluated on
general benchmarks. The adaptation of these models for specific domains, such
as remote sensing imagery, remains an underexplored area. In remote sensing,
precise building instance segmentation is vital for applications like urban
planning. While Convolutional Neural Networks (CNNs) perform well, their
generalization can be limited. For this aim, we present a novel approach to
adapt foundation models to address existing models' generalization dropback.
Among several models, our focus centers on the Segment Anything Model (SAM), a
potent foundation model renowned for its prowess in class-agnostic image
segmentation capabilities. We start by identifying the limitations of SAM,
revealing its suboptimal performance when applied to remote sensing imagery.
Moreover, SAM does not offer recognition abilities and thus fails to classify
and tag localized objects. To address these limitations, we introduce different
prompting strategies, including integrating a pre-trained CNN as a prompt
generator. This novel approach augments SAM with recognition abilities, a first
of its kind. We evaluated our method on three remote sensing datasets,
including the WHU Buildings dataset, the Massachusetts Buildings dataset, and
the AICrowd Mapping Challenge. For out-of-distribution performance on the WHU
dataset, we achieve a 5.47\% increase in IoU and a 4.81\% improvement in
F1-score. For in-distribution performance on the WHU dataset, we observe a
2.72\% and 1.58\% increase in True-Positive-IoU and True-Positive-F1 score,
respectively. Our code is publicly available at this Repo
(https://github.com/geoaigroup/GEOAI-ECRS2023), hoping to inspire further
exploration of foundation models for domain-specific tasks within the remote
sensing community. | cs.CV | Zero-Shot Refinement of Buildings' Segmentation Models using SAM |
2023-10-03T07:05:37Z | http://arxiv.org/pdf/2310.01839v2 | http://arxiv.org/abs/2310.01839v2 | Bi-Cheng Yan, Hsin-Wei Wang, Yi-Cheng Wang, Jiun-Ting Li, Chi-Han Lin, Berlin Chen | 2023-10-04T06:51:24Z | eess.AS, cs.CL, cs.SD | Automatic pronunciation assessment (APA) manages to quantify the
pronunciation proficiency of a second language (L2) learner in a language.
Prevailing approaches to APA normally leverage neural models trained with a
regression loss function, such as the mean-squared error (MSE) loss, for
proficiency level prediction. Despite most regression models can effectively
capture the ordinality of proficiency levels in the feature space, they are
confronted with a primary obstacle that different phoneme categories with the
same proficiency level are inevitably forced to be close to each other,
retaining less phoneme-discriminative information. On account of this, we
devise a phonemic contrast ordinal (PCO) loss for training regression-based APA
models, which aims to preserve better phonemic distinctions between phoneme
categories meanwhile considering ordinal relationships of the regression target
output. Specifically, we introduce a phoneme-distinct regularizer into the MSE
loss, which encourages feature representations of different phoneme categories
to be far apart while simultaneously pulling closer the representations
belonging to the same phoneme category by means of weighted distances. An
extensive set of experiments carried out on the speechocean762 benchmark
dataset suggest the feasibility and effectiveness of our model in relation to
some existing state-of-the-art models. | eess.AS | Preserving Phonemic Distinctions for Ordinal Regression: A Novel Loss Function for Automatic Pronunciation Assessment |
2023-10-03T07:01:23Z | http://arxiv.org/pdf/2310.01837v2 | http://arxiv.org/abs/2310.01837v2 | Abdul Karim Gizzini, Mustafa Shukor, Ali J. Ghandour | 2023-11-28T13:36:58Z | cs.CV, cs.AI, cs.CL, cs.LG | Current AI-based methods do not provide comprehensible physical
interpretations of the utilized data, extracted features, and
predictions/inference operations. As a result, deep learning models trained
using high-resolution satellite imagery lack transparency and explainability
and can be merely seen as a black box, which limits their wide-level adoption.
Experts need help understanding the complex behavior of AI models and the
underlying decision-making process. The explainable artificial intelligence
(XAI) field is an emerging field providing means for robust, practical, and
trustworthy deployment of AI models. Several XAI techniques have been proposed
for image classification tasks, whereas the interpretation of image
segmentation remains largely unexplored. This paper offers to bridge this gap
by adapting the recent XAI classification algorithms and making them usable for
muti-class image segmentation, where we mainly focus on buildings' segmentation
from high-resolution satellite images. To benchmark and compare the performance
of the proposed approaches, we introduce a new XAI evaluation methodology and
metric based on "Entropy" to measure the model uncertainty. Conventional XAI
evaluation methods rely mainly on feeding area-of-interest regions from the
image back to the pre-trained (utility) model and then calculating the average
change in the probability of the target class. Those evaluation metrics lack
the needed robustness, and we show that using Entropy to monitor the model
uncertainty in segmenting the pixels within the target class is more suitable.
We hope this work will pave the way for additional XAI research for image
segmentation and applications in the remote sensing discipline. | cs.CV | Extending CAM-based XAI methods for Remote Sensing Imagery Segmentation |
2023-10-03T06:51:48Z | http://arxiv.org/pdf/2310.01828v2 | http://arxiv.org/abs/2310.01828v2 | Hossein Shreim, Abdul Karim Gizzini, Ali J. Ghandour | 2023-11-25T13:35:34Z | cs.CV, cs.AI, cs.CL, cs.LG | eXplainable Artificial Intelligence (XAI) has emerged as an essential
requirement when dealing with mission-critical applications, ensuring
transparency and interpretability of the employed black box AI models. The
significance of XAI spans various domains, from healthcare to finance, where
understanding the decision-making process of deep learning algorithms is
essential. Most AI-based computer vision models are often black boxes; hence,
providing explainability of deep neural networks in image processing is crucial
for their wide adoption and deployment in medical image analysis, autonomous
driving, and remote sensing applications. Recently, several XAI methods for
image classification tasks have been introduced. On the contrary, image
segmentation has received comparatively less attention in the context of
explainability, although it is a fundamental task in computer vision
applications, especially in remote sensing. Only some research proposes
gradient-based XAI algorithms for image segmentation. This paper adapts the
recent gradient-free Sobol XAI method for semantic segmentation. To measure the
performance of the Sobol method for segmentation, we propose a quantitative XAI
evaluation method based on a learnable noise model. The main objective of this
model is to induce noise on the explanation maps, where higher induced noise
signifies low accuracy and vice versa. A benchmark analysis is conducted to
evaluate and compare performance of three XAI methods, including Seg-Grad-CAM,
Seg-Grad-CAM++ and Seg-Sobol using the proposed noise-based evaluation
technique. This constitutes the first attempt to run and evaluate XAI methods
using high-resolution satellite images. | cs.CV | Trainable Noise Model as an XAI evaluation method: application on Sobol for remote sensing image segmentation |
2023-10-03T06:42:28Z | http://arxiv.org/pdf/2310.01825v2 | http://arxiv.org/abs/2310.01825v2 | Mohamad Hasan Zahweh, Hasan Nasrallah, Mustafa Shukor, Ghaleb Faour, Ali J. Ghandour | 2023-11-27T10:39:13Z | cs.CV, cs.AI, cs.CL, cs.LG | Parameter Efficient Fine Tuning (PEFT) techniques have recently experienced
significant growth and have been extensively employed to adapt large vision and
language models to various domains, enabling satisfactory model performance
with minimal computational needs. Despite these advances, more research has yet
to delve into potential PEFT applications in real-life scenarios, particularly
in the critical domains of remote sensing and crop monitoring. The diversity of
climates across different regions and the need for comprehensive large-scale
datasets have posed significant obstacles to accurately identify crop types
across varying geographic locations and changing growing seasons. This study
seeks to bridge this gap by comprehensively exploring the feasibility of
cross-area and cross-year out-of-distribution generalization using the
State-of-the-Art (SOTA) wheat crop monitoring model. The aim of this work is to
explore PEFT approaches for crop monitoring. Specifically, we focus on adapting
the SOTA TSViT model to address winter wheat field segmentation, a critical
task for crop monitoring and food security. This adaptation process involves
integrating different PEFT techniques, including BigFit, LoRA, Adaptformer, and
prompt tuning. Using PEFT techniques, we achieved notable results comparable to
those achieved using full fine-tuning methods while training only a mere 0.7%
parameters of the whole TSViT architecture. The in-house labeled data-set,
referred to as the Beqaa-Lebanon dataset, comprises high-quality annotated
polygons for wheat and non-wheat classes with a total surface of 170 kmsq, over
five consecutive years. Using Sentinel-2 images, our model achieved a 84%
F1-score. We intend to publicly release the Lebanese winter wheat data set,
code repository, and model weights. | cs.CV | Empirical Study of PEFT techniques for Winter Wheat Segmentation |
2023-10-03T05:17:08Z | http://arxiv.org/pdf/2310.01801v3 | http://arxiv.org/abs/2310.01801v3 | Suyu Ge, Yunan Zhang, Liyuan Liu, Minjia Zhang, Jiawei Han, Jianfeng Gao | 2024-01-29T06:25:00Z | cs.CL | In this study, we introduce adaptive KV cache compression, a plug-and-play
method that reduces the memory footprint of generative inference for Large
Language Models (LLMs). Different from the conventional KV cache that retains
key and value vectors for all context tokens, we conduct targeted profiling to
discern the intrinsic structure of attention modules. Based on the recognized
structure, we then construct the KV cache in an adaptive manner: evicting
long-range contexts on attention heads emphasizing local contexts, discarding
non-special tokens on attention heads centered on special tokens, and only
employing the standard KV cache for attention heads that broadly attend to all
tokens. Moreover, with the lightweight attention profiling used to guide the
construction of the adaptive KV cache, FastGen can be deployed without
resource-intensive fine-tuning or re-training. In our experiments across
various asks, FastGen demonstrates substantial reduction on GPU memory
consumption with negligible generation quality loss. We will release our code
and the compatible CUDA kernel for reproducibility. | cs.CL | Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs |
2023-10-03T04:56:12Z | http://arxiv.org/pdf/2310.01798v2 | http://arxiv.org/abs/2310.01798v2 | Jie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng, Adams Wei Yu, Xinying Song, Denny Zhou | 2024-03-14T04:27:52Z | cs.CL, cs.AI | Large Language Models (LLMs) have emerged as a groundbreaking technology with
their unparalleled text generation capabilities across various applications.
Nevertheless, concerns persist regarding the accuracy and appropriateness of
their generated content. A contemporary methodology, self-correction, has been
proposed as a remedy to these issues. Building upon this premise, this paper
critically examines the role and efficacy of self-correction within LLMs,
shedding light on its true potential and limitations. Central to our
investigation is the notion of intrinsic self-correction, whereby an LLM
attempts to correct its initial responses based solely on its inherent
capabilities, without the crutch of external feedback. In the context of
reasoning, our research indicates that LLMs struggle to self-correct their
responses without external feedback, and at times, their performance even
degrades after self-correction. Drawing from these insights, we offer
suggestions for future research and practical applications in this field. | cs.CL | Large Language Models Cannot Self-Correct Reasoning Yet |
2023-10-03T04:14:17Z | http://arxiv.org/pdf/2310.01783v1 | http://arxiv.org/abs/2310.01783v1 | Weixin Liang, Yuhui Zhang, Hancheng Cao, Binglu Wang, Daisy Ding, Xinyu Yang, Kailas Vodrahalli, Siyu He, Daniel Smith, Yian Yin, Daniel McFarland, James Zou | 2023-10-03T04:14:17Z | cs.LG, cs.AI, cs.CL, cs.HC | Expert feedback lays the foundation of rigorous research. However, the rapid
growth of scholarly production and intricate knowledge specialization challenge
the conventional scientific feedback mechanisms. High-quality peer reviews are
increasingly difficult to obtain. Researchers who are more junior or from
under-resourced settings have especially hard times getting timely feedback.
With the breakthrough of large language models (LLM) such as GPT-4, there is
growing interest in using LLMs to generate scientific feedback on research
manuscripts. However, the utility of LLM-generated feedback has not been
systematically studied. To address this gap, we created an automated pipeline
using GPT-4 to provide comments on the full PDFs of scientific papers. We
evaluated the quality of GPT-4's feedback through two large-scale studies. We
first quantitatively compared GPT-4's generated feedback with human peer
reviewer feedback in 15 Nature family journals (3,096 papers in total) and the
ICLR machine learning conference (1,709 papers). The overlap in the points
raised by GPT-4 and by human reviewers (average overlap 30.85% for Nature
journals, 39.23% for ICLR) is comparable to the overlap between two human
reviewers (average overlap 28.58% for Nature journals, 35.25% for ICLR). The
overlap between GPT-4 and human reviewers is larger for the weaker papers. We
then conducted a prospective user study with 308 researchers from 110 US
institutions in the field of AI and computational biology to understand how
researchers perceive feedback generated by our GPT-4 system on their own
papers. Overall, more than half (57.4%) of the users found GPT-4 generated
feedback helpful/very helpful and 82.4% found it more beneficial than feedback
from at least some human reviewers. While our findings show that LLM-generated
feedback can help researchers, we also identify several limitations. | cs.LG | Can large language models provide useful feedback on research papers? A large-scale empirical analysis |
2023-10-03T03:56:26Z | http://arxiv.org/pdf/2310.01777v2 | http://arxiv.org/abs/2310.01777v2 | Heejun Lee, Jina Kim, Jeffrey Willette, Sung Ju Hwang | 2024-03-25T04:04:05Z | cs.CL, cs.LG | The transformer architecture has driven breakthroughs in recent years on
tasks which require modeling pairwise relationships between sequential
elements, as is the case in natural language understanding. However, long
seqeuences pose a problem due to the quadratic complexity of the attention
operation. Previous research has aimed to lower the complexity by sparsifying
or linearly approximating the attention matrix. Yet, these approaches cannot
straightforwardly distill knowledge from a teacher's attention matrix and often
require complete retraining from scratch. Furthermore, previous sparse and
linear approaches lose interpretability if they cannot produce full attention
matrices. To address these challenges, we propose SEA: Sparse linear attention
with an Estimated Attention mask. SEA estimates the attention matrix with
linear complexity via kernel-based linear attention, then subsequently creates
a sparse attention matrix with a top-k selection to perform a sparse attention
operation. For language modeling tasks (Wikitext2), previous linear and sparse
attention methods show roughly two-fold worse perplexity scores over the
quadratic OPT-1.3B baseline, while SEA achieves better perplexity than
OPT-1.3B, using roughly half the memory of OPT-1.3B, providing interpretable
attention matrix. We believe that our work will have a large practical impact,
as it opens the possibility of running large transformers on resource-limited
devices with less memory. | cs.CL | SEA: Sparse Linear Attention with Estimated Attention Mask |
2023-10-03T02:18:06Z | http://arxiv.org/pdf/2310.01749v2 | http://arxiv.org/abs/2310.01749v2 | Brian DuSell, David Chiang | 2024-01-24T16:28:43Z | cs.CL | Attention, specifically scaled dot-product attention, has proven effective
for natural language, but it does not have a mechanism for handling
hierarchical patterns of arbitrary nesting depth, which limits its ability to
recognize certain syntactic structures. To address this shortcoming, we propose
stack attention: an attention operator that incorporates stacks, inspired by
their theoretical connections to context-free languages (CFLs). We show that
stack attention is analogous to standard attention, but with a latent model of
syntax that requires no syntactic supervision. We propose two variants: one
related to deterministic pushdown automata (PDAs) and one based on
nondeterministic PDAs, which allows transformers to recognize arbitrary CFLs.
We show that transformers with stack attention are very effective at learning
CFLs that standard transformers struggle on, achieving strong results on a CFL
with theoretically maximal parsing difficulty. We also show that stack
attention is more effective at natural language modeling under a constrained
parameter budget, and we include results on machine translation. | cs.CL | Stack Attention: Improving the Ability of Transformers to Model Hierarchical Patterns |
2023-10-03T01:47:49Z | http://arxiv.org/pdf/2310.01732v1 | http://arxiv.org/abs/2310.01732v1 | Guanghui Qin, Benjamin Van Durme | 2023-10-03T01:47:49Z | cs.CL, cs.AI, cs.LG, I.2.7; I.2.6 | Embedding text sequences is a widespread requirement in modern language
understanding. Existing approaches focus largely on constant-size
representations. This is problematic, as the amount of information contained in
text often varies with the length of the input. We propose a solution called
Nugget, which encodes language into a representation based on a dynamically
selected subset of input tokens. These nuggets are learned through tasks like
autoencoding and machine translation, and intuitively segment language into
meaningful units. We demonstrate Nugget outperforms related approaches in tasks
involving semantic comparison. Finally, we illustrate these compact units allow
for expanding the contextual window of a language model (LM), suggesting new
future LMs that can condition on significantly larger amounts of content. | cs.CL | Nugget: Neural Agglomerative Embeddings of Text |
2023-10-03T01:34:55Z | http://arxiv.org/pdf/2311.12833v1 | http://arxiv.org/abs/2311.12833v1 | Xianzhong Ding, Le Chen, Murali Emani, Chunhua Liao, Pei-Hung Lin, Tristan Vanderbruggen, Zhen Xie, Alberto E. Cerpa, Wan Du | 2023-10-03T01:34:55Z | cs.DC, cs.AI, cs.CL | Large Language Models (LLMs), including the LLaMA model, have exhibited their
efficacy across various general-domain natural language processing (NLP) tasks.
However, their performance in high-performance computing (HPC) domain tasks has
been less than optimal due to the specialized expertise required to interpret
the model responses. In response to this challenge, we propose HPC-GPT, a novel
LLaMA-based model that has been supervised fine-tuning using generated QA
(Question-Answer) instances for the HPC domain. To evaluate its effectiveness,
we concentrate on two HPC tasks: managing AI models and datasets for HPC, and
data race detection. By employing HPC-GPT, we demonstrate comparable
performance with existing methods on both tasks, exemplifying its excellence in
HPC-related scenarios. Our experiments on open-source benchmarks yield
extensive results, underscoring HPC-GPT's potential to bridge the performance
gap between LLMs and HPC-specific tasks. With HPC-GPT, we aim to pave the way
for LLMs to excel in HPC domains, simplifying the utilization of language
models in complex computing applications. | cs.DC | HPC-GPT: Integrating Large Language Model for High-Performance Computing |
2023-10-03T01:02:44Z | http://arxiv.org/pdf/2310.01717v2 | http://arxiv.org/abs/2310.01717v2 | Behzad Shayegh, Yanshuai Cao, Xiaodan Zhu, Jackie C. K. Cheung, Lili Mou | 2024-04-26T00:41:48Z | cs.CL, cs.AI, cs.LG | We investigate the unsupervised constituency parsing task, which organizes
words and phrases of a sentence into a hierarchical structure without using
linguistically annotated data. We observe that existing unsupervised parsers
capture differing aspects of parsing structures, which can be leveraged to
enhance unsupervised parsing performance. To this end, we propose a notion of
"tree averaging," based on which we further propose a novel ensemble method for
unsupervised parsing. To improve inference efficiency, we further distill the
ensemble knowledge into a student model; such an ensemble-then-distill process
is an effective approach to mitigate the over-smoothing problem existing in
common multi-teacher distilling methods. Experiments show that our method
surpasses all previous approaches, consistently demonstrating its effectiveness
and robustness across various runs, with different ensemble components, and
under domain-shift conditions. | cs.CL | Ensemble Distillation for Unsupervised Constituency Parsing |
2023-10-03T00:08:23Z | http://arxiv.org/pdf/2310.01708v1 | http://arxiv.org/abs/2310.01708v1 | D. Umerenkov, G. Zubkova, A. Nesterov | 2023-10-03T00:08:23Z | cs.CL | Clinical Decision Support Systems (CDSS) utilize evidence-based knowledge and
patient data to offer real-time recommendations, with Large Language Models
(LLMs) emerging as a promising tool to generate plain-text explanations for
medical decisions. This study explores the effectiveness and reliability of
LLMs in generating explanations for diagnoses based on patient complaints.
Three experienced doctors evaluated LLM-generated explanations of the
connection between patient complaints and doctor and model-assigned diagnoses
across several stages. Experimental results demonstrated that LLM explanations
significantly increased doctors' agreement rates with given diagnoses and
highlighted potential errors in LLM outputs, ranging from 5% to 30%. The study
underscores the potential and challenges of LLMs in healthcare and emphasizes
the need for careful integration and evaluation to ensure patient safety and
optimal clinical utility. | cs.CL | Deciphering Diagnoses: How Large Language Models Explanations Influence Clinical Decision Making |
2023-10-02T23:29:23Z | http://arxiv.org/pdf/2310.04445v2 | http://arxiv.org/abs/2310.04445v2 | Muhammad Ahmed Shah, Roshan Sharma, Hira Dhamyal, Raphael Olivier, Ankit Shah, Joseph Konan, Dareen Alharthi, Hazim T Bukhari, Massa Baali, Soham Deshmukh, Michael Kuhlmann, Bhiksha Raj, Rita Singh | 2023-10-21T17:59:41Z | cs.CL, cs.AI, cs.LG | It has been shown that Large Language Model (LLM) alignments can be
circumvented by appending specially crafted attack suffixes with harmful
queries to elicit harmful responses. To conduct attacks against private target
models whose characterization is unknown, public models can be used as proxies
to fashion the attack, with successful attacks being transferred from public
proxies to private target models. The success rate of attack depends on how
closely the proxy model approximates the private model. We hypothesize that for
attacks to be transferrable, it is sufficient if the proxy can approximate the
target model in the neighborhood of the harmful query. Therefore, in this
paper, we propose \emph{Local Fine-Tuning (LoFT)}, \textit{i.e.}, fine-tuning
proxy models on similar queries that lie in the lexico-semantic neighborhood of
harmful queries to decrease the divergence between the proxy and target models.
First, we demonstrate three approaches to prompt private target models to
obtain similar queries given harmful queries. Next, we obtain data for local
fine-tuning by eliciting responses from target models for the generated similar
queries. Then, we optimize attack suffixes to generate attack prompts and
evaluate the impact of our local fine-tuning on the attack's success rate.
Experiments show that local fine-tuning of proxy models improves attack
transferability and increases attack success rate by $39\%$, $7\%$, and $0.5\%$
(absolute) on target models ChatGPT, GPT-4, and Claude respectively. | cs.CL | LoFT: Local Proxy Fine-tuning For Improving Transferability Of Adversarial Attacks Against Large Language Model |
2023-10-02T23:16:25Z | http://arxiv.org/pdf/2310.01693v1 | http://arxiv.org/abs/2310.01693v1 | Matthew Finlayson, John Hewitt, Alexander Koller, Swabha Swayamdipta, Ashish Sabharwal | 2023-10-02T23:16:25Z | cs.CL, 68T50, I.2.7 | Despite their ubiquity in language generation, it remains unknown why
truncation sampling heuristics like nucleus sampling are so effective. We
provide a theoretical explanation for the effectiveness of the truncation
sampling by proving that truncation methods that discard tokens below some
probability threshold (the most common type of truncation) can guarantee that
all sampled tokens have nonzero true probability. However, thresholds are a
coarse heuristic, and necessarily discard some tokens with nonzero true
probability as well. In pursuit of a more precise sampling strategy, we show
that we can leverage a known source of model errors, the softmax bottleneck, to
prove that certain tokens have nonzero true probability, without relying on a
threshold. Based on our findings, we develop an experimental truncation
strategy and the present pilot studies demonstrating the promise of this type
of algorithm. Our evaluations show that our method outperforms its
threshold-based counterparts under automatic and human evaluation metrics for
low-entropy (i.e., close to greedy) open-ended text generation. Our theoretical
findings and pilot experiments provide both insight into why truncation
sampling works, and make progress toward more expressive sampling algorithms
that better surface the generative capabilities of large language models. | cs.CL | Closing the Curious Case of Neural Text Degeneration |
2023-10-02T23:12:21Z | http://arxiv.org/pdf/2310.01691v1 | http://arxiv.org/abs/2310.01691v1 | Zijun Wu, Yongkang Wu, Lili Mou | 2023-10-02T23:12:21Z | cs.CL, cs.AI | Prompt tuning in natural language processing (NLP) has become an increasingly
popular method for adapting large language models to specific tasks. However,
the transferability of these prompts, especially continuous prompts, between
different models remains a challenge. In this work, we propose a zero-shot
continuous prompt transfer method, where source prompts are encoded into
relative space and the corresponding target prompts are searched for
transferring to target models. Experimental results confirm the effectiveness
of our method, showing that 'task semantics' in continuous prompts can be
generalized across various language models. Moreover, we find that combining
'task semantics' from multiple source models can further enhance the
generalizability of transfer. | cs.CL | Zero-Shot Continuous Prompt Transfer: Generalizing Task Semantics Across Language Models |
2023-10-02T23:03:30Z | http://arxiv.org/pdf/2310.01688v1 | http://arxiv.org/abs/2310.01688v1 | Samuele Cornell, Jee-weon Jung, Shinji Watanabe, Stefano Squartini | 2023-10-02T23:03:30Z | eess.AS, cs.CL, cs.SD | This paper presents a novel framework for joint speaker diarization (SD) and
automatic speech recognition (ASR), named SLIDAR (sliding-window
diarization-augmented recognition). SLIDAR can process arbitrary length inputs
and can handle any number of speakers, effectively solving ``who spoke what,
when'' concurrently. SLIDAR leverages a sliding window approach and consists of
an end-to-end diarization-augmented speech transcription (E2E DAST) model which
provides, locally, for each window: transcripts, diarization and speaker
embeddings. The E2E DAST model is based on an encoder-decoder architecture and
leverages recent techniques such as serialized output training and
``Whisper-style" prompting. The local outputs are then combined to get the
final SD+ASR result by clustering the speaker embeddings to get global speaker
identities. Experiments performed on monaural recordings from the AMI corpus
confirm the effectiveness of the method in both close-talk and far-field speech
scenarios. | eess.AS | One model to rule them all ? Towards End-to-End Joint Speaker Diarization and Speech Recognition |
2023-10-02T22:35:40Z | http://arxiv.org/pdf/2310.04444v3 | http://arxiv.org/abs/2310.04444v3 | Aman Bhargava, Cameron Witkowski, Manav Shah, Matt Thomson | 2024-01-03T06:38:36Z | cs.CL, cs.AI, cs.LG, cs.NE | Prompt engineering is crucial for deploying LLMs but is poorly understood
mathematically. We formalize LLM systems as a class of discrete stochastic
dynamical systems to explore prompt engineering through the lens of control
theory. We investigate the reachable set of output token sequences $R_y(\mathbf
x_0)$ for which there exists a control input sequence $\mathbf u$ for each
$\mathbf y \in R_y(\mathbf x_0)$ that steers the LLM to output $\mathbf y$ from
initial state sequence $\mathbf x_0$. We offer analytic analysis on the
limitations on the controllability of self-attention in terms of reachable set,
where we prove an upper bound on the reachable set of outputs $R_y(\mathbf
x_0)$ as a function of the singular values of the parameter matrices. We
present complementary empirical analysis on the controllability of a panel of
LLMs, including Falcon-7b, Llama-7b, and Falcon-40b. Our results demonstrate a
lower bound on the reachable set of outputs $R_y(\mathbf x_0)$ w.r.t. initial
state sequences $\mathbf x_0$ sampled from the Wikitext dataset. We find that
the correct next Wikitext token following sequence $\mathbf x_0$ is reachable
over 97% of the time with prompts of $k\leq 10$ tokens. We also establish that
the top 75 most likely next tokens, as estimated by the LLM itself, are
reachable at least 85% of the time with prompts of $k\leq 10$ tokens.
Intriguingly, short prompt sequences can dramatically alter the likelihood of
specific outputs, even making the least likely tokens become the most likely
ones. This control-centric analysis of LLMs demonstrates the significant and
poorly understood role of input sequences in steering output probabilities,
offering a foundational perspective for enhancing language model system
capabilities. | cs.CL | What's the Magic Word? A Control Theory of LLM Prompting |
2023-10-02T21:24:26Z | http://arxiv.org/pdf/2310.04443v2 | http://arxiv.org/abs/2310.04443v2 | Hao Xue, Flora D. Salim | 2023-10-13T05:07:49Z | cs.CL, cs.AI | Question answering (QA) systems have attracted much attention from the
artificial intelligence community as they can learn to answer questions based
on the given knowledge source (e.g., images in visual question answering).
However, the research into question answering systems with human mobility data
remains unexplored. Mining human mobility data is crucial for various
applications such as smart city planning, pandemic management, and personalised
recommendation system. In this paper, we aim to tackle this gap and introduce a
novel task, that is, human mobility question answering (MobQA). The aim of the
task is to let the intelligent system learn from mobility data and answer
related questions. This task presents a new paradigm change in mobility
prediction research and further facilitates the research of human mobility
recommendation systems. To better support this novel research topic, this
vision paper also proposes an initial design of the dataset and a potential
deep learning model framework for the introduced MobQA task. We hope that this
paper will provide novel insights and open new directions in human mobility
research and question answering research. | cs.CL | Human Mobility Question Answering (Vision Paper) |
2023-10-02T20:45:41Z | http://arxiv.org/pdf/2310.01627v2 | http://arxiv.org/abs/2310.01627v2 | Lane Lawley, Christopher J. MacLellan | 2024-04-22T19:06:09Z | cs.HC, cs.AI, cs.CL | Machine learning often requires millions of examples to produce static,
black-box models. In contrast, interactive task learning (ITL) emphasizes
incremental knowledge acquisition from limited instruction provided by humans
in modalities such as natural language. However, ITL systems often suffer from
brittle, error-prone language parsing, which limits their usability. Large
language models (LLMs) are resistant to brittleness but are not interpretable
and cannot learn incrementally. We present VAL, an ITL system with a new
philosophy for LLM/symbolic integration. By using LLMs only for specific
tasks--such as predicate and argument selection--within an algorithmic
framework, VAL reaps the benefits of LLMs to support interactive learning of
hierarchical task knowledge from natural language. Acquired knowledge is human
interpretable and generalizes to support execution of novel tasks without
additional training. We studied users' interactions with VAL in a video game
setting, finding that most users could successfully teach VAL using language
they felt was natural. | cs.HC | VAL: Interactive Task Learning with GPT Dialog Parsing |
2023-10-02T19:54:30Z | http://arxiv.org/pdf/2310.01603v1 | http://arxiv.org/abs/2310.01603v1 | Xiaoyi Tian, Kristy Elizabeth Boyer | 2023-10-02T19:54:30Z | cs.CL, cs.AI, cs.HC | Natural Language Processing (NLP) plays a significant role in our daily lives
and has become an essential part of Artificial Intelligence (AI) education in
K-12. As children grow up with NLP-powered applications, it is crucial to
introduce NLP concepts to them, fostering their understanding of language
processing, language generation, and ethical implications of AI and NLP. This
paper presents a comprehensive review of digital learning environments for
teaching NLP in K-12. Specifically, it explores existing digital learning
tools, discusses how they support specific NLP tasks and procedures, and
investigates their explainability and evaluation results in educational
contexts. By examining the strengths and limitations of these tools, this
literature review sheds light on the current state of NLP learning tools in
K-12 education. It aims to guide future research efforts to refine existing
tools, develop new ones, and explore more effective and inclusive strategies
for integrating NLP into K-12 educational contexts. | cs.CL | A Review of Digital Learning Environments for Teaching Natural Language Processing in K-12 Education |
2023-10-02T19:03:11Z | http://arxiv.org/pdf/2310.01568v1 | http://arxiv.org/abs/2310.01568v1 | Haining Wang | 2023-10-02T19:03:11Z | cs.CL, cs.CY | Authorship identification has proven unsettlingly effective in inferring the
identity of the author of an unsigned document, even when sensitive personal
information has been carefully omitted. In the digital era, individuals leave a
lasting digital footprint through their written content, whether it is posted
on social media, stored on their employer's computers, or located elsewhere.
When individuals need to communicate publicly yet wish to remain anonymous,
there is little available to protect them from unwanted authorship
identification. This unprecedented threat to privacy is evident in scenarios
such as whistle-blowing. Proposed defenses against authorship identification
attacks primarily aim to obfuscate one's writing style, thereby making it
unlinkable to their pre-existing writing, while concurrently preserving the
original meaning and grammatical integrity. The presented work offers a
comprehensive review of the advancements in this research area spanning over
the past two decades and beyond. It emphasizes the methodological frameworks of
modification and generation-based strategies devised to evade authorship
identification attacks, highlighting joint efforts from the differential
privacy community. Limitations of current research are discussed, with a
spotlight on open challenges and potential research avenues. | cs.CL | Defending Against Authorship Identification Attacks |
2023-10-02T18:52:35Z | http://arxiv.org/pdf/2310.01558v2 | http://arxiv.org/abs/2310.01558v2 | Ori Yoran, Tomer Wolfson, Ori Ram, Jonathan Berant | 2024-05-05T15:58:24Z | cs.CL, cs.AI | Retrieval-augmented language models (RALMs) hold promise to produce language
understanding systems that are are factual, efficient, and up-to-date. An
important desideratum of RALMs, is that retrieved information helps model
performance when it is relevant, and does not harm performance when it is not.
This is particularly important in multi-hop reasoning scenarios, where misuse
of irrelevant evidence can lead to cascading errors. However, recent work has
shown that retrieval augmentation can sometimes have a negative effect on
performance. In this work, we present a thorough analysis on five open-domain
question answering benchmarks, characterizing cases when retrieval reduces
accuracy. We then propose two methods to mitigate this issue. First, a simple
baseline that filters out retrieved passages that do not entail question-answer
pairs according to a natural language inference (NLI) model. This is effective
in preventing performance reduction, but at a cost of also discarding relevant
passages. Thus, we propose a method for automatically generating data to
fine-tune the language model to properly leverage retrieved passages, using a
mix of relevant and irrelevant contexts at training time. We empirically show
that even 1,000 examples suffice to train the model to be robust to irrelevant
contexts while maintaining high performance on examples with relevant ones. | cs.CL | Making Retrieval-Augmented Language Models Robust to Irrelevant Context |
2023-10-02T17:59:57Z | http://arxiv.org/pdf/2310.01415v3 | http://arxiv.org/abs/2310.01415v3 | Jiageng Mao, Yuxi Qian, Junjie Ye, Hang Zhao, Yue Wang | 2023-12-05T05:26:29Z | cs.CV, cs.AI, cs.CL, cs.RO | We present a simple yet effective approach that can transform the OpenAI
GPT-3.5 model into a reliable motion planner for autonomous vehicles. Motion
planning is a core challenge in autonomous driving, aiming to plan a driving
trajectory that is safe and comfortable. Existing motion planners predominantly
leverage heuristic methods to forecast driving trajectories, yet these
approaches demonstrate insufficient generalization capabilities in the face of
novel and unseen driving scenarios. In this paper, we propose a novel approach
to motion planning that capitalizes on the strong reasoning capabilities and
generalization potential inherent to Large Language Models (LLMs). The
fundamental insight of our approach is the reformulation of motion planning as
a language modeling problem, a perspective not previously explored.
Specifically, we represent the planner inputs and outputs as language tokens,
and leverage the LLM to generate driving trajectories through a language
description of coordinate positions. Furthermore, we propose a novel
prompting-reasoning-finetuning strategy to stimulate the numerical reasoning
potential of the LLM. With this strategy, the LLM can describe highly precise
trajectory coordinates and also its internal decision-making process in natural
language. We evaluate our approach on the large-scale nuScenes dataset, and
extensive experiments substantiate the effectiveness, generalization ability,
and interpretability of our GPT-based motion planner. Code is now available at
https://github.com/PointsCoder/GPT-Driver. | cs.CV | GPT-Driver: Learning to Drive with GPT |
2023-10-02T17:59:07Z | http://arxiv.org/pdf/2310.01405v3 | http://arxiv.org/abs/2310.01405v3 | Andy Zou, Long Phan, Sarah Chen, James Campbell, Phillip Guo, Richard Ren, Alexander Pan, Xuwang Yin, Mantas Mazeika, Ann-Kathrin Dombrowski, Shashwat Goel, Nathaniel Li, Michael J. Byun, Zifan Wang, Alex Mallen, Steven Basart, Sanmi Koyejo, Dawn Song, Matt Fredrikson, J. Zico Kolter, Dan Hendrycks | 2023-10-10T08:00:53Z | cs.LG, cs.AI, cs.CL, cs.CV, cs.CY | In this paper, we identify and characterize the emerging area of
representation engineering (RepE), an approach to enhancing the transparency of
AI systems that draws on insights from cognitive neuroscience. RepE places
population-level representations, rather than neurons or circuits, at the
center of analysis, equipping us with novel methods for monitoring and
manipulating high-level cognitive phenomena in deep neural networks (DNNs). We
provide baselines and an initial analysis of RepE techniques, showing that they
offer simple yet effective solutions for improving our understanding and
control of large language models. We showcase how these methods can provide
traction on a wide range of safety-relevant problems, including honesty,
harmlessness, power-seeking, and more, demonstrating the promise of top-down
transparency research. We hope that this work catalyzes further exploration of
RepE and fosters advancements in the transparency and safety of AI systems. | cs.LG | Representation Engineering: A Top-Down Approach to AI Transparency |
2023-10-02T17:47:10Z | http://arxiv.org/pdf/2310.01387v1 | http://arxiv.org/abs/2310.01387v1 | Amanda Bertsch, Alex Xie, Graham Neubig, Matthew R. Gormley | 2023-10-02T17:47:10Z | cs.CL | Minimum Bayes Risk (MBR) decoding is a method for choosing the outputs of a
machine learning system based not on the output with the highest probability,
but the output with the lowest risk (expected error) among multiple candidates.
It is a simple but powerful method: for an additional cost at inference time,
MBR provides reliable several-point improvements across metrics for a wide
variety of tasks without any additional data or training. Despite this, MBR is
not frequently applied in NLP works, and knowledge of the method itself is
limited. We first provide an introduction to the method and the recent
literature. We show that several recent methods that do not reference MBR can
be written as special cases of MBR; this reformulation provides additional
theoretical justification for the performance of these methods, explaining some
results that were previously only empirical. We provide theoretical and
empirical results about the effectiveness of various MBR variants and make
concrete recommendations for the application of MBR in NLP models, including
future directions in this area. | cs.CL | It's MBR All the Way Down: Modern Generation Techniques Through the Lens of Minimum Bayes Risk |
2023-10-02T17:46:09Z | http://arxiv.org/pdf/2310.01386v2 | http://arxiv.org/abs/2310.01386v2 | Jen-tse Huang, Wenxuan Wang, Eric John Li, Man Ho Lam, Shujie Ren, Youliang Yuan, Wenxiang Jiao, Zhaopeng Tu, Michael R. Lyu | 2024-01-22T13:58:50Z | cs.CL | Large Language Models (LLMs) have recently showcased their remarkable
capacities, not only in natural language processing tasks but also across
diverse domains such as clinical medicine, legal consultation, and education.
LLMs become more than mere applications, evolving into assistants capable of
addressing diverse user requests. This narrows the distinction between human
beings and artificial intelligence agents, raising intriguing questions
regarding the potential manifestation of personalities, temperaments, and
emotions within LLMs. In this paper, we propose a framework, PsychoBench, for
evaluating diverse psychological aspects of LLMs. Comprising thirteen scales
commonly used in clinical psychology, PsychoBench further classifies these
scales into four distinct categories: personality traits, interpersonal
relationships, motivational tests, and emotional abilities. Our study examines
five popular models, namely text-davinci-003, gpt-3.5-turbo, gpt-4, LLaMA-2-7b,
and LLaMA-2-13b. Additionally, we employ a jailbreak approach to bypass the
safety alignment protocols and test the intrinsic natures of LLMs. We have made
PsychoBench openly accessible via https://github.com/CUHK-ARISE/PsychoBench. | cs.CL | Who is ChatGPT? Benchmarking LLMs' Psychological Portrayal Using PsychoBench |
2023-10-02T17:42:37Z | http://arxiv.org/pdf/2310.01382v2 | http://arxiv.org/abs/2310.01382v2 | Ajay Jaiswal, Zhe Gan, Xianzhi Du, Bowen Zhang, Zhangyang Wang, Yinfei Yang | 2024-03-17T00:38:59Z | cs.CL, cs.LG | Despite their remarkable achievements, modern Large Language Models (LLMs)
face exorbitant computational and memory footprints. Recently, several works
have shown significant success in training-free and data-free compression
(pruning and quantization) of LLMs that achieve 50 - 60% sparsity and reduce
the bit width to 3 or 4 bits per weight, with negligible degradation of
perplexity over the uncompressed baseline. As recent research efforts are
focused on developing increasingly sophisticated compression methods, our work
takes a step back and re-evaluates the effectiveness of existing SoTA
compression methods, which rely on a fairly simple and widely questioned
metric, perplexity (even for dense LLMs). We introduce Knowledge-Intensive
Compressed LLM BenchmarK (LLM-KICK), a collection of carefully curated tasks to
redefine the evaluation protocol for compressed LLMs, which have significant
alignment with their dense counterparts and perplexity fail to capture subtle
change in their true capabilities. LLM-KICK unveils many favorable merits and
unfortunate plights of current SoTA compression methods: all pruning methods
suffer significant performance degradation, sometimes at trivial sparsity
ratios (e.g., 25-30%), and fail for N:M sparsity in knowledge-intensive tasks;
current quantization methods are more successful than pruning; yet, pruned LLMs
even at $\geq 50$% sparsity are robust in-context retrieval and summarization
systems; among others. LLM-KICK is designed to holistically access compressed
LLMs' ability for language understanding, reasoning, generation, in-context
retrieval, in-context summarization, etc. We hope our study can foster the
development of better LLM compression methods. The reproduced codes are
available at https://github.com/VITA-Group/llm-kick. | cs.CL | Compressing LLMs: The Truth is Rarely Pure and Never Simple |
2023-10-02T17:42:22Z | http://arxiv.org/pdf/2310.01381v3 | http://arxiv.org/abs/2310.01381v3 | Roi Benita, Michael Elad, Joseph Keshet | 2024-03-10T22:31:45Z | cs.SD, cs.CL, eess.AS | Diffusion models have recently been shown to be relevant for high-quality
speech generation. Most work has been focused on generating spectrograms, and
as such, they further require a subsequent model to convert the spectrogram to
a waveform (i.e., a vocoder). This work proposes a diffusion probabilistic
end-to-end model for generating a raw speech waveform. The proposed model is
autoregressive, generating overlapping frames sequentially, where each frame is
conditioned on a portion of the previously generated one. Hence, our model can
effectively synthesize an unlimited speech duration while preserving
high-fidelity synthesis and temporal coherence. We implemented the proposed
model for unconditional and conditional speech generation, where the latter can
be driven by an input sequence of phonemes, amplitudes, and pitch values.
Working on the waveform directly has some empirical advantages. Specifically,
it allows the creation of local acoustic behaviors, like vocal fry, which makes
the overall waveform sounds more natural. Furthermore, the proposed diffusion
model is stochastic and not deterministic; therefore, each inference generates
a slightly different waveform variation, enabling abundance of valid
realizations. Experiments show that the proposed model generates speech with
superior quality compared with other state-of-the-art neural speech generation
systems. | cs.SD | DiffAR: Denoising Diffusion Autoregressive Model for Raw Speech Waveform Generation |
2023-10-02T17:40:01Z | http://arxiv.org/pdf/2310.01377v1 | http://arxiv.org/abs/2310.01377v1 | Ganqu Cui, Lifan Yuan, Ning Ding, Guanming Yao, Wei Zhu, Yuan Ni, Guotong Xie, Zhiyuan Liu, Maosong Sun | 2023-10-02T17:40:01Z | cs.CL, cs.AI, cs.LG | Reinforcement learning from human feedback (RLHF) has become a pivot
technique in aligning large language models (LLMs) with human preferences. In
RLHF practice, preference data plays a crucial role in bridging human
proclivity and LLMs. However, the scarcity of diverse, naturalistic datasets of
human preferences on LLM outputs at scale poses a great challenge to RLHF as
well as feedback learning research within the open-source community. Current
preference datasets, either proprietary or limited in size and prompt variety,
result in limited RLHF adoption in open-source models and hinder further
exploration. In this study, we propose ULTRAFEEDBACK, a large-scale,
high-quality, and diversified preference dataset designed to overcome these
limitations and foster RLHF development. To create ULTRAFEEDBACK, we compile a
diverse array of instructions and models from multiple sources to produce
comparative data. We meticulously devise annotation instructions and employ
GPT-4 to offer detailed feedback in both numerical and textual forms.
ULTRAFEEDBACK establishes a reproducible and expandable preference data
construction pipeline, serving as a solid foundation for future RLHF and
feedback learning research. Utilizing ULTRAFEEDBACK, we train various models to
demonstrate its effectiveness, including the reward model UltraRM, chat
language model UltraLM-13B-PPO, and critique model UltraCM. Experimental
results indicate that our models outperform existing open-source models,
achieving top performance across multiple benchmarks. Our data and models are
available at https://github.com/thunlp/UltraFeedback. | cs.CL | UltraFeedback: Boosting Language Models with High-quality Feedback |
2023-10-02T17:23:48Z | http://arxiv.org/pdf/2310.01361v2 | http://arxiv.org/abs/2310.01361v2 | Lirui Wang, Yiyang Ling, Zhecheng Yuan, Mohit Shridhar, Chen Bao, Yuzhe Qin, Bailin Wang, Huazhe Xu, Xiaolong Wang | 2024-01-21T21:01:12Z | cs.LG, cs.CL, cs.CV, cs.RO | Collecting large amounts of real-world interaction data to train general
robotic policies is often prohibitively expensive, thus motivating the use of
simulation data. However, existing methods for data generation have generally
focused on scene-level diversity (e.g., object instances and poses) rather than
task-level diversity, due to the human effort required to come up with and
verify novel tasks. This has made it challenging for policies trained on
simulation data to demonstrate significant task-level generalization. In this
paper, we propose to automatically generate rich simulation environments and
expert demonstrations by exploiting a large language models' (LLM) grounding
and coding ability. Our approach, dubbed GenSim, has two modes: goal-directed
generation, wherein a target task is given to the LLM and the LLM proposes a
task curriculum to solve the target task, and exploratory generation, wherein
the LLM bootstraps from previous tasks and iteratively proposes novel tasks
that would be helpful in solving more complex tasks. We use GPT4 to expand the
existing benchmark by ten times to over 100 tasks, on which we conduct
supervised finetuning and evaluate several LLMs including finetuned GPTs and
Code Llama on code generation for robotic simulation tasks. Furthermore, we
observe that LLMs-generated simulation programs can enhance task-level
generalization significantly when used for multitask policy training. We
further find that with minimal sim-to-real adaptation, the multitask policies
pretrained on GPT4-generated simulation tasks exhibit stronger transfer to
unseen long-horizon tasks in the real world and outperform baselines by 25%.
See the project website (https://liruiw.github.io/gensim) for code, demos, and
videos. | cs.LG | GenSim: Generating Robotic Simulation Tasks via Large Language Models |
2023-10-02T17:16:26Z | http://arxiv.org/pdf/2310.01352v4 | http://arxiv.org/abs/2310.01352v4 | Xi Victoria Lin, Xilun Chen, Mingda Chen, Weijia Shi, Maria Lomeli, Rich James, Pedro Rodriguez, Jacob Kahn, Gergely Szilvasy, Mike Lewis, Luke Zettlemoyer, Scott Yih | 2024-05-06T07:50:35Z | cs.CL, cs.AI | Retrieval-augmented language models (RALMs) improve performance by accessing
long-tail and up-to-date knowledge from external data stores, but are
challenging to build. Existing approaches require either expensive
retrieval-specific modifications to LM pre-training or use post-hoc integration
of the data store that leads to suboptimal performance. We introduce
Retrieval-Augmented Dual Instruction Tuning (RA-DIT), a lightweight fine-tuning
methodology that provides a third option by retrofitting any LLM with retrieval
capabilities. Our approach operates in two distinct fine-tuning steps: (1) one
updates a pre-trained LM to better use retrieved information, while (2) the
other updates the retriever to return more relevant results, as preferred by
the LM. By fine-tuning over tasks that require both knowledge utilization and
contextual awareness, we demonstrate that each stage yields significant
performance improvements, and using both leads to additional gains. Our best
model, RA-DIT 65B, achieves state-of-the-art performance across a range of
knowledge-intensive zero- and few-shot learning benchmarks, significantly
outperforming existing in-context RALM approaches by up to +8.9% in 0-shot
setting and +1.4% in 5-shot setting on average. | cs.CL | RA-DIT: Retrieval-Augmented Dual Instruction Tuning |
2023-10-02T17:02:57Z | http://arxiv.org/pdf/2310.01339v2 | http://arxiv.org/abs/2310.01339v2 | Miguel Ángel Medina-Ramírez, Cayetano Guerra-Artal, Mario Hernández-Tejera | 2023-10-23T17:15:55Z | cs.CL, cs.AI | Task-oriented dialogue systems (TODS) have become crucial for users to
interact with machines and computers using natural language. One of its key
components is the dialogue manager, which guides the conversation towards a
good goal for the user by providing the best possible response. Previous works
have proposed rule-based systems (RBS), reinforcement learning (RL), and
supervised learning (SL) as solutions for the correct dialogue management; in
other words, select the best response given input by the user. However, this
work argues that the leading cause of DMs not achieving maximum performance
resides in the quality of the datasets rather than the models employed thus
far; this means that dataset errors, like mislabeling, originate a large
percentage of failures in dialogue management. We studied the main errors in
the most widely used datasets, Multiwoz 2.1 and SGD, to demonstrate this
hypothesis. To do this, we have designed a synthetic dialogue generator to
fully control the amount and type of errors introduced in the dataset. Using
this generator, we demonstrated that errors in the datasets contribute
proportionally to the performance of the models | cs.CL | Improving Dialogue Management: Quality Datasets vs Models |
2023-10-02T17:01:56Z | http://arxiv.org/pdf/2310.01469v2 | http://arxiv.org/abs/2310.01469v2 | Jia-Yu Yao, Kun-Peng Ning, Zhen-Hui Liu, Mu-Nan Ning, Li Yuan | 2023-10-04T17:53:49Z | cs.CL, cs.AI | Large Language Models (LLMs), including GPT-3.5, LLaMA, and PaLM, seem to be
knowledgeable and able to adapt to many tasks. However, we still can not
completely trust their answer, since LLMs suffer from
hallucination--fabricating non-existent facts to cheat users without
perception. And the reasons for their existence and pervasiveness remain
unclear. In this paper, we demonstrate that non-sense prompts composed of
random tokens can also elicit the LLMs to respond with hallucinations. This
phenomenon forces us to revisit that hallucination may be another view of
adversarial examples, and it shares similar features with conventional
adversarial examples as the basic feature of LLMs. Therefore, we formalize an
automatic hallucination triggering method as the hallucination attack in an
adversarial way. Finally, we explore basic feature of attacked adversarial
prompts and propose a simple yet effective defense strategy. Our code is
released on GitHub. | cs.CL | LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examples |
2023-10-02T16:55:37Z | http://arxiv.org/pdf/2310.01468v3 | http://arxiv.org/abs/2310.01468v3 | Yizhe Zhang, Jiarui Lu, Navdeep Jaitly | 2024-02-20T21:24:43Z | cs.CL, cs.AI, cs.HC, cs.LG | Large language models (LLMs) are effective at answering questions that are
clearly asked. However, when faced with ambiguous queries they can act
unpredictably and produce incorrect outputs. This underscores the need for the
development of intelligent agents capable of asking clarification questions to
resolve ambiguities effectively. This capability requires complex
understanding, state tracking, reasoning and planning over multiple
conversational turns. However, directly measuring this can be challenging. In
this paper, we offer a surrogate problem which assesses an LLMs's capability to
deduce an entity unknown to itself, but revealed to a judge, by asking the
judge a series of queries. This \textit{entity-deducing game} can serve as an
evaluation framework to probe the conversational reasoning and planning
capabilities of language models. We systematically evaluate various LLMs and
discover significant differences in their performance on this task. We find
that strong LLMs like GPT-4 outperform human players by a large margin. We
further employ Behavior Cloning (BC) to examine whether a weaker model is
capable of imitating a stronger model and generalizing to data or domains,
using only the demonstrations from a stronger model. We finally propose to use
Reinforcement Learning to enhance reasoning and planning capacity of Vicuna
models through episodes of game playing, which lead to significant performance
improvement. We hope that this problem offers insights into how autonomous
agents could be trained to behave more intelligently in ambiguous
circumstances. | cs.CL | Probing the Multi-turn Planning Capabilities of LLMs via 20 Question Games |
2023-10-02T16:51:32Z | http://arxiv.org/pdf/2310.01334v2 | http://arxiv.org/abs/2310.01334v2 | Pingzhi Li, Zhenyu Zhang, Prateek Yadav, Yi-Lin Sung, Yu Cheng, Mohit Bansal, Tianlong Chen | 2024-03-14T11:01:15Z | cs.LG, cs.AI, cs.CL | Sparsely activated Mixture-of-Experts (SMoE) has shown promise to scale up
the learning capacity of neural networks, however, they have issues like (a)
High Memory Usage, due to duplication of the network layers into multiple
copies as experts; and (b) Redundancy in Experts, as common learning-based
routing policies suffer from representational collapse. Therefore, vanilla SMoE
models are memory inefficient and non-scalable, especially for
resource-constrained downstream scenarios. In this paper, we ask: Can we craft
a compact SMoE model by consolidating expert information? What is the best
recipe to merge multiple experts into fewer but more knowledgeable experts? Our
pilot investigation reveals that conventional model merging methods fail to be
effective in such expert merging for SMoE. The potential reasons are: (1)
redundant information overshadows critical experts; (2) appropriate neuron
permutation for each expert is missing to bring all of them in alignment. To
address this, we propose M-SMoE, which leverages routing statistics to guide
expert merging. Specifically, it starts with neuron permutation alignment for
experts; then, dominant experts and their "group members" are formed; lastly,
every expert group is merged into a single expert by utilizing each expert's
activation frequency as their weight for merging, thus diminishing the impact
of insignificant experts. Moreover, we observed that our proposed merging
promotes a low dimensionality in the merged expert's weight space, naturally
paving the way for additional compression. Hence, our final method, MC-SMoE
(i.e., Merge, then Compress SMoE), further decomposes the merged experts into
low-rank and structural sparse alternatives. Extensive experiments across 8
benchmarks validate the effectiveness of MC-SMoE. For instance, our MC-SMoE
achieves up to 80% memory and a 20% FLOPs reduction, with virtually no loss in
performance. | cs.LG | Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy |
2023-10-02T16:48:47Z | http://arxiv.org/pdf/2310.01329v2 | http://arxiv.org/abs/2310.01329v2 | Qingqing Cao, Sewon Min, Yizhong Wang, Hannaneh Hajishirzi | 2024-05-03T05:41:55Z | cs.CL, cs.AI | Retrieval augmentation addresses many critical problems in large language
models such as hallucination, staleness, and privacy leaks. However, running
retrieval-augmented language models (LMs) is slow and difficult to scale due to
processing large amounts of retrieved text. We introduce binary token
representations (BTR), which use 1-bit vectors to precompute every token in
passages, significantly reducing computation during inference. Despite the
potential loss of accuracy, our new calibration techniques and training
objectives restore performance. Combined with offline and runtime compression,
this only requires 127GB of disk space for encoding 3 billion tokens in
Wikipedia. Our experiments show that on five knowledge-intensive NLP tasks, BTR
accelerates state-of-the-art inference by up to 4x and reduces storage by over
100x while maintaining over 95% task performance. | cs.CL | BTR: Binary Token Representations for Efficient Retrieval Augmented Language Models |
2023-10-02T16:43:14Z | http://arxiv.org/pdf/2310.01467v1 | http://arxiv.org/abs/2310.01467v1 | Jingwei Sun, Ziyue Xu, Hongxu Yin, Dong Yang, Daguang Xu, Yiran Chen, Holger R. Roth | 2023-10-02T16:43:14Z | cs.CL, cs.AI | Pre-trained language models (PLM) have revolutionized the NLP landscape,
achieving stellar performances across diverse tasks. These models, while
benefiting from vast training data, often require fine-tuning on specific data
to cater to distinct downstream tasks. However, this data adaptation process
has inherent security and privacy concerns, primarily when leveraging
user-generated, device-residing data. Federated learning (FL) provides a
solution, allowing collaborative model fine-tuning without centralized data
collection. However, applying FL to finetune PLMs is hampered by challenges,
including restricted model parameter access, high computational requirements,
and communication overheads. This paper introduces Federated Black-box Prompt
Tuning (FedBPT), a framework designed to address these challenges. FedBPT does
not require the clients to access the model parameters. By focusing on training
optimal prompts and utilizing gradient-free optimization methods, FedBPT
reduces the number of exchanged variables, boosts communication efficiency, and
minimizes computational and storage costs. Experiments highlight the
framework's ability to drastically cut communication and memory costs while
maintaining competitive performance. Ultimately, FedBPT presents a promising
solution for efficient, privacy-preserving fine-tuning of PLM in the age of
large language models. | cs.CL | FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models |
2023-10-02T16:27:36Z | http://arxiv.org/pdf/2310.01320v3 | http://arxiv.org/abs/2310.01320v3 | Shenzhi Wang, Chang Liu, Zilong Zheng, Siyuan Qi, Shuo Chen, Qisen Yang, Andrew Zhao, Chaofei Wang, Shiji Song, Gao Huang | 2023-10-24T12:51:28Z | cs.AI, cs.CL, cs.CY, cs.LG, cs.MA | Recent breakthroughs in large language models (LLMs) have brought remarkable
success in the field of LLM-as-Agent. Nevertheless, a prevalent assumption is
that the information processed by LLMs is consistently honest, neglecting the
pervasive deceptive or misleading information in human society and AI-generated
content. This oversight makes LLMs susceptible to malicious manipulations,
potentially resulting in detrimental outcomes. This study utilizes the
intricate Avalon game as a testbed to explore LLMs' potential in deceptive
environments. Avalon, full of misinformation and requiring sophisticated logic,
manifests as a "Game-of-Thoughts". Inspired by the efficacy of humans'
recursive thinking and perspective-taking in the Avalon game, we introduce a
novel framework, Recursive Contemplation (ReCon), to enhance LLMs' ability to
identify and counteract deceptive information. ReCon combines formulation and
refinement contemplation processes; formulation contemplation produces initial
thoughts and speech, while refinement contemplation further polishes them.
Additionally, we incorporate first-order and second-order perspective
transitions into these processes respectively. Specifically, the first-order
allows an LLM agent to infer others' mental states, and the second-order
involves understanding how others perceive the agent's mental state. After
integrating ReCon with different LLMs, extensive experiment results from the
Avalon game indicate its efficacy in aiding LLMs to discern and maneuver around
deceptive information without extra fine-tuning and data. Finally, we offer a
possible explanation for the efficacy of ReCon and explore the current
limitations of LLMs in terms of safety, reasoning, speaking style, and format,
potentially furnishing insights for subsequent research. | cs.AI | Avalon's Game of Thoughts: Battle Against Deception through Recursive Contemplation |
2023-10-02T16:13:08Z | http://arxiv.org/pdf/2310.01307v2 | http://arxiv.org/abs/2310.01307v2 | Han Xu, Jie Ren, Pengfei He, Shenglai Zeng, Yingqian Cui, Amy Liu, Hui Liu, Jiliang Tang | 2023-10-03T16:40:35Z | cs.CL, cs.AI, cs.LG | ChatGPT is one of the most popular language models which achieve amazing
performance on various natural language tasks. Consequently, there is also an
urgent need to detect the texts generated ChatGPT from human written. One of
the extensively studied methods trains classification models to distinguish
both. However, existing studies also demonstrate that the trained models may
suffer from distribution shifts (during test), i.e., they are ineffective to
predict the generated texts from unseen language tasks or topics. In this work,
we aim to have a comprehensive investigation on these methods' generalization
behaviors under distribution shift caused by a wide range of factors, including
prompts, text lengths, topics, and language tasks. To achieve this goal, we
first collect a new dataset with human and ChatGPT texts, and then we conduct
extensive studies on the collected dataset. Our studies unveil insightful
findings which provide guidance for developing future methodologies or data
collection strategies for ChatGPT detection. | cs.CL | On the Generalization of Training-based ChatGPT Detection Methods |
2023-10-02T16:00:37Z | http://arxiv.org/pdf/2310.01299v1 | http://arxiv.org/abs/2310.01299v1 | Wei Sun, Mingxiao Li, Damien Sileo, Jesse Davis, Marie-Francine Moens | 2023-10-02T16:00:37Z | cs.CL, cs.AI | Medical Question Answering~(medical QA) systems play an essential role in
assisting healthcare workers in finding answers to their questions. However, it
is not sufficient to merely provide answers by medical QA systems because users
might want explanations, that is, more analytic statements in natural language
that describe the elements and context that support the answer. To do so, we
propose a novel approach for generating natural language explanations for
answers predicted by medical QA systems. As high-quality medical explanations
require additional medical knowledge, so that our system extract knowledge from
medical textbooks to enhance the quality of explanations during the explanation
generation process. Concretely, we designed an expectation-maximization
approach that makes inferences about the evidence found in these texts,
offering an efficient way to focus attention on lengthy evidence passages.
Experimental results, conducted on two datasets MQAE-diag and MQAE, demonstrate
the effectiveness of our framework for reasoning with textual evidence. Our
approach outperforms state-of-the-art models, achieving a significant
improvement of \textbf{6.86} and \textbf{9.43} percentage points on the Rouge-1
score; \textbf{8.23} and \textbf{7.82} percentage points on the Bleu-4 score on
the respective datasets. | cs.CL | Generating Explanations in Medical Question-Answering by Expectation Maximization Inference over Evidence |
2023-10-02T15:59:10Z | http://arxiv.org/pdf/2310.01297v1 | http://arxiv.org/abs/2310.01297v1 | Andrew D. Gordon, Carina Negreanu, José Cambronero, Rasika Chakravarthy, Ian Drosos, Hao Fang, Bhaskar Mitra, Hannah Richardson, Advait Sarkar, Stephanie Simmons, Jack Williams, Ben Zorn | 2023-10-02T15:59:10Z | cs.HC, cs.AI, cs.CL, cs.PL | Users are increasingly being warned to check AI-generated content for
correctness. Still, as LLMs (and other generative models) generate more complex
output, such as summaries, tables, or code, it becomes harder for the user to
audit or evaluate the output for quality or correctness. Hence, we are seeing
the emergence of tool-assisted experiences to help the user double-check a
piece of AI-generated content. We refer to these as co-audit tools. Co-audit
tools complement prompt engineering techniques: one helps the user construct
the input prompt, while the other helps them check the output response. As a
specific example, this paper describes recent research on co-audit tools for
spreadsheet computations powered by generative models. We explain why co-audit
experiences are essential for any application of generative AI where quality is
important and errors are consequential (as is common in spreadsheet
computations). We propose a preliminary list of principles for co-audit, and
outline research challenges. | cs.HC | Co-audit: tools to help humans double-check AI-generated content |
2023-10-02T15:43:53Z | http://arxiv.org/pdf/2310.01290v2 | http://arxiv.org/abs/2310.01290v2 | Wenxuan Ding, Shangbin Feng, Yuhan Liu, Zhaoxuan Tan, Vidhisha Balachandran, Tianxing He, Yulia Tsvetkov | 2024-06-25T06:25:41Z | cs.CL, cs.AI | We propose Knowledge Crosswords, a geometric knowledge reasoning benchmark
consisting of incomplete knowledge networks bounded by structured factual
constraints, where LLMs are tasked with inferring the missing facts to meet all
constraints. The novel setting of geometric knowledge reasoning necessitates
new LM abilities beyond existing atomic/linear multi-hop QA, such as
backtracking, verifying facts and constraints, reasoning with uncertainty, and
more. Knowledge Crosswords contains 2,101 individual problems, covering diverse
knowledge domains, and is further divided into three difficulty levels. We
conduct extensive experiments to evaluate existing LLMs and approaches on
Knowledge Crosswords. Results demonstrate that baseline approaches struggle
with larger knowledge networks and semantically-equivalent entity distractors.
In light of their limitations, we propose two new approaches, Staged Prompting
and Verify-All, to augment LLMs' abilities for error-aware backtracking and
constraint verification. Our Verify-All significantly outperforms prior methods
and is more robust towards problems in the hard subset. Further analysis shows
that geometric knowledge reasoning poses new challenges to LLMs' knowledge
abilities, particularly in robustness towards varying option orders, complex
structural constraints in knowledge networks, "none of the above" scenarios,
and more. | cs.CL | Knowledge Crosswords: Geometric Knowledge Reasoning with Large Language Models |
2023-10-02T15:16:31Z | http://arxiv.org/pdf/2310.01271v2 | http://arxiv.org/abs/2310.01271v2 | Xue Zongyue, Liu Huanghai, Hu Yiran, Kong Kangle, Wang Chenlu, Liu Yun, Shen Weixing | 2023-10-10T04:34:56Z | cs.CL, cs.IR | As a pivotal task in natural language processing, element extraction has
gained significance in the legal domain. Extracting legal elements from
judicial documents helps enhance interpretative and analytical capacities of
legal cases, and thereby facilitating a wide array of downstream applications
in various domains of law. Yet existing element extraction datasets are limited
by their restricted access to legal knowledge and insufficient coverage of
labels. To address this shortfall, we introduce a more comprehensive,
large-scale criminal element extraction dataset, comprising 15,831 judicial
documents and 159 labels. This dataset was constructed through two main steps:
first, designing the label system by our team of legal experts based on prior
legal research which identified critical factors driving and processes
generating sentencing outcomes in criminal cases; second, employing the legal
knowledge to annotate judicial documents according to the label system and
annotation guideline. The Legal Element ExtraCtion dataset (LEEC) represents
the most extensive and domain-specific legal element extraction dataset for the
Chinese legal system. Leveraging the annotated data, we employed various SOTA
models that validates the applicability of LEEC for Document Event Extraction
(DEE) task. The LEEC dataset is available on https://github.com/THUlawtech/LEEC . | cs.CL | LEEC: A Legal Element Extraction Dataset with an Extensive Domain-Specific Label System |
2023-10-02T14:51:16Z | http://arxiv.org/pdf/2310.01260v1 | http://arxiv.org/abs/2310.01260v1 | Yujian Betterest Li, Kai Wu | 2023-10-02T14:51:16Z | cs.CL, cs.AI | Prompt engineering is a new paradigm for enhancing the performance of trained
neural network models. For optimizing text-style prompts, existing methods
usually individually operate small portions of a text step by step, which
either breaks the fluency or could not globally adjust a prompt. Since large
language models (LLMs) have powerful ability of generating coherent texts token
by token, can we utilize LLMs for improving prompts? Based on this motivation,
in this paper, considering a trained LLM as a text generator, we attempt to
design a black-box evolution algorithm for automatically optimizing texts,
namely SPELL (Semantic Prompt Evolution based on a LLM). The proposed method is
evaluated with different LLMs and evolution parameters in different text tasks.
Experimental results show that SPELL could rapidly improve the prompts indeed.
We further explore the evolution process and discuss on the limitations,
potential possibilities and future work. | cs.CL | SPELL: Semantic Prompt Evolution based on a LLM |
2023-10-02T14:32:07Z | http://arxiv.org/pdf/2310.01248v2 | http://arxiv.org/abs/2310.01248v2 | Yuelyu Ji, Yuheng Song, Wei Wang, Ruoyi Xu, Zhongqian Xie, Huiyun Liu | 2023-10-12T20:10:37Z | cs.CL | Text generation in image-based platforms, particularly for music-related
content, requires precise control over text styles and the incorporation of
emotional expression. However, existing approaches often need help to control
the proportion of external factors in generated text and rely on discrete
inputs, lacking continuous control conditions for desired text generation. This
study proposes Continuous Parameterization for Controlled Text Generation
(CPCTG) to overcome these limitations. Our approach leverages a Language Model
(LM) as a style learner, integrating Semantic Cohesion (SC) and Emotional
Expression Proportion (EEP) considerations. By enhancing the reward method and
manipulating the CPCTG level, our experiments on playlist description and music
topic generation tasks demonstrate significant improvements in ROUGE scores,
indicating enhanced relevance and coherence in the generated text. | cs.CL | Improving Emotional Expression and Cohesion in Image-Based Playlist Description and Music Topics: A Continuous Parameterization Approach |
2023-10-02T14:01:36Z | http://arxiv.org/pdf/2310.01217v3 | http://arxiv.org/abs/2310.01217v3 | Markus Frohmann, Carolin Holtermann, Shahed Masoudian, Anne Lauscher, Navid Rekabsaz | 2024-05-17T14:23:08Z | cs.LG, cs.AI, cs.CL | Multi-task learning (MTL) has shown considerable practical benefits,
particularly when using language models (LMs). While this is commonly achieved
by learning $n$ tasks under a joint optimization procedure, some methods, such
as AdapterFusion, divide the problem into two stages: (i) task learning, where
knowledge specific to a task is encapsulated within sets of parameters (e.g.,
adapters), and (ii) transfer, where this already learned knowledge is leveraged
for a target task. This separation of concerns provides numerous benefits
(e.g., promoting reusability). However, current two-stage MTL introduces a
substantial number of additional parameters. We address this issue by
leveraging the usefulness of linearly scaling the output representations of
source adapters for transfer learning. We introduce ScaLearn, a simple and
highly parameter-efficient two-stage MTL method that capitalizes on the
knowledge of the source tasks by learning a minimal set of scaling parameters
that enable effective transfer to a target task. Our experiments on three
benchmarks (GLUE, SuperGLUE, and HumSet) and two encoder LMs show that ScaLearn
consistently outperforms strong baselines with a small number of transfer
parameters (~ $0.35$% of those of AdapterFusion). Remarkably, we observe that
ScaLearn maintains its strong abilities even when further reducing parameters,
achieving competitive results with only $8$ transfer parameters per target
task. Our proposed approach thus demonstrates the power of simple scaling as a
promise for more efficient task transfer. | cs.LG | ScaLearn: Simple and Highly Parameter-Efficient Task Transfer by Learning to Scale |
2023-10-02T13:53:03Z | http://arxiv.org/pdf/2310.01208v1 | http://arxiv.org/abs/2310.01208v1 | Zongxi Li, Xianming Li, Yuzhang Liu, Haoran Xie, Jing Li, Fu-lee Wang, Qing Li, Xiaoqin Zhong | 2023-10-02T13:53:03Z | cs.CL | The recent success of Large Language Models (LLMs) has gained significant
attention in both academia and industry. Substantial efforts have been made to
enhance the zero- and few-shot generalization capabilities of open-source LLMs
through finetuning. Currently, the prevailing approach is instruction-tuning,
which trains LLMs to complete real-world tasks by generating responses guided
by natural language instructions. It is worth noticing that such an approach
may underperform in sequence and token classification tasks. Unlike text
generation tasks, classification tasks have a limited label space, where
precise label prediction is more appreciated than generating diverse and
human-like responses. Prior research has unveiled that instruction-tuned LLMs
cannot outperform BERT, prompting us to explore the potential of leveraging
latent representations from LLMs for supervised label prediction. In this
paper, we introduce a label-supervised adaptation for LLMs, which aims to
finetuning the model with discriminant labels. We evaluate this approach with
Label Supervised LLaMA (LS-LLaMA), based on LLaMA-2-7B, a relatively
small-scale LLM, and can be finetuned on a single GeForce RTX4090 GPU. We
extract latent representations from the final LLaMA layer and project them into
the label space to compute the cross-entropy loss. The model is finetuned by
Low-Rank Adaptation (LoRA) to minimize this loss. Remarkably, without intricate
prompt engineering or external knowledge, LS-LLaMA substantially outperforms
LLMs ten times its size in scale and demonstrates consistent improvements
compared to robust baselines like BERT-Large and RoBERTa-Large in text
classification. Moreover, by removing the causal mask from decoders, LS-unLLaMA
achieves the state-of-the-art performance in named entity recognition (NER).
Our work will shed light on a novel approach to adapting LLMs for various
downstream tasks. | cs.CL | Label Supervised LLaMA Finetuning |
2023-10-02T13:48:16Z | http://arxiv.org/pdf/2310.01206v2 | http://arxiv.org/abs/2310.01206v2 | Atsuki Yamaguchi, Terufumi Morishita | 2023-10-03T13:19:40Z | cs.CL, cs.AI | We present appjsonify, a Python-based PDF-to-JSON conversion toolkit for
academic papers. It parses a PDF file using several visual-based document
layout analysis models and rule-based text processing approaches. appjsonify is
a flexible tool that allows users to easily configure the processing pipeline
to handle a specific format of a paper they wish to process. We are publicly
releasing appjsonify as an easy-to-install toolkit available via PyPI and
GitHub. | cs.CL | appjsonify: An Academic Paper PDF-to-JSON Conversion Toolkit |
2023-10-02T13:26:43Z | http://arxiv.org/pdf/2310.01188v2 | http://arxiv.org/abs/2310.01188v2 | Gabriele Sarti, Grzegorz Chrupała, Malvina Nissim, Arianna Bisazza | 2024-03-13T08:45:53Z | cs.CL, cs.AI, cs.HC, cs.LG, I.2.7 | Establishing whether language models can use contextual information in a
human-plausible way is important to ensure their trustworthiness in real-world
settings. However, the questions of when and which parts of the context affect
model generations are typically tackled separately, with current plausibility
evaluations being practically limited to a handful of artificial benchmarks. To
address this, we introduce Plausibility Evaluation of Context Reliance
(PECoRe), an end-to-end interpretability framework designed to quantify context
usage in language models' generations. Our approach leverages model internals
to (i) contrastively identify context-sensitive target tokens in generated
texts and (ii) link them to contextual cues justifying their prediction. We use
\pecore to quantify the plausibility of context-aware machine translation
models, comparing model rationales with human annotations across several
discourse-level phenomena. Finally, we apply our method to unannotated model
translations to identify context-mediated predictions and highlight instances
of (im)plausible context usage throughout generation. | cs.CL | Quantifying the Plausibility of Context Reliance in Neural Machine Translation |
2023-10-02T13:24:00Z | http://arxiv.org/pdf/2310.01459v1 | http://arxiv.org/abs/2310.01459v1 | Runcong Zhao, Wenjia Zhang, Jiazheng Li, Lixing Zhu, Yanran Li, Yulan He, Lin Gui | 2023-10-02T13:24:00Z | cs.CL, cs.AI, cs.HC | In this paper, we introduce NarrativePlay, a novel system that allows users
to role-play a fictional character and interact with other characters in
narratives such as novels in an immersive environment. We leverage Large
Language Models (LLMs) to generate human-like responses, guided by personality
traits extracted from narratives. The system incorporates auto-generated visual
display of narrative settings, character portraits, and character speech,
greatly enhancing user experience. Our approach eschews predefined sandboxes,
focusing instead on main storyline events extracted from narratives from the
perspective of a user-selected character. NarrativePlay has been evaluated on
two types of narratives, detective and adventure stories, where users can
either explore the world or improve their favorability with the narrative
characters through conversations. | cs.CL | NarrativePlay: Interactive Narrative Understanding |
2023-10-02T12:25:05Z | http://arxiv.org/pdf/2310.01138v1 | http://arxiv.org/abs/2310.01138v1 | Iffat Maab, Edison Marrese-Taylor, Yutaka Matsuo | 2023-10-02T12:25:05Z | cs.CL | Media bias detection requires comprehensive integration of information
derived from multiple news sources. Sentence-level political bias detection in
news is no exception, and has proven to be a challenging task that requires an
understanding of bias in consideration of the context. Inspired by the fact
that humans exhibit varying degrees of writing styles, resulting in a diverse
range of statements with different local and global contexts, previous work in
media bias detection has proposed augmentation techniques to exploit this fact.
Despite their success, we observe that these techniques introduce noise by
over-generalizing bias context boundaries, which hinders performance. To
alleviate this issue, we propose techniques to more carefully search for
context using a bias-sensitive, target-aware approach for data augmentation.
Comprehensive experiments on the well-known BASIL dataset show that when
combined with pre-trained models such as BERT, our augmentation techniques lead
to state-of-the-art results. Our approach outperforms previous methods
significantly, obtaining an F1-score of 58.15 over state-of-the-art bias
detection task. | cs.CL | Target-Aware Contextual Political Bias Detection in News |
2023-10-02T12:11:17Z | http://arxiv.org/pdf/2310.01132v4 | http://arxiv.org/abs/2310.01132v4 | Jacob Whitehill, Jennifer LoCasale-Crouch | 2024-04-16T21:30:14Z | cs.CL, cs.AI | With the aim to provide teachers with more specific, frequent, and actionable
feedback about their teaching, we explore how Large Language Models (LLMs) can
be used to estimate ``Instructional Support'' domain scores of the CLassroom
Assessment Scoring System (CLASS), a widely used observation protocol. We
design a machine learning architecture that uses either zero-shot prompting of
Meta's Llama2, and/or a classic Bag of Words (BoW) model, to classify
individual utterances of teachers' speech (transcribed automatically using
OpenAI's Whisper) for the presence of Instructional Support. Then, these
utterance-level judgments are aggregated over a 15-min observation session to
estimate a global CLASS score. Experiments on two CLASS-coded datasets of
toddler and pre-kindergarten classrooms indicate that (1) automatic CLASS
Instructional Support estimation accuracy using the proposed method (Pearson
$R$ up to $0.48$) approaches human inter-rater reliability (up to $R=0.55$);
(2) LLMs generally yield slightly greater accuracy than BoW for this task,
though the best models often combined features extracted from both LLM and BoW;
and (3) for classifying individual utterances, there is still room for
improvement of automated methods compared to human-level judgments. Finally,
(4) we illustrate how the model's outputs can be visualized at the utterance
level to provide teachers with explainable feedback on which utterances were
most positively or negatively correlated with specific CLASS dimensions. | cs.CL | Automated Evaluation of Classroom Instructional Support with LLMs and BoWs: Connecting Global Predictions to Specific Feedback |
2023-10-02T11:53:48Z | http://arxiv.org/pdf/2310.02281v1 | http://arxiv.org/abs/2310.02281v1 | Yajing Feng, Laurence Devillers | 2023-10-02T11:53:48Z | eess.AS, cs.AI, cs.CL, cs.LG, cs.SD | Speech Emotion recognition (SER) in call center conversations has emerged as
a valuable tool for assessing the quality of interactions between clients and
agents. In contrast to controlled laboratory environments, real-life
conversations take place under uncontrolled conditions and are subject to
contextual factors that influence the expression of emotions. In this paper, we
present our approach to constructing a large-scale reallife dataset (CusEmo)
for continuous SER in customer service call center conversations. We adopted
the dimensional emotion annotation approach to capture the subtlety,
complexity, and continuity of emotions in real-life call center conversations,
while annotating contextual information. The study also addresses the
challenges encountered during the application of the End-to-End (E2E) SER
system to the dataset, including determining the appropriate label sampling
rate and input segment length, as well as integrating contextual information
(interlocutor's gender and empathy level) with different weights using
multitask learning. The result shows that incorporating the empathy level
information improved the model's performance. | eess.AS | End-to-End Continuous Speech Emotion Recognition in Real-life Customer Service Call Center Conversations |
2023-10-02T11:49:05Z | http://arxiv.org/pdf/2310.01119v2 | http://arxiv.org/abs/2310.01119v2 | Jean Kaddour, Qi Liu | 2024-01-08T13:09:24Z | cs.CL, cs.LG | The in-context learning ability of large language models (LLMs) enables them
to generalize to novel downstream tasks with relatively few labeled examples.
However, they require enormous computational resources to be deployed.
Alternatively, smaller models can solve specific tasks if fine-tuned with
enough labeled examples. These examples, however, are expensive to obtain. In
pursuit of the best of both worlds, we study synthetic data generation of
fine-tuning training data via fine-tuned teacher LLMs to improve the downstream
performance of much smaller models. In four text classification and two text
generation tasks, we find that both data generation and annotation dramatically
improve the respective downstream model's performance, occasionally
necessitating only a minor fraction of the original training dataset. | cs.CL | Synthetic Data Generation in Low-Resource Settings via Fine-Tuning of Large Language Models |
2023-10-02T11:03:57Z | http://arxiv.org/pdf/2310.01089v1 | http://arxiv.org/abs/2310.01089v1 | Jianan Zhao, Le Zhuo, Yikang Shen, Meng Qu, Kai Liu, Michael Bronstein, Zhaocheng Zhu, Jian Tang | 2023-10-02T11:03:57Z | cs.CL, cs.LG | Large Language Models (LLMs) have gained the ability to assimilate human
knowledge and facilitate natural language interactions with both humans and
other LLMs. However, despite their impressive achievements, LLMs have not made
significant advancements in the realm of graph machine learning. This
limitation arises because graphs encapsulate distinct relational data, making
it challenging to transform them into natural language that LLMs understand. In
this paper, we bridge this gap with a novel framework, GraphText, that
translates graphs into natural language. GraphText derives a graph-syntax tree
for each graph that encapsulates both the node attributes and inter-node
relationships. Traversal of the tree yields a graph text sequence, which is
then processed by an LLM to treat graph tasks as text generation tasks.
Notably, GraphText offers multiple advantages. It introduces training-free
graph reasoning: even without training on graph data, GraphText with ChatGPT
can achieve on par with, or even surpassing, the performance of
supervised-trained graph neural networks through in-context learning (ICL).
Furthermore, GraphText paves the way for interactive graph reasoning, allowing
both humans and LLMs to communicate with the model seamlessly using natural
language. These capabilities underscore the vast, yet-to-be-explored potential
of LLMs in the domain of graph machine learning. | cs.CL | GraphText: Graph Reasoning in Text Space |
2023-10-02T11:03:20Z | http://arxiv.org/pdf/2310.01088v1 | http://arxiv.org/abs/2310.01088v1 | Kentaro Mitsui, Yukiya Hono, Kei Sawada | 2023-10-02T11:03:20Z | cs.CL, cs.LG, cs.SD, eess.AS | The advent of large language models (LLMs) has made it possible to generate
natural written dialogues between two agents. However, generating human-like
spoken dialogues from these written dialogues remains challenging. Spoken
dialogues have several unique characteristics: they frequently include
backchannels and laughter, and the smoothness of turn-taking significantly
influences the fluidity of conversation. This study proposes CHATS - CHatty
Agents Text-to-Speech - a discrete token-based system designed to generate
spoken dialogues based on written dialogues. Our system can generate speech for
both the speaker side and the listener side simultaneously, using only the
transcription from the speaker side, which eliminates the need for
transcriptions of backchannels or laughter. Moreover, CHATS facilitates natural
turn-taking; it determines the appropriate duration of silence after each
utterance in the absence of overlap, and it initiates the generation of
overlapping speech based on the phoneme sequence of the next utterance in case
of overlap. Experimental evaluations indicate that CHATS outperforms the
text-to-speech baseline, producing spoken dialogues that are more interactive
and fluid while retaining clarity and intelligibility. | cs.CL | Towards human-like spoken dialogue generation between AI agents from written dialogue |
2023-10-02T10:35:23Z | http://arxiv.org/pdf/2310.01074v2 | http://arxiv.org/abs/2310.01074v2 | Chenhan Yuan, Qianqian Xie, Jimin Huang, Sophia Ananiadou | 2023-10-08T12:45:18Z | cs.CL, cs.AI | Temporal reasoning is a crucial NLP task, providing a nuanced understanding
of time-sensitive contexts within textual data. Although recent advancements in
LLMs have demonstrated their potential in temporal reasoning, the predominant
focus has been on tasks such as temporal expression and temporal relation
extraction. These tasks are primarily designed for the extraction of direct and
past temporal cues and to engage in simple reasoning processes. A significant
gap remains when considering complex reasoning tasks such as event forecasting,
which requires multi-step temporal reasoning on events and prediction on the
future timestamp. Another notable limitation of existing methods is their
incapability to provide an illustration of their reasoning process, hindering
explainability. In this paper, we introduce the first task of explainable
temporal reasoning, to predict an event's occurrence at a future timestamp
based on context which requires multiple reasoning over multiple events, and
subsequently provide a clear explanation for their prediction. Our task offers
a comprehensive evaluation of both the LLMs' complex temporal reasoning
ability, the future event prediction ability, and explainability-a critical
attribute for AI applications. To support this task, we present the first
multi-source instruction-tuning dataset of explainable temporal reasoning
(ExpTime) with 26k derived from the temporal knowledge graph datasets and their
temporal reasoning paths, using a novel knowledge-graph-instructed-generation
strategy. Based on the dataset, we propose the first open-source LLM series
TimeLlaMA based on the foundation LlaMA2, with the ability of instruction
following for explainable temporal reasoning. We compare the performance of our
method and a variety of LLMs, where our method achieves the state-of-the-art
performance of temporal prediction and explanation. | cs.CL | Back to the Future: Towards Explainable Temporal Reasoning with Large Language Models |
2023-10-02T10:14:43Z | http://arxiv.org/pdf/2310.01061v2 | http://arxiv.org/abs/2310.01061v2 | Linhao Luo, Yuan-Fang Li, Gholamreza Haffari, Shirui Pan | 2024-02-24T03:03:12Z | cs.CL, cs.AI | Large language models (LLMs) have demonstrated impressive reasoning abilities
in complex tasks. However, they lack up-to-date knowledge and experience
hallucinations during reasoning, which can lead to incorrect reasoning
processes and diminish their performance and trustworthiness. Knowledge graphs
(KGs), which capture vast amounts of facts in a structured format, offer a
reliable source of knowledge for reasoning. Nevertheless, existing KG-based LLM
reasoning methods only treat KGs as factual knowledge bases and overlook the
importance of their structural information for reasoning. In this paper, we
propose a novel method called reasoning on graphs (RoG) that synergizes LLMs
with KGs to enable faithful and interpretable reasoning. Specifically, we
present a planning-retrieval-reasoning framework, where RoG first generates
relation paths grounded by KGs as faithful plans. These plans are then used to
retrieve valid reasoning paths from the KGs for LLMs to conduct faithful
reasoning. Furthermore, RoG not only distills knowledge from KGs to improve the
reasoning ability of LLMs through training but also allows seamless integration
with any arbitrary LLMs during inference. Extensive experiments on two
benchmark KGQA datasets demonstrate that RoG achieves state-of-the-art
performance on KG reasoning tasks and generates faithful and interpretable
reasoning results. | cs.CL | Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning |
2023-10-02T09:47:40Z | http://arxiv.org/pdf/2310.01045v2 | http://arxiv.org/abs/2310.01045v2 | Lei Li, Yekun Chai, Shuohuan Wang, Yu Sun, Hao Tian, Ningyu Zhang, Hua Wu | 2024-02-11T16:58:02Z | cs.CL | Reward modeling (a.k.a., preference modeling) is instrumental for aligning
large language models with human preferences, particularly within the context
of reinforcement learning from human feedback (RLHF). While conventional reward
models (RMs) have exhibited remarkable scalability, they oft struggle with
fundamental functionality such as arithmetic computation, code execution, and
factual lookup. In this paper, we propose a tool-augmented preference modeling
approach, named Themis, to address these limitations by empowering RMs with
access to external environments, including calculators and search engines. This
approach not only fosters synergy between tool utilization and reward grading
but also enhances interpretive capacity and scoring reliability. Our study
delves into the integration of external tools into RMs, enabling them to
interact with diverse external sources and construct task-specific tool
engagement and reasoning traces in an autoregressive manner. We validate our
approach across a wide range of domains, incorporating seven distinct external
tools. Our experimental results demonstrate a noteworthy overall improvement of
17.7% across eight tasks in preference ranking. Furthermore, our approach
outperforms Gopher 280B by 7.3% on TruthfulQA task in zero-shot evaluation. In
human evaluations, RLHF trained with Themis attains an average win rate of 32%
when compared to baselines across four distinct tasks. Additionally, we provide
a comprehensive collection of tool-related RM datasets, incorporating data from
seven distinct tool APIs, totaling 15,000 instances. We have made the code,
data, and model checkpoints publicly available to facilitate and inspire
further research
advancements\footnote{\url{https://github.com/ernie-research/Tool-Augmented-Reward-Model}}. | cs.CL | Tool-Augmented Reward Modeling |
2023-10-02T09:35:27Z | http://arxiv.org/pdf/2310.01041v2 | http://arxiv.org/abs/2310.01041v2 | Haozhe Ji, Pei Ke, Hongning Wang, Minlie Huang | 2024-06-05T08:23:11Z | cs.CL | Despite the remarkable advances in language modeling, current mainstream
decoding methods still struggle to generate texts that align with human texts
across different aspects. In particular, sampling-based methods produce
less-repetitive texts which are often disjunctive in discourse, while
search-based methods maintain topic coherence at the cost of increased
repetition. Overall, these methods fall short in achieving holistic alignment
across a broad range of aspects. In this work, we frame decoding from a
language model as an optimization problem with the goal of strictly matching
the expected performance with human texts measured by multiple metrics of
desired aspects simultaneously. The resulting decoding distribution enjoys an
analytical solution that scales the input language model distribution via a
sequence-level energy function defined by these metrics. And most importantly,
we prove that this induced distribution is guaranteed to improve the perplexity
on human texts, which suggests a better approximation to the underlying
distribution of human texts. To facilitate tractable sampling from this
globally normalized distribution, we adopt the Sampling-Importance-Resampling
technique. Experiments on various domains and model scales demonstrate the
superiority of our method in metrics alignment with human texts and human
evaluation over strong baselines. | cs.CL | Language Model Decoding as Direct Metrics Optimization |
2023-10-02T08:58:29Z | http://arxiv.org/pdf/2310.00996v3 | http://arxiv.org/abs/2310.00996v3 | Zhivar Sourati, Filip Ilievski, Pia Sommerauer, Yifan Jiang | 2024-04-23T01:16:00Z | cs.CL | As a core cognitive skill that enables the transferability of information
across domains, analogical reasoning has been extensively studied for both
humans and computational models. However, while cognitive theories of analogy
often focus on narratives and study the distinction between surface,
relational, and system similarities, existing work in natural language
processing has a narrower focus as far as relational analogies between word
pairs. This gap brings a natural question: can state-of-the-art large language
models (LLMs) detect system analogies between narratives? To gain insight into
this question and extend word-based relational analogies to relational system
analogies, we devise a comprehensive computational framework that
operationalizes dominant theories of analogy, using narrative elements to
create surface and system mappings. Leveraging the interplay between these
mappings, we create a binary task and benchmark for Analogical Reasoning on
Narratives (ARN), covering four categories of far (cross-domain)/near
(within-domain) analogies and disanalogies. We show that while all LLMs can
largely recognize near analogies, even the largest ones struggle with far
analogies in a zero-shot setting, with GPT4.0 scoring below random. Guiding the
models through solved examples and chain-of-thought reasoning enhances their
analogical reasoning ability. Yet, since even in the few-shot setting, the best
model only performs halfway between random and humans, ARN opens exciting
directions for computational analogical reasoners. | cs.CL | ARN: Analogical Reasoning on Narratives |
2023-10-02T08:22:34Z | http://arxiv.org/pdf/2310.00970v1 | http://arxiv.org/abs/2310.00970v1 | Yiyao Yu, Junjie Wang, Yuxiang Zhang, Lin Zhang, Yujiu Yang, Tetsuya Sakai | 2023-10-02T08:22:34Z | cs.CL | Artificial intelligence (AI) technologies should adhere to human norms to
better serve our society and avoid disseminating harmful or misleading
information, particularly in Conversational Information Retrieval (CIR).
Previous work, including approaches and datasets, has not always been
successful or sufficiently robust in taking human norms into consideration. To
this end, we introduce a workflow that integrates ethical alignment, with an
initial ethical judgment stage for efficient data screening. To address the
need for ethical judgment in CIR, we present the QA-ETHICS dataset, adapted
from the ETHICS benchmark, which serves as an evaluation tool by unifying
scenarios and label meanings. However, each scenario only considers one ethical
concept. Therefore, we introduce the MP-ETHICS dataset to evaluate a scenario
under multiple ethical concepts, such as justice and Deontology. In addition,
we suggest a new approach that achieves top performance in both binary and
multi-label ethical judgment tasks. Our research provides a practical method
for introducing ethical alignment into the CIR workflow. The data and code are
available at https://github.com/wanng-ide/ealm . | cs.CL | EALM: Introducing Multidimensional Ethical Alignment in Conversational Information Retrieval |
2023-10-02T06:57:45Z | http://arxiv.org/pdf/2310.00935v1 | http://arxiv.org/abs/2310.00935v1 | Yike Wang, Shangbin Feng, Heng Wang, Weijia Shi, Vidhisha Balachandran, Tianxing He, Yulia Tsvetkov | 2023-10-02T06:57:45Z | cs.CL | Large language models (LLMs) often encounter knowledge conflicts, scenarios
where discrepancy arises between the internal parametric knowledge of LLMs and
non-parametric information provided in the prompt context. In this work we ask
what are the desiderata for LLMs when a knowledge conflict arises and whether
existing LLMs fulfill them. We posit that LLMs should 1) identify knowledge
conflicts, 2) pinpoint conflicting information segments, and 3) provide
distinct answers or viewpoints in conflicting scenarios. To this end, we
introduce KNOWLEDGE CONFLICT, an evaluation framework for simulating contextual
knowledge conflicts and quantitatively evaluating to what extent LLMs achieve
these goals. KNOWLEDGE CONFLICT includes diverse and complex situations of
knowledge conflict, knowledge from diverse entities and domains, two synthetic
conflict creation methods, and settings with progressively increasing
difficulty to reflect realistic knowledge conflicts. Extensive experiments with
the KNOWLEDGE CONFLICT framework reveal that while LLMs perform well in
identifying the existence of knowledge conflicts, they struggle to determine
the specific conflicting knowledge and produce a response with distinct answers
amidst conflicting information. To address these challenges, we propose new
instruction-based approaches that augment LLMs to better achieve the three
goals. Further analysis shows that abilities to tackle knowledge conflicts are
greatly impacted by factors such as knowledge domain and prompt text, while
generating robust responses to knowledge conflict scenarios remains an open
research question. | cs.CL | Resolving Knowledge Conflicts in Large Language Models |
2023-10-02T06:57:25Z | http://arxiv.org/pdf/2310.01452v2 | http://arxiv.org/abs/2310.01452v2 | Duy C. Hoang, Quang H. Nguyen, Saurav Manchanda, MinLong Peng, Kok-Seng Wong, Khoa D. Doan | 2024-06-09T06:06:28Z | cs.CL, cs.AI | Despite outstanding performance in a variety of NLP tasks, recent studies
have revealed that NLP models are vulnerable to adversarial attacks that
slightly perturb the input to cause the models to misbehave. Among these
attacks, adversarial word-level perturbations are well-studied and effective
attack strategies. Since these attacks work in black-box settings, they do not
require access to the model architecture or model parameters and thus can be
detrimental to existing NLP applications. To perform an attack, the adversary
queries the victim model many times to determine the most important words in an
input text and to replace these words with their corresponding synonyms. In
this work, we propose a lightweight and attack-agnostic defense whose main goal
is to perplex the process of generating an adversarial example in these
query-based black-box attacks; that is to fool the textual fooler. This
defense, named AdvFooler, works by randomizing the latent representation of the
input at inference time. Different from existing defenses, AdvFooler does not
necessitate additional computational overhead during training nor relies on
assumptions about the potential adversarial perturbation set while having a
negligible impact on the model's accuracy. Our theoretical and empirical
analyses highlight the significance of robustness resulting from confusing the
adversary via randomizing the latent space, as well as the impact of
randomization on clean accuracy. Finally, we empirically demonstrate near
state-of-the-art robustness of AdvFooler against representative adversarial
word-level attacks on two benchmark datasets. | cs.CL | Fooling the Textual Fooler via Randomizing Latent Representations |
2023-10-02T05:23:34Z | http://arxiv.org/pdf/2310.00905v2 | http://arxiv.org/abs/2310.00905v2 | Wenxuan Wang, Zhaopeng Tu, Chang Chen, Youliang Yuan, Jen-tse Huang, Wenxiang Jiao, Michael R. Lyu | 2024-06-20T14:15:23Z | cs.CL, cs.AI | Safety lies at the core of developing and deploying large language models
(LLMs). However, previous safety benchmarks only concern the safety in one
language, e.g. the majority language in the pretraining data such as English.
In this work, we build the first multilingual safety benchmark for LLMs,
XSafety, in response to the global deployment of LLMs in practice. XSafety
covers 14 kinds of commonly used safety issues across 10 languages that span
several language families. We utilize XSafety to empirically study the
multilingual safety for 4 widely-used LLMs, including both close-API and
open-source models. Experimental results show that all LLMs produce
significantly more unsafe responses for non-English queries than English ones,
indicating the necessity of developing safety alignment for non-English
languages. In addition, we propose several simple and effective prompting
methods to improve the multilingual safety of ChatGPT by evoking safety
knowledge and improving cross-lingual generalization of safety alignment. Our
prompting method can significantly reduce the ratio of unsafe responses from
19.1% to 9.7% for non-English queries. We release our data at
https://github.com/Jarviswang94/Multilingual_safety_benchmark. | cs.CL | All Languages Matter: On the Multilingual Safety of Large Language Models |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.