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
|
|---|---|---|---|---|---|---|---|---|
2024-06-24T11:16:31Z | http://arxiv.org/pdf/2406.16536v1 | http://arxiv.org/abs/2406.16536v1 | Kunting Li, Yong Hu, Liang He, Fandong Meng, Jie Zhou | 2024-06-24T11:16:31Z | cs.CL | Chinese Spell Checking (CSC) aims to detect and correct spelling errors in
sentences. Despite Large Language Models (LLMs) exhibit robust capabilities and
are widely applied in various tasks, their performance on CSC is often
unsatisfactory. We find that LLMs fail to meet the Chinese character-level
constraints of the CSC task, namely equal length and phonetic similarity,
leading to a performance bottleneck. Further analysis reveal that this issue
stems from the granularity of tokenization, as current mixed character-word
tokenization struggles to satisfy these character-level constraints. To address
this issue, we propose C-LLM, a Large Language Model-based Chinese Spell
Checking method that learns to check errors Character by Character.
Character-level tokenization enables the model to learn character-level
alignment, effectively mitigating issues related to character-level
constraints. Furthermore, CSC is simplified to replication-dominated and
substitution-supplemented tasks. Experiments on two CSC benchmarks demonstrate
that C-LLM achieves an average improvement of 10% over existing methods.
Specifically, it shows a 2.1% improvement in general scenarios and a
significant 12% improvement in vertical domain scenarios, establishing
state-of-the-art performance. The source code can be accessed at
https://github.com/ktlKTL/C-LLM. | cs.CL | C-LLM: Learn to Check Chinese Spelling Errors Character by Character |
2024-06-24T11:16:26Z | http://arxiv.org/pdf/2406.16535v1 | http://arxiv.org/abs/2406.16535v1 | Hakaze Cho, Yoshihiro Sakai, Mariko Kato, Kenshiro Tanaka, Akira Ishii, Naoya Inoue | 2024-06-24T11:16:26Z | cs.CL, cs.AI, cs.LG | In-Context Learning (ICL) typically utilizes classification criteria from
probabilities of manually selected label tokens. However, we argue that such
token-based classification criteria lead to suboptimal decision boundaries,
despite delicate calibrations through translation and constrained rotation. To
address this problem, we propose Hidden Calibration, which renounces token
probabilities and uses the nearest centroid classifier on the LM's last hidden
states. In detail, we use the nearest centroid classification on the hidden
states, assigning the category of the nearest centroid previously observed from
a few-shot calibration set to the test sample as the predicted label. Our
experiments on 3 models and 10 classification datasets indicate that Hidden
Calibration consistently outperforms current token-based calibrations by about
20%. Our further analysis demonstrates that Hidden Calibration finds better
classification criteria with less inter-categories overlap, and LMs provide
linearly separable intra-category clusters with the help of demonstrations,
which supports Hidden Calibration and gives new insights into the conventional
ICL. | cs.CL | Token-based Decision Criteria Are Suboptimal in In-context Learning |
2024-06-24T11:08:28Z | http://arxiv.org/pdf/2406.16529v1 | http://arxiv.org/abs/2406.16529v1 | Hao Yue, Shaopeng Lai, Chengyi Yang, Liang Zhang, Junfeng Yao, Jinsong Su | 2024-06-24T11:08:28Z | cs.CL | Cross-document Relation Extraction aims to predict the relation between
target entities located in different documents. In this regard, the dominant
models commonly retain useful information for relation prediction via bridge
entities, which allows the model to elaborately capture the intrinsic
interdependence between target entities. However, these studies ignore the
non-bridge entities, each of which co-occurs with only one target entity and
offers the semantic association between target entities for relation
prediction. Besides, the commonly-used dataset--CodRED contains substantial NA
instances, leading to the prediction bias during inference. To address these
issues, in this paper, we propose a novel graph-based cross-document RE model
with non-bridge entity enhancement and prediction debiasing. Specifically, we
use a unified entity graph to integrate numerous non-bridge entities with
target entities and bridge entities, modeling various associations between
them, and then use a graph recurrent network to encode this graph. Finally, we
introduce a novel debiasing strategy to calibrate the original prediction
distribution. Experimental results on the closed and open settings show that
our model significantly outperforms all baselines, including the GPT-3.5-turbo
and InstructUIE, achieving state-of-the-art performance. Particularly, our
model obtains 66.23% and 55.87% AUC points in the official
leaderboard\footnote{\url{https://codalab.lisn.upsaclay.fr/competitions/3770#results}}
under the two settings, respectively, ranking the first place in all
submissions since December 2023. Our code is available at
https://github.com/DeepLearnXMU/CoRE-NEPD. | cs.CL | Towards Better Graph-based Cross-document Relation Extraction via Non-bridge Entity Enhancement and Prediction Debiasing |
2024-06-24T11:07:01Z | http://arxiv.org/pdf/2406.16528v1 | http://arxiv.org/abs/2406.16528v1 | Anthony G Cohn, Robert E Blackwell | 2024-06-24T11:07:01Z | cs.CL | We investigate the abilities of a representative set of Large language Models
(LLMs) to reason about cardinal directions (CDs). To do so, we create two
datasets: the first, co-created with ChatGPT, focuses largely on recall of
world knowledge about CDs; the second is generated from a set of templates,
comprehensively testing an LLM's ability to determine the correct CD given a
particular scenario. The templates allow for a number of degrees of variation
such as means of locomotion of the agent involved, and whether set in the first
, second or third person. Even with a temperature setting of zero, Our
experiments show that although LLMs are able to perform well in the simpler
dataset, in the second more complex dataset no LLM is able to reliably
determine the correct CD, even with a temperature setting of zero. | cs.CL | Evaluating the Ability of Large Language Models to Reason about Cardinal Directions |
2024-06-24T11:04:43Z | http://arxiv.org/pdf/2406.16527v1 | http://arxiv.org/abs/2406.16527v1 | Zheng Fang, Miguel Arana-Catania, Felix-Anselm van Lier, Juliana Outes Velarde, Harry Bregazzi, Mara Airoldi, Eleanor Carter, Rob Procter | 2024-06-24T11:04:43Z | cs.CL, cs.CY, cs.DL, cs.LG | The sheer number of research outputs published every year makes systematic
reviewing increasingly time- and resource-intensive. This paper explores the
use of machine learning techniques to help navigate the systematic review
process. ML has previously been used to reliably 'screen' articles for review -
that is, identify relevant articles based on reviewers' inclusion criteria. The
application of ML techniques to subsequent stages of a review, however, such as
data extraction and evidence mapping, is in its infancy. We therefore set out
to develop a series of tools that would assist in the profiling and analysis of
1,952 publications on the theme of 'outcomes-based contracting'. Tools were
developed for the following tasks: assign publications into 'policy area'
categories; identify and extract key information for evidence mapping, such as
organisations, laws, and geographical information; connect the evidence base to
an existing dataset on the same topic; and identify subgroups of articles that
may share thematic content. An interactive tool using these techniques and a
public dataset with their outputs have been released. Our results demonstrate
the utility of ML techniques to enhance evidence accessibility and analysis
within the systematic review processes. These efforts show promise in
potentially yielding substantial efficiencies for future systematic reviewing
and for broadening their analytical scope. Our work suggests that there may be
implications for the ease with which policymakers and practitioners can access
evidence. While ML techniques seem poised to play a significant role in
bridging the gap between research and policy by offering innovative ways of
gathering, accessing, and analysing data from systematic reviews, we also
highlight their current limitations and the need to exercise caution in their
application, particularly given the potential for errors and biases. | cs.CL | SyROCCo: Enhancing Systematic Reviews using Machine Learning |
2024-06-24T10:59:26Z | http://arxiv.org/pdf/2406.16524v1 | http://arxiv.org/abs/2406.16524v1 | Haryo Akbarianto Wibowo, Thamar Solorio, Alham Fikri Aji | 2024-06-24T10:59:26Z | cs.CL, 68T50 | Knowledge distillation (KD) has proven to be a successful strategy to improve
the performance of a smaller model in many NLP tasks. However, most of the work
in KD only explores monolingual scenarios. In this paper, we investigate the
value of KD in multilingual settings. We find the significance of KD and model
initialization by analyzing how well the student model acquires multilingual
knowledge from the teacher model. Our proposed method emphasizes copying the
teacher model's weights directly to the student model to enhance
initialization. Our finding shows that model initialization using copy-weight
from the fine-tuned teacher contributes the most compared to the distillation
process itself across various multilingual settings. Furthermore, we
demonstrate that efficient weight initialization preserves multilingual
capabilities even in low-resource scenarios. | cs.CL | The Privileged Students: On the Value of Initialization in Multilingual Knowledge Distillation |
2024-06-24T10:55:31Z | http://arxiv.org/pdf/2406.16521v1 | http://arxiv.org/abs/2406.16521v1 | Jimin Sohn, Jeihee Cho, Junyong Lee, Songmu Heo, Ji-Eun Han, David R. Mortensen | 2024-06-24T10:55:31Z | cs.CL, cs.AI | Positive thinking is thought to be an important component of self-motivation
in various practical fields such as education and the workplace. Previous work,
including sentiment transfer and positive reframing, has focused on the
positive side of language. However, self-motivation that drives people to reach
their goals has not yet been studied from a computational perspective.
Moreover, negative feedback has not yet been explored, even though positive and
negative feedback are both necessary to grow self-motivation. To facilitate
self-motivation, we propose CArrot and STICk (CASTIC) dataset, consisting of
12,590 sentences with 5 different strategies for enhancing self-motivation. Our
data and code are publicly available at here. | cs.CL | Carrot and Stick: Inducing Self-Motivation with Positive & Negative Feedback |
2024-06-24T10:27:07Z | http://arxiv.org/pdf/2406.16508v1 | http://arxiv.org/abs/2406.16508v1 | Sho Takase, Ryokan Ri, Shun Kiyono, Takuya Kato | 2024-06-24T10:27:07Z | cs.CL | This paper empirically investigates the relationship between subword
vocabulary size and the performance of large language models (LLMs) to provide
insights on how to define the vocabulary size. Experimental results show that
larger vocabulary sizes lead to better performance in LLMs. Moreover, we
consider a continual training scenario where a pre-trained language model is
trained on a different target language. We introduce a simple method to use a
new vocabulary instead of the pre-defined one. We show that using the new
vocabulary outperforms the model with the vocabulary used in pre-training. | cs.CL | Large Vocabulary Size Improves Large Language Models |
2024-06-24T10:05:23Z | http://arxiv.org/pdf/2406.16495v2 | http://arxiv.org/abs/2406.16495v2 | Jingze Shi, Ting Xie, Bingheng Wu, Chunjun Zheng, Kai Wang | 2024-06-25T02:20:14Z | cs.CL, cs.AI | Recent research has shown that combining Mamba with Transformer architecture,
which has selective state space and quadratic self-attention mechanism,
outperforms using Mamba or Transformer architecture alone in language modeling
tasks. The quadratic self-attention mechanism effectively alleviates the
shortcomings of selective state space in handling long-term dependencies of any
element in the sequence. We propose a position information injection method
that connects the selective state space model with the quadratic attention, and
integrates these two architectures with hybrid experts with cross-sharing
domains, so that we can enjoy the advantages of both. We design a new
architecture with a more biomimetic idea: Observer-Thinker-Conceiver-Expresser
(OTCE), which can compete with well-known medium-scale open-source language
models on a small scale in language modeling tasks. | cs.CL | OTCE: Hybrid SSM and Attention with Cross Domain Mixture of Experts to construct Observer-Thinker-Conceiver-Expresser |
2024-06-24T09:57:44Z | http://arxiv.org/pdf/2406.16490v1 | http://arxiv.org/abs/2406.16490v1 | Hoorieh Sabzevari, Mohammadmostafa Rostamkhani, Sauleh Eetemadi | 2024-06-24T09:57:44Z | cs.CL | This study investigates the performance of the zero-shot method in
classifying data using three large language models, alongside two models with
large input token sizes and the two pre-trained models on legal data. Our main
dataset comes from the domain of U.S. civil procedure. It includes summaries of
legal cases, specific questions, potential answers, and detailed explanations
for why each solution is relevant, all sourced from a book aimed at law
students. By comparing different methods, we aimed to understand how
effectively they handle the complexities found in legal datasets. Our findings
show how well the zero-shot method of large language models can understand
complicated data. We achieved our highest F1 score of 64% in these experiments. | cs.CL | eagerlearners at SemEval2024 Task 5: The Legal Argument Reasoning Task in Civil Procedure |
2024-06-24T09:55:31Z | http://arxiv.org/pdf/2406.16489v1 | http://arxiv.org/abs/2406.16489v1 | Adam Frej, Adrian Kaminski, Piotr Marciniak, Szymon Szmajdzinski, Soveatin Kuntur, Anna Wroblewska | 2024-06-24T09:55:31Z | cs.CL | This study addresses the critical challenge of detecting DeepFake tweets by
leveraging advanced natural language processing (NLP) techniques to distinguish
between genuine and AI-generated texts. Given the increasing prevalence of
misinformation, our research utilizes the TweepFake dataset to train and
evaluate various machine learning models. The objective is to identify
effective strategies for recognizing DeepFake content, thereby enhancing the
integrity of digital communications. By developing reliable methods for
detecting AI-generated misinformation, this work contributes to a more
trustworthy online information environment. | cs.CL | Deepfake tweets automatic detection |
2024-06-24T09:32:28Z | http://arxiv.org/pdf/2406.16478v1 | http://arxiv.org/abs/2406.16478v1 | Lucie Galland, Catherine Pelachaud, Florian Pecune | 2024-06-24T09:32:28Z | cs.CL | The study of multimodal interaction in therapy can yield a comprehensive
understanding of therapist and patient behavior that can be used to develop a
multimodal virtual agent supporting therapy. This investigation aims to uncover
how therapists skillfully blend therapy's task goal (employing classical steps
of Motivational Interviewing) with the social goal (building a trusting
relationship and expressing empathy). Furthermore, we seek to categorize
patients into various ``types'' requiring tailored therapeutic approaches. To
this intent, we present multimodal annotations of a corpus consisting of
simulated motivational interviewing conversations, wherein actors portray the
roles of patients and therapists. We introduce EMMI, composed of two publicly
available MI corpora, AnnoMI and the Motivational Interviewing Dataset, for
which we add multimodal annotations. We analyze these annotations to
characterize functional behavior for developing a virtual agent performing
motivational interviews emphasizing social and empathic behaviors. Our analysis
found three clusters of patients expressing significant differences in behavior
and adaptation of the therapist's behavior to those types. This shows the
importance of a therapist being able to adapt their behavior depending on the
current situation within the dialog and the type of user. | cs.CL | EMMI -- Empathic Multimodal Motivational Interviews Dataset: Analyses and Annotations |
2024-06-24T09:30:36Z | http://arxiv.org/pdf/2406.16477v1 | http://arxiv.org/abs/2406.16477v1 | Aiwen Jiang, Zhi Wei, Long Peng, Feiqiang Liu, Wenbo Li, Mingwen Wang | 2024-06-24T09:30:36Z | cs.CV, cs.CL | Image super-resolution pursuits reconstructing high-fidelity high-resolution
counterpart for low-resolution image. In recent years, diffusion-based models
have garnered significant attention due to their capabilities with rich prior
knowledge. The success of diffusion models based on general text prompts has
validated the effectiveness of textual control in the field of text2image.
However, given the severe degradation commonly presented in low-resolution
images, coupled with the randomness characteristics of diffusion models,
current models struggle to adequately discern semantic and degradation
information within severely degraded images. This often leads to obstacles such
as semantic loss, visual artifacts, and visual hallucinations, which pose
substantial challenges for practical use. To address these challenges, this
paper proposes to leverage degradation-aligned language prompt for accurate,
fine-grained, and high-fidelity image restoration. Complementary priors
including semantic content descriptions and degradation prompts are explored.
Specifically, on one hand, image-restoration prompt alignment decoder is
proposed to automatically discern the degradation degree of LR images, thereby
generating beneficial degradation priors for image restoration. On the other
hand, much richly tailored descriptions from pretrained multimodal large
language model elicit high-level semantic priors closely aligned with human
perception, ensuring fidelity control for image restoration. Comprehensive
comparisons with state-of-the-art methods have been done on several popular
synthetic and real-world benchmark datasets. The quantitative and qualitative
analysis have demonstrated that the proposed method achieves a new
state-of-the-art perceptual quality level, especially in real-world cases based
on reference-free metrics. | cs.CV | DaLPSR: Leverage Degradation-Aligned Language Prompt for Real-World Image Super-Resolution |
2024-06-24T09:18:15Z | http://arxiv.org/pdf/2406.16469v1 | http://arxiv.org/abs/2406.16469v1 | Yujin Baek, ChaeHun Park, Jaeseok Kim, Yu-Jung Heo, Du-Seong Chang, Jaegul Choo | 2024-06-24T09:18:15Z | cs.CL, cs.CV | To create culturally inclusive vision-language models (VLMs), the foremost
requirement is developing a test benchmark that can diagnose the models'
ability to respond to questions reflecting cultural elements. This paper
addresses the necessity for such benchmarks, noting that existing research has
relied on human annotators' manual efforts, which impedes diversity and
efficiency. We propose a semi-automated pipeline for constructing cultural VLM
benchmarks to enhance diversity and efficiency. This pipeline leverages
human-VLM collaboration, where VLMs generate questions based on guidelines,
human-annotated examples, and image-wise relevant knowledge, which are then
reviewed by native speakers for quality and cultural relevance. The
effectiveness of our adaptable pipeline is demonstrated through a specific
application: creating a dataset tailored to Korean culture, dubbed K-Viscuit.
The resulting benchmark features two types of questions: Type 1 questions
measure visual recognition abilities, while Type 2 assess fine-grained visual
reasoning skills. This ensures a thorough diagnosis of VLM models across
various aspects. Our evaluation using K-Viscuit revealed that open-source
models notably lag behind proprietary models in understanding Korean culture,
highlighting areas for improvement. We provided diverse analyses of VLM
performance across different cultural aspects. Besides, we explored the
potential of incorporating external knowledge retrieval to enhance the
generation process, suggesting future directions for improving cultural
interpretation ability of VLMs. Our dataset and code will be made publicly
available. | cs.CL | Evaluating Visual and Cultural Interpretation: The K-Viscuit Benchmark with Human-VLM Collaboration |
2024-06-24T09:13:42Z | http://arxiv.org/pdf/2406.16464v2 | http://arxiv.org/abs/2406.16464v2 | Junjie Chen, Subin Huang | 2024-06-26T05:40:16Z | cs.CL, cs.AI, cs.CV | The prevalence of sarcasm in social media, conveyed through text-image
combinations, presents significant challenges for sentiment analysis and
intention mining. Current multi-modal sarcasm detection methods have been
proven to struggle with biases from spurious cues, leading to a superficial
understanding of the complex interactions between text and image. To address
these issues, we propose InterCLIP-MEP, a robust framework for multi-modal
sarcasm detection. InterCLIP-MEP introduces a refined variant of CLIP,
Interactive CLIP (InterCLIP), as the backbone, enhancing sample representations
by embedding cross-modality information in each encoder. Furthermore, a novel
training strategy is designed to adapt InterCLIP for a Memory-Enhanced
Predictor (MEP). MEP uses dynamic dual-channel memory to store valuable
historical knowledge of test samples and then leverages this memory as a
non-parametric classifier to derive the final prediction. By using InterCLIP to
encode text-image interactions more effectively and incorporating MEP,
InterCLIP-MEP offers a more robust recognition of multi-modal sarcasm.
Experiments demonstrate that InterCLIP-MEP achieves state-of-the-art
performance on the MMSD2.0 benchmark. Code and data are available at
https://github.com/CoderChen01/InterCLIP-MEP. | cs.CL | InterCLIP-MEP: Interactive CLIP and Memory-Enhanced Predictor for Multi-modal Sarcasm Detection |
2024-06-24T08:43:21Z | http://arxiv.org/pdf/2406.16450v1 | http://arxiv.org/abs/2406.16450v1 | Xiuying Wei, Skander Moalla, Razvan Pascanu, Caglar Gulcehre | 2024-06-24T08:43:21Z | cs.CL | State-of-the-art results in large language models (LLMs) often rely on scale,
which becomes computationally expensive. This has sparked a research agenda to
reduce these models' parameter count and computational costs without
significantly impacting their performance. Our study focuses on
transformer-based LLMs, specifically targeting the computationally intensive
feedforward networks (FFN), which are less studied than attention blocks. We
consider three candidate linear layer approximations in the FFN by combining
efficient low-rank and block-diagonal matrices. In contrast to many previous
works that examined these approximations, our study i) explores these
structures from the training-from-scratch perspective, ii) scales up to 1.3B
parameters, and iii) is conducted within recent Transformer-based LLMs rather
than convolutional architectures. We first demonstrate they can lead to actual
computational gains in various scenarios, including online decoding when using
a pre-merge technique. Additionally, we propose a novel training regime, called
\textit{self-guided training}, aimed at improving the poor training dynamics
that these approximations exhibit when used from initialization. Experiments on
the large RefinedWeb dataset show that our methods are both efficient and
effective for training and inference. Interestingly, these structured FFNs
exhibit steeper scaling curves than the original models. Further applying
self-guided training to the structured matrices with 32\% FFN parameters and
2.5$\times$ speed-up enables only a 0.4 perplexity increase under the same
training FLOPs. Finally, we develop the wide and structured networks surpassing
the current medium-sized and large-sized Transformer in perplexity and
throughput performance. Our code is available at
\url{https://github.com/CLAIRE-Labo/StructuredFFN/tree/main}. | cs.CL | Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers |
2024-06-24T08:32:48Z | http://arxiv.org/pdf/2406.16441v1 | http://arxiv.org/abs/2406.16441v1 | Tao Sun, Linzheng Chai, Jian Yang, Yuwei Yin, Hongcheng Guo, Jiaheng Liu, Bing Wang, Liqun Yang, Zhoujun Li | 2024-06-24T08:32:48Z | cs.CL | Intermediate reasoning or acting steps have successfully improved large
language models (LLMs) for handling various downstream natural language
processing (NLP) tasks. When applying LLMs for code generation, recent works
mainly focus on directing the models to articulate intermediate
natural-language reasoning steps, as in chain-of-thought (CoT) prompting, and
then output code with the natural language or other structured intermediate
steps. However, such output is not suitable for code translation or generation
tasks since the standard CoT has different logical structures and forms of
expression with the code. In this work, we introduce the universal code
(UniCode) as the intermediate representation. It is a description of algorithm
steps using a mix of conventions of programming languages, such as assignment
operator, conditional operator, and loop. Hence, we collect an instruction
dataset UniCoder-Instruct to train our model UniCoder on multi-task learning
objectives. UniCoder-Instruct comprises natural-language questions, code
solutions, and the corresponding universal code. The alignment between the
intermediate universal code representation and the final code solution
significantly improves the quality of the generated code. The experimental
results demonstrate that UniCoder with the universal code significantly
outperforms the previous prompting methods by a large margin, showcasing the
effectiveness of the structural clues in pseudo-code. | cs.CL | UniCoder: Scaling Code Large Language Model via Universal Code |
2024-06-24T08:06:56Z | http://arxiv.org/pdf/2406.16416v1 | http://arxiv.org/abs/2406.16416v1 | Xue zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Yufeng Chen, Jinan Xu, Jie Zhou | 2024-06-24T08:06:56Z | cs.CL | Multilingual knowledge editing (MKE) aims to simultaneously revise factual
knowledge across multilingual languages within large language models (LLMs).
However, most existing MKE methods just adapt existing monolingual editing
methods to multilingual scenarios, overlooking the deep semantic connections of
the same factual knowledge between different languages, thereby limiting edit
performance. To address this issue, we first investigate how LLMs represent
multilingual factual knowledge and discover that the same factual knowledge in
different languages generally activates a shared set of neurons, which we call
language-agnostic factual neurons. These neurons represent the semantic
connections between multilingual knowledge and are mainly located in certain
layers. Inspired by this finding, we propose a new MKE method by locating and
modifying Language-Agnostic Factual Neurons (LAFN) to simultaneously edit
multilingual knowledge. Specifically, we first generate a set of paraphrases
for each multilingual knowledge to be edited to precisely locate the
corresponding language-agnostic factual neurons. Then we optimize the update
values for modifying these located neurons to achieve simultaneous modification
of the same factual knowledge in multiple languages. Experimental results on
Bi-ZsRE and MzsRE benchmarks demonstrate that our method outperforms existing
MKE methods and achieves remarkable edit performance, indicating the importance
of considering the semantic connections among multilingual knowledge. | cs.CL | Multilingual Knowledge Editing with Language-Agnostic Factual Neurons |
2024-06-24T08:05:21Z | http://arxiv.org/pdf/2406.16408v1 | http://arxiv.org/abs/2406.16408v1 | Yan Lanciault, Christophe Reutenauer | 2024-06-24T08:05:21Z | math.CO, cs.CL | Motivated by the theory of trapezoidal words, whose sequences of cardinality
of factors by length are symmetric, we introduce a bivariate variant of this
symmetry. We show that this symmetry characterizes Christoffel words, and
establish other related results. | math.CO | A Symmetry Property of Christoffel Words |
2024-06-24T07:47:34Z | http://arxiv.org/pdf/2406.16382v1 | http://arxiv.org/abs/2406.16382v1 | Zhanyue Qin, Haochuan Wang, Deyuan Liu, Ziyang Song, Cunhang Fan, Zhao Lv, Jinlin Wu, Zhen Lei, Zhiying Tu, Dianhui Chu, Xiaoyan Yu, Dianbo Sui | 2024-06-24T07:47:34Z | cs.CL | Sequential decision-making refers to algorithms that take into account the
dynamics of the environment, where early decisions affect subsequent decisions.
With large language models (LLMs) demonstrating powerful capabilities between
tasks, we can't help but ask: Can Current LLMs Effectively Make Sequential
Decisions? In order to answer this question, we propose the UNO Arena based on
the card game UNO to evaluate the sequential decision-making capability of LLMs
and explain in detail why we choose UNO. In UNO Arena, We evaluate the
sequential decision-making capability of LLMs dynamically with novel metrics
based Monte Carlo methods. We set up random players, DQN-based reinforcement
learning players, and LLM players (e.g. GPT-4, Gemini-pro) for comparison
testing. Furthermore, in order to improve the sequential decision-making
capability of LLMs, we propose the TUTRI player, which can involves having LLMs
reflect their own actions wtih the summary of game history and the game
strategy. Numerous experiments demonstrate that the TUTRI player achieves a
notable breakthrough in the performance of sequential decision-making compared
to the vanilla LLM player. | cs.CL | UNO Arena for Evaluating Sequential Decision-Making Capability of Large Language Models |
2024-06-24T07:42:32Z | http://arxiv.org/pdf/2406.16377v1 | http://arxiv.org/abs/2406.16377v1 | Deng Cai, Huayang Li, Tingchen Fu, Siheng Li, Weiwen Xu, Shuaiyi Li, Bowen Cao, Zhisong Zhang, Xinting Huang, Leyang Cui, Yan Wang, Lemao Liu, Taro Watanabe, Shuming Shi | 2024-06-24T07:42:32Z | cs.CL, cs.AI | Despite the general capabilities of pre-trained large language models (LLMs),
they still need further adaptation to better serve practical applications. In
this paper, we demonstrate the interchangeability of three popular and distinct
adaptation tools: parameter updating, reward modeling, and in-context
prompting. This interchangeability establishes a triangular framework with six
transformation directions, each of which facilitates a variety of applications.
Our work offers a holistic view that unifies numerous existing studies and
suggests potential research directions. We envision our work as a useful
roadmap for future research on LLMs. | cs.CL | On the Transformations across Reward Model, Parameter Update, and In-Context Prompt |
2024-06-24T07:32:35Z | http://arxiv.org/pdf/2406.16374v1 | http://arxiv.org/abs/2406.16374v1 | Dongyang Li, Taolin Zhang, Longtao Huang, Chengyu Wang, Xiaofeng He, Hui Xue | 2024-06-24T07:32:35Z | cs.CL | Knowledge-enhanced pre-trained language models (KEPLMs) leverage relation
triples from knowledge graphs (KGs) and integrate these external data sources
into language models via self-supervised learning. Previous works treat
knowledge enhancement as two independent operations, i.e., knowledge injection
and knowledge integration. In this paper, we propose to learn
Knowledge-Enhanced language representations with Hierarchical Reinforcement
Learning (KEHRL), which jointly addresses the problems of detecting positions
for knowledge injection and integrating external knowledge into the model in
order to avoid injecting inaccurate or irrelevant knowledge. Specifically, a
high-level reinforcement learning (RL) agent utilizes both internal and prior
knowledge to iteratively detect essential positions in texts for knowledge
injection, which filters out less meaningful entities to avoid diverting the
knowledge learning direction. Once the entity positions are selected, a
relevant triple filtration module is triggered to perform low-level RL to
dynamically refine the triples associated with polysemic entities through
binary-valued actions. Experiments validate KEHRL's effectiveness in probing
factual knowledge and enhancing the model's performance on various natural
language understanding tasks. | cs.CL | KEHRL: Learning Knowledge-Enhanced Language Representations with Hierarchical Reinforcement Learning |
2024-06-24T07:27:01Z | http://arxiv.org/pdf/2406.16372v1 | http://arxiv.org/abs/2406.16372v1 | Dongyang Li, Taolin Zhang, Jiali Deng, Longtao Huang, Chengyu Wang, Xiaofeng He, Hui Xue | 2024-06-24T07:27:01Z | cs.CL | Cross-lingual representation learning transfers knowledge from resource-rich
data to resource-scarce ones to improve the semantic understanding abilities of
different languages. However, previous works rely on shallow unsupervised data
generated by token surface matching, regardless of the global context-aware
semantics of the surrounding text tokens. In this paper, we propose an
Unsupervised Pseudo Semantic Data Augmentation (UniPSDA) mechanism for
cross-lingual natural language understanding to enrich the training data
without human interventions. Specifically, to retrieve the tokens with similar
meanings for the semantic data augmentation across different languages, we
propose a sequential clustering process in 3 stages: within a single language,
across multiple languages of a language family, and across languages from
multiple language families. Meanwhile, considering the multi-lingual knowledge
infusion with context-aware semantics while alleviating computation burden, we
directly replace the key constituents of the sentences with the above-learned
multi-lingual family knowledge, viewed as pseudo-semantic. The infusion process
is further optimized via three de-biasing techniques without introducing any
neural parameters. Extensive experiments demonstrate that our model
consistently improves the performance on general zero-shot cross-lingual
natural language understanding tasks, including sequence classification,
information extraction, and question answering. | cs.CL | UniPSDA: Unsupervised Pseudo Semantic Data Augmentation for Zero-Shot Cross-Lingual Natural Language Understanding |
2024-06-24T06:53:36Z | http://arxiv.org/pdf/2406.16356v1 | http://arxiv.org/abs/2406.16356v1 | Rem Hida, Junki Ohmura, Toshiyuki Sekiya | 2024-06-24T06:53:36Z | cs.CL | Instruction-tuned Large Language Models (LLMs) have achieved remarkable
performance across various benchmark tasks. While providing instructions to
LLMs for guiding their generations is user-friendly, assessing their
instruction-following capabilities is still unclarified due to a lack of
evaluation metrics. In this paper, we focus on evaluating the
instruction-following ability of LLMs in the context of story-ending
generation, which requires diverse and context-specific instructions. We
propose an automatic evaluation pipeline that utilizes a machine reading
comprehension (MRC) model to determine whether the generated story-ending
reflects instruction. Our findings demonstrate that our proposed metric aligns
with human evaluation. Furthermore, our experiments confirm that recent
open-source LLMs can achieve instruction-following performance close to
GPT-3.5, as assessed through automatic evaluation. | cs.CL | Evaluation of Instruction-Following Ability for Large Language Models on Story-Ending Generation |
2024-06-24T06:27:47Z | http://arxiv.org/pdf/2406.16342v1 | http://arxiv.org/abs/2406.16342v1 | Yoo Yeon Sung, Eve Fleisig, Ishani Mondal, Jordan Lee Boyd-Graber | 2024-06-24T06:27:47Z | cs.CL | Adversarial benchmarks validate model abilities by providing samples that
fool models but not humans. However, despite the proliferation of datasets that
claim to be adversarial, there does not exist an established metric to evaluate
how adversarial these datasets are. To address this lacuna, we introduce
ADVSCORE, a metric which quantifies how adversarial and discriminative an
adversarial dataset is and exposes the features that make data adversarial. We
then use ADVSCORE to underpin a dataset creation pipeline that incentivizes
writing a high-quality adversarial dataset. As a proof of concept, we use
ADVSCORE to collect an adversarial question answering (QA) dataset, ADVQA, from
our pipeline. The high-quality questions in ADVQA surpasses three adversarial
benchmarks across domains at fooling several models but not humans. We validate
our result based on difficulty estimates from 9,347 human responses on four
datasets and predictions from three models. Moreover, ADVSCORE uncovers which
adversarial tactics used by human writers fool models (e.g., GPT-4) but not
humans. Through ADVSCORE and its analyses, we offer guidance on revealing
language model vulnerabilities and producing reliable adversarial examples. | cs.CL | ADVSCORE: A Metric for the Evaluation and Creation of Adversarial Benchmarks |
2024-06-24T06:26:50Z | http://arxiv.org/pdf/2406.16341v1 | http://arxiv.org/abs/2406.16341v1 | Yeonsu Kwon, Jiho Kim, Gyubok Lee, Seongsu Bae, Daeun Kyung, Wonchul Cha, Tom Pollard, Alistair Johnson, Edward Choi | 2024-06-24T06:26:50Z | cs.CL | Electronic Health Records (EHRs) are integral for storing comprehensive
patient medical records, combining structured data (e.g., medications) with
detailed clinical notes (e.g., physician notes). These elements are essential
for straightforward data retrieval and provide deep, contextual insights into
patient care. However, they often suffer from discrepancies due to unintuitive
EHR system designs and human errors, posing serious risks to patient safety. To
address this, we developed EHRCon, a new dataset and task specifically designed
to ensure data consistency between structured tables and unstructured notes in
EHRs. EHRCon was crafted in collaboration with healthcare professionals using
the MIMIC-III EHR dataset, and includes manual annotations of 3,943 entities
across 105 clinical notes checked against database entries for consistency.
EHRCon has two versions, one using the original MIMIC-III schema, and another
using the OMOP CDM schema, in order to increase its applicability and
generalizability. Furthermore, leveraging the capabilities of large language
models, we introduce CheckEHR, a novel framework for verifying the consistency
between clinical notes and database tables. CheckEHR utilizes an eight-stage
process and shows promising results in both few-shot and zero-shot settings.
The code is available at https://github.com/dustn1259/EHRCon. | cs.CL | EHRCon: Dataset for Checking Consistency between Unstructured Notes and Structured Tables in Electronic Health Records |
2024-06-24T06:10:13Z | http://arxiv.org/pdf/2406.16332v1 | http://arxiv.org/abs/2406.16332v1 | Wenhan Liu, Yutao Zhu, Zhicheng Dou | 2024-06-24T06:10:13Z | cs.IR, cs.CL | Recently, there has been increasing interest in applying large language
models (LLMs) as zero-shot passage rankers. However, few studies have explored
how to select appropriate in-context demonstrations for the passage ranking
task, which is the focus of this paper. Previous studies mainly apply a
demonstration retriever to retrieve demonstrations and use top-$k$
demonstrations for in-context learning (ICL). Although effective, this approach
overlooks the dependencies between demonstrations, leading to inferior
performance of few-shot ICL in the passage ranking task. In this paper, we
formulate the demonstration selection as a \textit{retrieve-then-rerank}
process and introduce the DemoRank framework. In this framework, we first use
LLM feedback to train a demonstration retriever and construct a novel
dependency-aware training samples to train a demonstration reranker to improve
few-shot ICL. The construction of such training samples not only considers
demonstration dependencies but also performs in an efficient way. Extensive
experiments demonstrate DemoRank's effectiveness in in-domain scenarios and
strong generalization to out-of-domain scenarios. Our codes are available
at~\url{https://github.com/8421BCD/DemoRank}. | cs.IR | DemoRank: Selecting Effective Demonstrations for Large Language Models in Ranking Task |
2024-06-24T05:57:55Z | http://arxiv.org/pdf/2406.16330v1 | http://arxiv.org/abs/2406.16330v1 | Deyuan Liu, Zhanyue Qin, Hairu Wang, Zhao Yang, Zecheng Wang, Fangying Rong, Qingbin Liu, Yanchao Hao, Xi Chen, Cunhang Fan, Zhao Lv, Zhiying Tu, Dianhui Chu, Bo Li, Dianbo Sui | 2024-06-24T05:57:55Z | cs.CL, cs.AI | While large language models (LLMs) excel in many domains, their complexity
and scale challenge deployment in resource-limited environments. Current
compression techniques, such as parameter pruning, often fail to effectively
utilize the knowledge from pruned parameters. To address these challenges, we
propose Manifold-Based Knowledge Alignment and Layer Merging Compression (MKA),
a novel approach that uses manifold learning and the Normalized Pairwise
Information Bottleneck (NPIB) measure to merge similar layers, reducing model
size while preserving essential performance. We evaluate MKA on multiple
benchmark datasets and various LLMs. Our findings show that MKA not only
preserves model performance but also achieves substantial compression ratios,
outperforming traditional pruning methods. Moreover, when coupled with
quantization, MKA delivers even greater compression. Specifically, on the MMLU
dataset using the Llama3-8B model, MKA achieves a compression ratio of 43.75%
with a minimal performance decrease of only 2.82\%. The proposed MKA method
offers a resource-efficient and performance-preserving model compression
technique for LLMs. | cs.CL | Pruning via Merging: Compressing LLMs via Manifold Alignment Based Layer Merging |
2024-06-24T05:13:19Z | http://arxiv.org/pdf/2406.16320v1 | http://arxiv.org/abs/2406.16320v1 | Michal Golovanevsky, William Rudman, Vedant Palit, Ritambhara Singh, Carsten Eickhoff | 2024-06-24T05:13:19Z | cs.CL | Vision-Language Models (VLMs) have gained community-spanning prominence due
to their ability to integrate visual and textual inputs to perform complex
tasks. Despite their success, the internal decision-making processes of these
models remain opaque, posing challenges in high-stakes applications. To address
this, we introduce NOTICE, the first Noise-free Text-Image Corruption and
Evaluation pipeline for mechanistic interpretability in VLMs. NOTICE
incorporates a Semantic Minimal Pairs (SMP) framework for image corruption and
Symmetric Token Replacement (STR) for text. This approach enables semantically
meaningful causal mediation analysis for both modalities, providing a robust
method for analyzing multimodal integration within models like BLIP. Our
experiments on the SVO-Probes, MIT-States, and Facial Expression Recognition
datasets reveal crucial insights into VLM decision-making, identifying the
significant role of middle-layer cross-attention heads. Further, we uncover a
set of ``universal cross-attention heads'' that consistently contribute across
tasks and modalities, each performing distinct functions such as implicit image
segmentation, object inhibition, and outlier inhibition. This work paves the
way for more transparent and interpretable multimodal systems. | cs.CL | What Do VLMs NOTICE? A Mechanistic Interpretability Pipeline for Noise-free Text-Image Corruption and Evaluation |
2024-06-24T04:56:26Z | http://arxiv.org/pdf/2406.16319v1 | http://arxiv.org/abs/2406.16319v1 | Irene Smith, Morgan Sonderegger, The Spade Consortium | 2024-06-24T04:56:26Z | cs.CL | This paper introduces a novel method for quantifying vowel overlap. There is
a tension in previous work between using multivariate measures, such as those
derived from empirical distributions, and the ability to control for unbalanced
data and extraneous factors, as is possible when using fitted model parameters.
The method presented here resolves this tension by jointly modelling all
acoustic dimensions of interest and by simulating distributions from the model
to compute a measure of vowel overlap. An additional benefit of this method is
that computation of uncertainty becomes straightforward. We evaluate this
method on corpus speech data targeting the PIN-PEN merger in four dialects of
English and find that using modelled distributions to calculate Bhattacharyya
affinity substantially improves results compared to empirical distributions,
while the difference between multivariate and univariate modelling is subtle. | cs.CL | Modelled Multivariate Overlap: A method for measuring vowel merger |
2024-06-24T04:50:12Z | http://arxiv.org/pdf/2406.16316v1 | http://arxiv.org/abs/2406.16316v1 | Yuu Jinnai | 2024-06-24T04:50:12Z | cs.CL, cs.AI, cs.CY, cs.LG | Alignment of the language model with human preferences is a common approach
to making a language model useful to end users. However, most alignment work is
done in English, and human preference datasets are dominated by English,
reflecting only the preferences of English-speaking annotators. Nevertheless,
it is common practice to use the English preference data, either directly or by
translating it into the target language, when aligning a multilingual language
model. The question is whether such an alignment strategy marginalizes the
preference of non-English speaking users. To this end, we investigate the
effect of aligning Japanese language models with (mostly) English resources. In
particular, we focus on evaluating whether the commonsense morality of the
resulting fine-tuned models is aligned with Japanese culture using the
JCommonsenseMorality (JCM) and ETHICS datasets. The experimental results show
that the fine-tuned model outperforms the SFT model. However, it does not
demonstrate the same level of improvement as a model fine-tuned using the JCM,
suggesting that while some aspects of commonsense morality are transferable,
others may not be. | cs.CL | Does Cross-Cultural Alignment Change the Commonsense Morality of Language Models? |
2024-06-24T04:48:29Z | http://arxiv.org/pdf/2406.16315v1 | http://arxiv.org/abs/2406.16315v1 | Hokuto Munakata, Ryo Terashima, Yusuke Fujita | 2024-06-24T04:48:29Z | eess.AS, cs.CL, cs.SD | We propose a data cleansing method that utilizes a neural analysis and
synthesis (NANSY++) framework to train an end-to-end neural diarization model
(EEND) for singer diarization. Our proposed model converts song data with
choral singing which is commonly contained in popular music and unsuitable for
generating a simulated dataset to the solo singing data. This cleansing is
based on NANSY++, which is a framework trained to reconstruct an input
non-overlapped audio signal. We exploit the pre-trained NANSY++ to convert
choral singing into clean, non-overlapped audio. This cleansing process
mitigates the mislabeling of choral singing to solo singing and helps the
effective training of EEND models even when the majority of available song data
contains choral singing sections. We experimentally evaluated the EEND model
trained with a dataset using our proposed method using annotated popular duet
songs. As a result, our proposed method improved 14.8 points in diarization
error rate. | eess.AS | Song Data Cleansing for End-to-End Neural Singer Diarization Using Neural Analysis and Synthesis Framework |
2024-06-24T04:17:03Z | http://arxiv.org/pdf/2406.16308v1 | http://arxiv.org/abs/2406.16308v1 | Aodong Li, Yunhan Zhao, Chen Qiu, Marius Kloft, Padhraic Smyth, Maja Rudolph, Stephan Mandt | 2024-06-24T04:17:03Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) have shown their potential in long-context
understanding and mathematical reasoning. In this paper, we study the problem
of using LLMs to detect tabular anomalies and show that pre-trained LLMs are
zero-shot batch-level anomaly detectors. That is, without extra
distribution-specific model fitting, they can discover hidden outliers in a
batch of data, demonstrating their ability to identify low-density data
regions. For LLMs that are not well aligned with anomaly detection and
frequently output factual errors, we apply simple yet effective data-generating
processes to simulate synthetic batch-level anomaly detection datasets and
propose an end-to-end fine-tuning strategy to bring out the potential of LLMs
in detecting real anomalies. Experiments on a large anomaly detection benchmark
(ODDS) showcase i) GPT-4 has on-par performance with the state-of-the-art
transductive learning-based anomaly detection methods and ii) the efficacy of
our synthetic dataset and fine-tuning strategy in aligning LLMs to this task. | cs.LG | Anomaly Detection of Tabular Data Using LLMs |
2024-06-24T04:08:35Z | http://arxiv.org/pdf/2406.16306v1 | http://arxiv.org/abs/2406.16306v1 | Bolian Li, Yifan Wang, Ananth Grama, Ruqi Zhang | 2024-06-24T04:08:35Z | cs.CL, cs.LG, stat.ML | Aligning large language models (LLMs) with human preferences is critical for
their deployment. Recently, decoding-time alignment has emerged as an effective
plug-and-play technique that requires no fine-tuning of model parameters.
However, generating text that achieves both high reward and high likelihood
remains a significant challenge. Existing methods often fail to generate
high-reward text or incur substantial computational costs. In this paper, we
propose Cascade Reward Sampling (CARDS) to address both issues, guaranteeing
the generation of high-reward and high-likelihood text with significantly low
costs. Based on our analysis of reward models (RMs) on incomplete text and our
observation that high-reward prefixes induce high-reward complete text, we use
rejection sampling to iteratively generate small semantic segments to form such
prefixes. The segment length is dynamically determined by the predictive
uncertainty of LLMs. This strategy guarantees desirable prefixes for subsequent
generations and significantly reduces wasteful token re-generations and the
number of reward model scoring. Our experiments demonstrate substantial gains
in both generation efficiency and alignment ratings compared to the baselines,
achieving five times faster text generation and 99\% win-ties in GPT-4/Claude-3
helpfulness evaluation. | cs.CL | Cascade Reward Sampling for Efficient Decoding-Time Alignment |
2024-06-24T03:59:17Z | http://arxiv.org/pdf/2406.17808v1 | http://arxiv.org/abs/2406.17808v1 | Jeffrey Willette, Heejun Lee, Youngwan Lee, Myeongjae Jeon, Sung Ju Hwang | 2024-06-24T03:59:17Z | cs.CL, cs.AI, cs.LG | The context window within a transformer provides a form of active memory for
the current task, which can be useful for few-shot learning and conditional
generation, both which depend heavily on previous context tokens. However, as
the context length grows, the computational cost increases quadratically.
Recent works have shown that saving a few initial tokens along with a
fixed-sized sliding window leads to stable streaming generation with linear
complexity in transformer-based Large Language Models (LLMs). However, they
make suboptimal use of the fixed window by naively evicting all tokens
unconditionally from the key-value (KV) cache once they reach the end of the
window, resulting in tokens being forgotten and no longer able to affect
subsequent predictions. To overcome this limitation, we propose a novel
mechanism for storing longer sliding window contexts with the same total cache
size by keeping separate cascading sub-cache buffers whereby each subsequent
buffer conditionally accepts a fraction of the relatively more important tokens
evicted from the previous buffer. Our method results in a dynamic KV cache that
can store tokens from the more distant past than a fixed, static sliding window
approach. Our experiments show improvements of 5.6% on long context generation
(LongBench), 1.2% in streaming perplexity (PG19), and 0.6% in language
understanding (MMLU STEM) using LLMs given the same fixed cache size.
Additionally, we provide an efficient implementation that improves the KV cache
latency from 1.33ms per caching operation to 0.54ms, a 59% speedup over
previous work. | cs.CL | Training-Free Exponential Extension of Sliding Window Context with Cascading KV Cache |
2024-06-24T03:52:52Z | http://arxiv.org/pdf/2406.16299v1 | http://arxiv.org/abs/2406.16299v1 | Yifei Gao, Jie Ou, Lei Wang, Yuting Xiao, Zhiyuan Xiang, Ruiting Dai, Jun Cheng | 2024-06-24T03:52:52Z | cs.CL, cs.AI, F.2.3 | Emergent Large Language Models (LLMs) use their extraordinary performance and
powerful deduction capacity to discern from traditional language models.
However, the expenses of computational resources and storage for these LLMs are
stunning, quantization then arises as a trending conversation. To address
accuracy decay caused by quantization, two streams of works in post-training
quantization methods stand out. One uses other weights to compensate existing
quantization error, while the other transfers the quantization difficulty to
other parts in the model. Combining both merits, we introduce Learnable
Singular value Increment (LSI) as an advanced solution. LSI uses Singular Value
Decomposition to extract singular values of the weights and make them learnable
to help weights compensate each other conditioned on activation. Incorporating
LSI with existing techniques, we achieve state-of-the-art performance in
diverse quantization settings, no matter in weight-only, weight-activation or
extremely low bit scenarios. By unleashing the potential of LSI, efficient
finetuning on quantized model is no longer a prohibitive problem. | cs.CL | Compensate Quantization Errors: Make Weights Hierarchical to Compensate Each Other |
2024-06-24T03:36:29Z | http://arxiv.org/pdf/2406.16294v1 | http://arxiv.org/abs/2406.16294v1 | Zixia Jia, Mengmeng Wang, Baichen Tong, Song-Chun Zhu, Zilong Zheng | 2024-06-24T03:36:29Z | cs.CL, cs.AI | Recent advances in Large Language Models (LLMs) have shown inspiring
achievements in constructing autonomous agents that rely on language
descriptions as inputs. However, it remains unclear how well LLMs can function
as few-shot or zero-shot embodied agents in dynamic interactive environments.
To address this gap, we introduce LangSuitE, a versatile and simulation-free
testbed featuring 6 representative embodied tasks in textual embodied worlds.
Compared with previous LLM-based testbeds, LangSuitE (i) offers adaptability to
diverse environments without multiple simulation engines, (ii) evaluates
agents' capacity to develop ``internalized world knowledge'' with embodied
observations, and (iii) allows easy customization of communication and action
strategies. To address the embodiment challenge, we devise a novel
chain-of-thought (CoT) schema, EmMem, which summarizes embodied states w.r.t.
history information. Comprehensive benchmark results illustrate challenges and
insights of embodied planning. LangSuitE represents a significant step toward
building embodied generalists in the context of language models. | cs.CL | LangSuitE: Planning, Controlling and Interacting with Large Language Models in Embodied Text Environments |
2024-06-24T03:36:19Z | http://arxiv.org/pdf/2406.16293v1 | http://arxiv.org/abs/2406.16293v1 | Zixia Jia, Junpeng Li, Shichuan Zhang, Anji Liu, Zilong Zheng | 2024-06-24T03:36:19Z | cs.CL, cs.AI | Traditional supervised learning heavily relies on human-annotated datasets,
especially in data-hungry neural approaches. However, various tasks, especially
multi-label tasks like document-level relation extraction, pose challenges in
fully manual annotation due to the specific domain knowledge and large class
sets. Therefore, we address the multi-label positive-unlabelled learning
(MLPUL) problem, where only a subset of positive classes is annotated. We
propose Mixture Learner for Partially Annotated Classification (MLPAC), an
RL-based framework combining the exploration ability of reinforcement learning
and the exploitation ability of supervised learning. Experimental results
across various tasks, including document-level relation extraction, multi-label
image classification, and binary PU learning, demonstrate the generalization
and effectiveness of our framework. | cs.CL | Combining Supervised Learning and Reinforcement Learning for Multi-Label Classification Tasks with Partial Labels |
2024-06-24T03:29:53Z | http://arxiv.org/pdf/2406.16288v1 | http://arxiv.org/abs/2406.16288v1 | Jooyoung Lee, Toshini Agrawal, Adaku Uchendu, Thai Le, Jinghui Chen, Dongwon Lee | 2024-06-24T03:29:53Z | cs.CL | Recent literature has highlighted potential risks to academic integrity
associated with large language models (LLMs), as they can memorize parts of
training instances and reproduce them in the generated texts without proper
attribution. In addition, given their capabilities in generating high-quality
texts, plagiarists can exploit LLMs to generate realistic paraphrases or
summaries indistinguishable from original work. In response to possible
malicious use of LLMs in plagiarism, we introduce PlagBench, a comprehensive
dataset consisting of 46.5K synthetic plagiarism cases generated using three
instruction-tuned LLMs across three writing domains. The quality of PlagBench
is ensured through fine-grained automatic evaluation for each type of
plagiarism, complemented by human annotation. We then leverage our proposed
dataset to evaluate the plagiarism detection performance of five modern LLMs
and three specialized plagiarism checkers. Our findings reveal that GPT-3.5
tends to generates paraphrases and summaries of higher quality compared to
Llama2 and GPT-4. Despite LLMs' weak performance in summary plagiarism
identification, they can surpass current commercial plagiarism detectors.
Overall, our results highlight the potential of LLMs to serve as robust
plagiarism detection tools. | cs.CL | PlagBench: Exploring the Duality of Large Language Models in Plagiarism Generation and Detection |
2024-06-24T02:50:09Z | http://arxiv.org/pdf/2406.16275v1 | http://arxiv.org/abs/2406.16275v1 | Choonghyun Park, Hyuhng Joon Kim, Junyeob Kim, Youna Kim, Taeuk Kim, Hyunsoo Cho, Hwiyeol Jo, Sang-goo Lee, Kang Min Yoo | 2024-06-24T02:50:09Z | cs.CL | AI Generated Text (AIGT) detectors are developed with texts from humans and
LLMs of common tasks. Despite the diversity of plausible prompt choices, these
datasets are generally constructed with a limited number of prompts. The lack
of prompt variation can introduce prompt-specific shortcut features that exist
in data collected with the chosen prompt, but do not generalize to others. In
this paper, we analyze the impact of such shortcuts in AIGT detection. We
propose Feedback-based Adversarial Instruction List Optimization (FAILOpt), an
attack that searches for instructions deceptive to AIGT detectors exploiting
prompt-specific shortcuts. FAILOpt effectively drops the detection performance
of the target detector, comparable to other attacks based on adversarial
in-context examples. We also utilize our method to enhance the robustness of
the detector by mitigating the shortcuts. Based on the findings, we further
train the classifier with the dataset augmented by FAILOpt prompt. The
augmented classifier exhibits improvements across generation models, tasks, and
attacks. Our code will be available at https://github.com/zxcvvxcz/FAILOpt. | cs.CL | Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection |
2024-06-24T02:03:57Z | http://arxiv.org/pdf/2406.16264v1 | http://arxiv.org/abs/2406.16264v1 | Marzena Karpinska, Katherine Thai, Kyle Lo, Tanya Goyal, Mohit Iyyer | 2024-06-24T02:03:57Z | cs.CL, cs.AI | Synthetic long-context LLM benchmarks (e.g., "needle-in-the-haystack") test
only surface-level retrieval capabilities, but how well can long-context LLMs
retrieve, synthesize, and reason over information across book-length inputs? We
address this question by creating NoCha, a dataset of 1,001 minimally different
pairs of true and false claims about 67 recently-published English fictional
books, written by human readers of those books. In contrast to existing
long-context benchmarks, our annotators confirm that the largest share of pairs
in NoCha require global reasoning over the entire book to verify. Our
experiments show that while human readers easily perform this task, it is
enormously challenging for all ten long-context LLMs that we evaluate: no
open-weight model performs above random chance (despite their strong
performance on synthetic benchmarks), while GPT-4o achieves the highest
accuracy at 55.8%. Further analysis reveals that (1) on average, models perform
much better on pairs that require only sentence-level retrieval vs. global
reasoning; (2) model-generated explanations for their decisions are often
inaccurate even for correctly-labeled claims; and (3) models perform
substantially worse on speculative fiction books that contain extensive
world-building. The methodology proposed in NoCha allows for the evolution of
the benchmark dataset and the easy analysis of future models. | cs.CL | One Thousand and One Pairs: A "novel" challenge for long-context language models |
2024-06-24T01:31:03Z | http://arxiv.org/pdf/2406.16254v1 | http://arxiv.org/abs/2406.16254v1 | Alessandro Stolfo, Ben Wu, Wes Gurnee, Yonatan Belinkov, Xingyi Song, Mrinmaya Sachan, Neel Nanda | 2024-06-24T01:31:03Z | cs.LG, cs.AI, cs.CL | Despite their widespread use, the mechanisms by which large language models
(LLMs) represent and regulate uncertainty in next-token predictions remain
largely unexplored. This study investigates two critical components believed to
influence this uncertainty: the recently discovered entropy neurons and a new
set of components that we term token frequency neurons. Entropy neurons are
characterized by an unusually high weight norm and influence the final layer
normalization (LayerNorm) scale to effectively scale down the logits. Our work
shows that entropy neurons operate by writing onto an unembedding null space,
allowing them to impact the residual stream norm with minimal direct effect on
the logits themselves. We observe the presence of entropy neurons across a
range of models, up to 7 billion parameters. On the other hand, token frequency
neurons, which we discover and describe here for the first time, boost or
suppress each token's logit proportionally to its log frequency, thereby
shifting the output distribution towards or away from the unigram distribution.
Finally, we present a detailed case study where entropy neurons actively manage
confidence in the setting of induction, i.e. detecting and continuing repeated
subsequences. | cs.LG | Confidence Regulation Neurons in Language Models |
2024-06-24T01:30:22Z | http://arxiv.org/pdf/2406.16253v2 | http://arxiv.org/abs/2406.16253v2 | Jiangshu Du, Yibo Wang, Wenting Zhao, Zhongfen Deng, Shuaiqi Liu, Renze Lou, Henry Peng Zou, Pranav Narayanan Venkit, Nan Zhang, Mukund Srinath, Haoran Ranran Zhang, Vipul Gupta, Yinghui Li, Tao Li, Fei Wang, Qin Liu, Tianlin Liu, Pengzhi Gao, Congying Xia, Chen Xing, Jiayang Cheng, Zhaowei Wang, Ying Su, Raj Sanjay Shah, Ruohao Guo, Jing Gu, Haoran Li, Kangda Wei, Zihao Wang, Lu Cheng, Surangika Ranathunga, Meng Fang, Jie Fu, Fei Liu, Ruihong Huang, Eduardo Blanco, Yixin Cao, Rui Zhang, Philip S. Yu, Wenpeng Yin | 2024-06-25T18:04:38Z | cs.CL | This work is motivated by two key trends. On one hand, large language models
(LLMs) have shown remarkable versatility in various generative tasks such as
writing, drawing, and question answering, significantly reducing the time
required for many routine tasks. On the other hand, researchers, whose work is
not only time-consuming but also highly expertise-demanding, face increasing
challenges as they have to spend more time reading, writing, and reviewing
papers. This raises the question: how can LLMs potentially assist researchers
in alleviating their heavy workload?
This study focuses on the topic of LLMs assist NLP Researchers, particularly
examining the effectiveness of LLM in assisting paper (meta-)reviewing and its
recognizability. To address this, we constructed the ReviewCritique dataset,
which includes two types of information: (i) NLP papers (initial submissions
rather than camera-ready) with both human-written and LLM-generated reviews,
and (ii) each review comes with "deficiency" labels and corresponding
explanations for individual segments, annotated by experts. Using
ReviewCritique, this study explores two threads of research questions: (i)
"LLMs as Reviewers", how do reviews generated by LLMs compare with those
written by humans in terms of quality and distinguishability? (ii) "LLMs as
Metareviewers", how effectively can LLMs identify potential issues, such as
Deficient or unprofessional review segments, within individual paper reviews?
To our knowledge, this is the first work to provide such a comprehensive
analysis. | cs.CL | LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing |
2024-06-23T22:56:34Z | http://arxiv.org/pdf/2406.16985v1 | http://arxiv.org/abs/2406.16985v1 | Yukun Zhang | 2024-06-23T22:56:34Z | cs.LG, cs.AI, cs.CL | This study presents a novel approach that leverages Neural Ordinary
Differential Equations (Neural ODEs) to unravel the intricate relationships
between inputs and outputs in Large Language Models (LLMs), and employs robust
control to fine-tune outputs to meet predefined standards. Central to our
methodology is the transformation of LLM inputs and outputs into a
lower-dimensional latent space, facilitating a detailed examination of the
information processing pathways within LLMs. Neural ODEs play a pivotal role in
this investigation by providing a dynamic model that captures the continuous
evolution of data within the LLMs. Additionally, robust control mechanisms are
applied to strategically adjust the model's outputs, ensuring they not only
maintain high quality and reliability but also adhere to specific performance
criteria. This fusion of Neural ODEs and robust control represents a
significant advancement in LLM interpretability, offering a comprehensive
framework that elucidates the previously opaque mechanisms of these complex
models. Our empirical results validate the effectiveness of this integrated
approach, making a substantial contribution to the field of explainable AI by
merging advanced machine learning techniques with the critical need for
transparency and control in AI outputs. | cs.LG | Unveiling LLM Mechanisms Through Neural ODEs and Control Theory |
2024-06-23T22:53:47Z | http://arxiv.org/pdf/2406.16235v1 | http://arxiv.org/abs/2406.16235v1 | Xiaochen Li, Zheng-Xin Yong, Stephen H. Bach | 2024-06-23T22:53:47Z | cs.CL, cs.AI, cs.CR, cs.LG | Detoxifying multilingual Large Language Models (LLMs) has become crucial due
to their increasing global use. In this work, we explore zero-shot
cross-lingual generalization of preference tuning in detoxifying LLMs. Unlike
previous studies that show limited cross-lingual generalization for other
safety tasks, we demonstrate that Direct Preference Optimization (DPO) training
with only English data can significantly reduce toxicity in multilingual
open-ended generations. For example, the probability of mGPT-1.3B generating
toxic continuations drops from 46.8% to 3.9% across 17 different languages
after training. Our results also extend to other multilingual LLMs, such as
BLOOM, Llama3, and Aya-23. Using mechanistic interpretability tools like causal
intervention and activation analysis, we identified the dual multilinguality
property of MLP layers in LLMs, which explains the cross-lingual generalization
of DPO. Finally, we show that bilingual sentence retrieval can predict the
cross-lingual transferability of DPO preference tuning. | cs.CL | Preference Tuning For Toxicity Mitigation Generalizes Across Languages |
2024-06-23T21:56:48Z | http://arxiv.org/pdf/2406.16229v1 | http://arxiv.org/abs/2406.16229v1 | Dang Nguyen, Jiuhai Chen, Tianyi Zhou | 2024-06-23T21:56:48Z | cs.CL | Large language models (LLMs), despite their breakthroughs on many challenging
benchmark tasks, lean to generate verbose responses and lack the
controllability of output complexity, which is usually preferred by human users
in practice. In this paper, we study how to precisely control multiple
linguistic complexities of LLM output by finetuning using off-the-shelf data.
To this end, we propose multi-control tuning (MCTune), which includes multiple
linguistic complexity values of ground-truth responses as controls in the input
for instruction tuning. We finetune LLaMA2-7B on Alpaca-GPT4 and WizardLM
datasets. Evaluations on widely used benchmarks demonstrate that our method
does not only improve LLMs' multi-complexity controllability substantially but
also retains or even enhances the quality of the responses as a side benefit. | cs.CL | Multi-Objective Linguistic Control of Large Language Models |
2024-06-23T21:32:15Z | http://arxiv.org/pdf/2406.16223v1 | http://arxiv.org/abs/2406.16223v1 | Rong Wang, Kun Sun | 2024-06-23T21:32:15Z | cs.CL, cs.AI | The traditional personality models only yield binary results. This paper
presents a novel approach for training personality detection models that
produce continuous output values, using mixed strategies. By leveraging the
PANDORA dataset, which includes extensive personality labeling of Reddit
comments, we developed models that predict the Big Five personality traits with
high accuracy. Our approach involves fine-tuning a RoBERTa-base model with
various strategies such as Multi-Layer Perceptron (MLP) integration, and
hyperparameter tuning. The results demonstrate that our models significantly
outperform traditional binary classification methods, offering precise
continuous outputs for personality traits, thus enhancing applications in AI,
psychology, human resources, marketing and health care fields. | cs.CL | Continuous Output Personality Detection Models via Mixed Strategy Training |
2024-06-23T19:49:10Z | http://arxiv.org/pdf/2406.16203v3 | http://arxiv.org/abs/2406.16203v3 | Hanzi Xu, Renze Lou, Jiangshu Du, Vahid Mahzoon, Elmira Talebianaraki, Zhuoan Zhou, Elizabeth Garrison, Slobodan Vucetic, Wenpeng Yin | 2024-07-03T13:18:50Z | cs.CL | In many classification tasks designed for AI or human to solve, gold labels
are typically included within the label space by default, often posed as "which
of the following is correct?" This standard setup has traditionally highlighted
the strong performance of advanced AI, particularly top-performing Large
Language Models (LLMs), in routine classification tasks. However, when the gold
label is intentionally excluded from the label space, it becomes evident that
LLMs still attempt to select from the available label candidates, even when
none are correct. This raises a pivotal question: Do LLMs truly demonstrate
their intelligence in understanding the essence of classification tasks?
In this study, we evaluate both closed-source and open-source LLMs across
representative classification tasks, arguing that the perceived performance of
LLMs is overstated due to their inability to exhibit the expected comprehension
of the task. This paper makes a threefold contribution: i) To our knowledge,
this is the first work to identify the limitations of LLMs in classification
tasks when gold labels are absent. We define this task as Classify-w/o-Gold and
propose it as a new testbed for LLMs. ii) We introduce a benchmark, Know-No,
comprising two existing classification tasks and one new task, to evaluate
Classify-w/o-Gold. iii) This work defines and advocates for a new evaluation
metric, OmniAccuracy, which assesses LLMs' performance in classification tasks
both when gold labels are present and absent. | cs.CL | LLMs' Classification Performance is Overclaimed |
2024-06-23T19:40:11Z | http://arxiv.org/pdf/2406.16201v1 | http://arxiv.org/abs/2406.16201v1 | Debeshee Das, Jie Zhang, Florian Tramèr | 2024-06-23T19:40:11Z | cs.CR, cs.CL, cs.LG | Membership inference (MI) attacks try to determine if a data sample was used
to train a machine learning model. For foundation models trained on unknown Web
data, MI attacks can be used to detect copyrighted training materials, measure
test set contamination, or audit machine unlearning. Unfortunately, we find
that evaluations of MI attacks for foundation models are flawed, because they
sample members and non-members from different distributions. For 8 published MI
evaluation datasets, we show that blind attacks -- that distinguish the member
and non-member distributions without looking at any trained model -- outperform
state-of-the-art MI attacks. Existing evaluations thus tell us nothing about
membership leakage of a foundation model's training data. | cs.CR | Blind Baselines Beat Membership Inference Attacks for Foundation Models |
2024-06-23T18:01:56Z | http://arxiv.org/pdf/2406.16176v1 | http://arxiv.org/abs/2406.16176v1 | Qiming Wu, Zichen Chen, Will Corcoran, Misha Sra, Ambuj K. Singh | 2024-06-23T18:01:56Z | cs.AI, cs.CL, cs.LG, H.2.8, I.2.6, I.2.7 | Large language models (LLMs) have achieved remarkable success in natural
language processing (NLP), demonstrating significant capabilities in processing
and understanding text data. However, recent studies have identified
limitations in LLMs' ability to reason about graph-structured data. To address
this gap, we introduce GraphEval2000, the first comprehensive graph dataset,
comprising 40 graph data structure problems along with 2000 test cases.
Additionally, we introduce an evaluation framework based on GraphEval2000,
designed to assess the graph reasoning abilities of LLMs through coding
challenges. Our dataset categorizes test cases into four primary and four
sub-categories, ensuring a comprehensive evaluation. We evaluate eight popular
LLMs on GraphEval2000, revealing that LLMs exhibit a better understanding of
directed graphs compared to undirected ones. While private LLMs consistently
outperform open-source models, the performance gap is narrowing. Furthermore,
to improve the usability of our evaluation framework, we propose Structured
Symbolic Decomposition (SSD), an instruction-based method designed to enhance
LLM performance on GraphEval2000. Results show that SSD improves the
performance of GPT-3.5, GPT-4, and GPT-4o on complex graph problems, with an
increase of 11.11\%, 33.37\%, and 33.37\%, respectively. | cs.AI | GraphEval2000: Benchmarking and Improving Large Language Models on Graph Datasets |
2024-06-23T17:18:19Z | http://arxiv.org/pdf/2406.16167v1 | http://arxiv.org/abs/2406.16167v1 | Harish Tayyar Madabushi | 2024-06-23T17:18:19Z | cs.CL | We present a novel extension to Retrieval Augmented Generation with the goal
of mitigating factual inaccuracies in the output of large language models.
Specifically, our method draws on the cognitive linguistic theory of frame
semantics for the indexing and retrieval of factual information relevant to
helping large language models answer queries. We conduct experiments to
demonstrate the effectiveness of this method both in terms of retrieval
effectiveness and in terms of the relevance of the frames and frame relations
automatically generated. Our results show that this novel mechanism of Frame
Semantic-based retrieval, designed to improve Retrieval Augmented Generation
(FS-RAG), is effective and offers potential for providing data-driven insights
into frame semantics theory. We provide open access to our program code and
prompts. | cs.CL | FS-RAG: A Frame Semantics Based Approach for Improved Factual Accuracy in Large Language Models |
2024-06-23T16:26:27Z | http://arxiv.org/pdf/2406.16152v1 | http://arxiv.org/abs/2406.16152v1 | Angana Borah, Aparna Garimella, Rada Mihalcea | 2024-06-23T16:26:27Z | cs.CL | When exposed to human-generated data, language models are known to learn and
amplify societal biases. While previous works introduced benchmarks that can be
used to assess the bias in these models, they rely on assumptions that may not
be universally true. For instance, a gender bias dimension commonly used by
these metrics is that of family--career, but this may not be the only common
bias in certain regions of the world. In this paper, we identify topical
differences in gender bias across different regions and propose a region-aware
bottom-up approach for bias assessment. Our proposed approach uses
gender-aligned topics for a given region and identifies gender bias dimensions
in the form of topic pairs that are likely to capture gender societal biases.
Several of our proposed bias topic pairs are on par with human perception of
gender biases in these regions in comparison to the existing ones, and we also
identify new pairs that are more aligned than the existing ones. In addition,
we use our region-aware bias topic pairs in a Word Embedding Association Test
(WEAT)-based evaluation metric to test for gender biases across different
regions in different data domains. We also find that LLMs have a higher
alignment to bias pairs for highly-represented regions showing the importance
of region-aware bias evaluation metric. | cs.CL | Towards Region-aware Bias Evaluation Metrics |
2024-06-23T15:55:07Z | http://arxiv.org/pdf/2407.02518v1 | http://arxiv.org/abs/2407.02518v1 | Hung Le, Yingbo Zhou, Caiming Xiong, Silvio Savarese, Doyen Sahoo | 2024-06-23T15:55:07Z | cs.SE, cs.AI, cs.CL, cs.CR, cs.MA, cs.PL | Large language models (LLMs) for code are typically trained to align with
natural language instructions to closely follow their intentions and
requirements. However, in many practical scenarios, it becomes increasingly
challenging for these models to navigate the intricate boundary between
helpfulness and safety, especially against highly complex yet potentially
malicious instructions. In this work, we introduce INDICT: a new framework that
empowers LLMs with Internal Dialogues of Critiques for both safety and
helpfulness guidance. The internal dialogue is a dual cooperative system
between a safety-driven critic and a helpfulness-driven critic. Each critic
provides analysis against the given task and corresponding generated response,
equipped with external knowledge queried through relevant code snippets and
tools like web search and code interpreter. We engage the dual critic system in
both code generation stage as well as code execution stage, providing
preemptive and post-hoc guidance respectively to LLMs. We evaluated INDICT on 8
diverse tasks across 8 programming languages from 5 benchmarks, using LLMs from
7B to 70B parameters. We observed that our approach can provide an advanced
level of critiques of both safety and helpfulness analysis, significantly
improving the quality of output codes ($+10\%$ absolute improvements in all
models). | cs.SE | INDICT: Code Generation with Internal Dialogues of Critiques for Both Security and Helpfulness |
2024-06-23T15:50:22Z | http://arxiv.org/pdf/2406.16144v1 | http://arxiv.org/abs/2406.16144v1 | Zezhong Wang, Xingshan Zeng, Weiwen Liu, Yufei Wang, Liangyou Li, Yasheng Wang, Lifeng Shang, Xin Jiang, Qun Liu, Kam-Fai Wong | 2024-06-23T15:50:22Z | cs.CL | Current research found the issue of Early Answering in large language models
(LLMs), where the models already have an answer before generating the
Chain-of-Thought (CoT). This phenomenon suggests a potential lack of necessary
dependency between the predicted answer and the reasoning process.
Consequently, two important questions arise: (1) Is CoT still necessary if the
model already has an answer? (2) Can the correctness of the answer serve as
valid evidence for the correctness of CoT? To address these questions, we
propose a method, namely Chain-of-Probe (CoP), to probe changes in the mind
during the model's reasoning. The probing results show that in a significant
number of question-answer cases, CoT appears to be unnecessary, and this
necessity correlates with the simplicity of the task, defined by reasoning
steps required. Furthermore, by analyzing patterns in mind change, we examine
the correctness of the model's reasoning. Our validation reveals that many
responses, although correct in their final answer, contain errors in their
reasoning process. To this end, we propose a strategic approach based on CoP to
prioritize answers with correct reasoning among multiple candidates, thereby
bolstering the reliability of the model's reasoning. | cs.CL | Chain-of-Probe: Examing the Necessity and Accuracy of CoT Step-by-Step |
2024-06-23T15:15:17Z | http://arxiv.org/pdf/2406.16135v1 | http://arxiv.org/abs/2406.16135v1 | Lynn Chua, Badih Ghazi, Yangsibo Huang, Pritish Kamath, Ravi Kumar, Pasin Manurangsi, Amer Sinha, Chulin Xie, Chiyuan Zhang | 2024-06-23T15:15:17Z | cs.CL, cs.LG | Large language models (LLMs) are typically multilingual due to pretraining on
diverse multilingual corpora. But can these models relate corresponding
concepts across languages, effectively being crosslingual? This study evaluates
six state-of-the-art LLMs on inherently crosslingual tasks. We observe that
while these models show promising surface-level crosslingual abilities on
machine translation and embedding space analyses, they struggle with deeper
crosslingual knowledge transfer, revealing a crosslingual knowledge barrier in
both general (MMLU benchmark) and domain-specific (Harry Potter quiz) contexts.
We observe that simple inference-time mitigation methods offer only limited
improvement. On the other hand, we propose fine-tuning of LLMs on
mixed-language data, which effectively reduces these gaps, even when using
out-of-domain datasets like WikiText. Our findings suggest the need for
explicit optimization to unlock the full crosslingual potential of LLMs. Our
code is publicly available at
https://github.com/google-research/crosslingual-knowledge-barriers. | cs.CL | Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models |
2024-06-23T14:22:59Z | http://arxiv.org/pdf/2406.16120v1 | http://arxiv.org/abs/2406.16120v1 | Muhammad Shakeel, Yui Sudo, Yifan Peng, Shinji Watanabe | 2024-06-23T14:22:59Z | eess.AS, cs.CL, cs.SD | Contextualized end-to-end automatic speech recognition has been an active
research area, with recent efforts focusing on the implicit learning of
contextual phrases based on the final loss objective. However, these approaches
ignore the useful contextual knowledge encoded in the intermediate layers. We
hypothesize that employing explicit biasing loss as an auxiliary task in the
encoder intermediate layers may better align text tokens or audio frames with
the desired objectives. Our proposed intermediate biasing loss brings more
regularization and contextualization to the network. Our method outperforms a
conventional contextual biasing baseline on the LibriSpeech corpus, achieving a
relative improvement of 22.5% in biased word error rate (B-WER) and up to 44%
compared to the non-contextual baseline with a biasing list size of 100.
Moreover, employing RNN-transducer-driven joint decoding further reduces the
unbiased word error rate (U-WER), resulting in a more robust network. | eess.AS | Contextualized End-to-end Automatic Speech Recognition with Intermediate Biasing Loss |
2024-06-23T13:50:08Z | http://arxiv.org/pdf/2406.16107v1 | http://arxiv.org/abs/2406.16107v1 | Emiru Tsunoo, Hayato Futami, Yosuke Kashiwagi, Siddhant Arora, Shinji Watanabe | 2024-06-23T13:50:08Z | eess.AS, cs.CL | Decoder-only language models (LMs) have been successfully adopted for
speech-processing tasks including automatic speech recognition (ASR). The LMs
have ample expressiveness and perform efficiently. This efficiency is a
suitable characteristic for streaming applications of ASR. In this work, we
propose to use a decoder-only architecture for blockwise streaming ASR. In our
approach, speech features are compressed using CTC output and context embedding
using blockwise speech subnetwork, and are sequentially provided as prompts to
the decoder. The decoder estimates the output tokens promptly at each block. To
this end, we also propose a novel training scheme using random-length prefix
prompts to make the model robust to the truncated prompts caused by blockwise
processing. An experimental comparison shows that our proposed decoder-only
streaming ASR achieves 8% relative word error rate reduction in the LibriSpeech
test-other set while being twice as fast as the baseline model. | eess.AS | Decoder-only Architecture for Streaming End-to-end Speech Recognition |
2024-06-23T11:58:26Z | http://arxiv.org/pdf/2406.17807v1 | http://arxiv.org/abs/2406.17807v1 | Meiling Tao. Xuechen Liang, Yiling Tao, Tianyu Shi | 2024-06-23T11:58:26Z | cs.CL, cs.AI | Recent advancements in large language models (LLMs) have unlocked the
potential for generating high-quality game commentary. However, producing
insightful and engaging commentary for complex games with incomplete
information remains a significant challenge. In this paper, we introduce a
novel commentary method that combine Reinforcement Learning (RL) and LLMs,
tailored specifically for the Chinese card game \textit{Guandan}. Our system
leverages RL to generate intricate card-playing scenarios and employs LLMs to
generate corresponding commentary text, effectively emulating the strategic
analysis and narrative prowess of professional commentators. The framework
comprises a state commentary guide, a Theory of Mind (ToM)-based strategy
analyzer, and a style retrieval module, which seamlessly collaborate to deliver
detailed and context-relevant game commentary in the Chinese language
environment. We empower LLMs with ToM capabilities and refine both retrieval
and information filtering mechanisms. This facilitates the generation of
personalized commentary content. Our experimental results showcase the
substantial enhancement in performance achieved by the proposed commentary
framework when applied to open-source LLMs, surpassing the performance of GPT-4
across multiple evaluation metrics. | cs.CL | Enhancing Commentary Strategies for Imperfect Information Card Games: A Study of Large Language Models in Guandan Commentary |
2024-06-23T11:57:53Z | http://arxiv.org/pdf/2406.16086v1 | http://arxiv.org/abs/2406.16086v1 | Gili Lior, Avi Caciularu, Arie Cattan, Shahar Levy, Ori Shapira, Gabriel Stanovsky | 2024-06-23T11:57:53Z | cs.CL | Various tasks, such as summarization, multi-hop question answering, or
coreference resolution, are naturally phrased over collections of real-world
documents. Such tasks present a unique set of challenges, revolving around the
lack of coherent narrative structure across documents, which often leads to
contradiction, omission, or repetition of information. Despite their real-world
application and challenging properties, there is currently no benchmark which
specifically measures the abilities of large language models (LLMs) on
multi-document tasks. To bridge this gap, we present SEAM (a Stochastic
Evaluation Approach for Multi-document tasks), a conglomerate benchmark over a
diverse set of multi-document datasets, setting conventional evaluation
criteria, input-output formats, and evaluation protocols. In particular, SEAM
addresses the sensitivity of LLMs to minor prompt variations through repeated
evaluations, where in each evaluation we sample uniformly at random the values
of arbitrary factors (e.g., the order of documents). We evaluate different LLMs
on SEAM finding that multi-document tasks pose a significant challenge for
LLMs, even for state-of-the-art models with 70B parameters. In addition, we
show that the stochastic approach uncovers underlying statistical trends which
cannot be observed in a static benchmark. We hope that SEAM will spur progress
via consistent and meaningful evaluation of multi-document tasks. | cs.CL | SEAM: A Stochastic Benchmark for Multi-Document Tasks |
2024-06-23T11:18:55Z | http://arxiv.org/pdf/2406.16079v1 | http://arxiv.org/abs/2406.16079v1 | Zheng Li, Dawei Zhu, Qilong Ma, Weimin Xiong, Sujian Li | 2024-06-23T11:18:55Z | cs.CL, cs.AI | Personality is a fundamental construct in psychology, reflecting an
individual's behavior, thinking, and emotional patterns. Previous researches
have made some progress in personality detection, primarily by utilizing the
whole text to predict personality. However, these studies generally tend to
overlook psychological knowledge: they rarely apply the well-established
correlations between emotion regulation and personality. Based on this, we
propose a new personality detection method called EERPD. This method introduces
the use of emotion regulation, a psychological concept highly correlated with
personality, for personality prediction. By combining this feature with emotion
features, it retrieves few-shot examples and provides process CoTs for
inferring labels from text. This approach enhances the understanding of LLM for
personality within text and improves the performance in personality detection.
Experimental results demonstrate that EERPD significantly enhances the accuracy
and robustness of personality detection, outperforming previous SOTA by
15.05/4.29 in average F1 on the two benchmark datasets. | cs.CL | EERPD: Leveraging Emotion and Emotion Regulation for Improving Personality Detection |
2024-06-23T11:11:46Z | http://arxiv.org/pdf/2406.16078v1 | http://arxiv.org/abs/2406.16078v1 | Yoichi Aoki, Keito Kudo, Tatsuki Kuribayashi, Shusaku Sone, Masaya Taniguchi, Keisuke Sakaguchi, Kentaro Inui | 2024-06-23T11:11:46Z | cs.CL | Multi-step reasoning is widely adopted in the community to explore the better
performance of language models (LMs). We report on the systematic strategy that
LMs use in this process. Our controlled experiments reveal that LMs rely more
heavily on heuristics, such as lexical overlap, in the earlier stages of
reasoning when more steps are required to reach an answer. Conversely, as LMs
progress closer to the final answer, their reliance on heuristics decreases.
This suggests that LMs track only a limited number of future steps and
dynamically combine heuristic strategies with logical ones in tasks involving
multi-step reasoning. | cs.CL | First Heuristic Then Rational: Dynamic Use of Heuristics in Language Model Reasoning |
2024-06-23T10:47:01Z | http://arxiv.org/pdf/2406.16071v1 | http://arxiv.org/abs/2406.16071v1 | Carlos Gómez-Rodríguez, Muhammad Imran, David Vilares, Elena Solera, Olga Kellert | 2024-06-23T10:47:01Z | cs.CL, 68T50, I.2.7 | Sentiment analysis is a key technology for companies and institutions to
gauge public opinion on products, services or events. However, for large-scale
sentiment analysis to be accessible to entities with modest computational
resources, it needs to be performed in a resource-efficient way. While some
efficient sentiment analysis systems exist, they tend to apply shallow
heuristics, which do not take into account syntactic phenomena that can
radically change sentiment. Conversely, alternatives that take syntax into
account are computationally expensive. The SALSA project, funded by the
European Research Council under a Proof-of-Concept Grant, aims to leverage
recently-developed fast syntactic parsing techniques to build sentiment
analysis systems that are lightweight and efficient, while still providing
accuracy and explainability through the explicit use of syntax. We intend our
approaches to be the backbone of a working product of interest for SMEs to use
in production. | cs.CL | Dancing in the syntax forest: fast, accurate and explainable sentiment analysis with SALSA |
2024-06-23T10:36:35Z | http://arxiv.org/pdf/2406.16069v1 | http://arxiv.org/abs/2406.16069v1 | Junyi Zhu, Shuochen Liu, Yu Yu, Bo Tang, Yibo Yan, Zhiyu Li, Feiyu Xiong, Tong Xu, Matthew B. Blaschko | 2024-06-23T10:36:35Z | cs.CL, cs.AI | Large language models (LLMs) excel in generating coherent text, but they
often struggle with context awareness, leading to inaccuracies in tasks
requiring faithful adherence to provided information. We introduce FastMem, a
novel method designed to enhance instruction fine-tuned LLMs' context awareness
through fast memorization of the prompt. FastMem maximizes the likelihood of
the prompt before inference by fine-tuning only the last Feed-Forward Network
(FFN) module. This targeted approach ensures efficient optimization without
overfitting, significantly improving the model's ability to comprehend and
accurately follow the context. Our experiments demonstrate substantial gains in
reading comprehension, text summarization and adherence to output structures.
For instance, FastMem improves the accuracy of Llama 3-8B-Inst on the NQ-SWAP
dataset from 59.1% to 71.6%, and reduces the output structure failure rate of
Qwen 1.5-4B-Chat from 34.9% to 25.5%. Extensive experimental results highlight
FastMem's potential to offer a robust solution to enhance the reliability and
accuracy of LLMs in various applications. Our code is available at:
https://github.com/IAAR-Shanghai/FastMem | cs.CL | FastMem: Fast Memorization of Prompt Improves Context Awareness of Large Language Models |
2024-06-23T09:51:06Z | http://arxiv.org/pdf/2406.16061v1 | http://arxiv.org/abs/2406.16061v1 | Salem Lahlou, Abdalgader Abubaker, Hakim Hacid | 2024-06-23T09:51:06Z | cs.LG, cs.CL | Preference optimization methods have been successfully applied to improve not
only the alignment of large language models (LLMs) with human values, but also
specific natural language tasks such as summarization and stylistic
continuations. This paper proposes using preference optimization methods on
Chain-of-Thought steps in order to improve the reasoning performances of
language models. While the chosen answers are obtained from datasets that
include reasoning traces, we propose two complementary schemes for generating
rejected answers: digit corruption, and weak LLM prompting. Our approach leads
to increased accuracy on the GSM8K, AQuA-RAT, and ARC benchmarks for
Falcon2-11B and Mistral-7B. For example, the approach can lead to up to a
relative 8.47% increase in accuracy on the GSM8K benchmark without any extra
annotations. This work suggests that spending resources on creating more
datasets of reasoning traces would further boost LLM performances on informal
reasoning tasks. | cs.LG | PORT: Preference Optimization on Reasoning Traces |
2024-06-23T06:54:47Z | http://arxiv.org/pdf/2406.16033v1 | http://arxiv.org/abs/2406.16033v1 | Tianyi Men, Pengfei Cao, Zhuoran Jin, Yubo Chen, Kang Liu, Jun Zhao | 2024-06-23T06:54:47Z | cs.CL | Planning, as the core module of agents, is crucial in various fields such as
embodied agents, web navigation, and tool using. With the development of large
language models (LLMs), some researchers treat large language models as
intelligent agents to stimulate and evaluate their planning capabilities.
However, the planning mechanism is still unclear. In this work, we focus on
exploring the look-ahead planning mechanism in large language models from the
perspectives of information flow and internal representations. First, we study
how planning is done internally by analyzing the multi-layer perception (MLP)
and multi-head self-attention (MHSA) components at the last token. We find that
the output of MHSA in the middle layers at the last token can directly decode
the decision to some extent. Based on this discovery, we further trace the
source of MHSA by information flow, and we reveal that MHSA mainly extracts
information from spans of the goal states and recent steps. According to
information flow, we continue to study what information is encoded within it.
Specifically, we explore whether future decisions have been encoded in advance
in the representation of flow. We demonstrate that the middle and upper layers
encode a few short-term future decisions to some extent when planning is
successful. Overall, our research analyzes the look-ahead planning mechanisms
of LLMs, facilitating future research on LLMs performing planning tasks. | cs.CL | Unlocking the Future: Exploring Look-Ahead Planning Mechanistic Interpretability in Large Language Models |
2024-06-23T06:38:56Z | http://arxiv.org/pdf/2406.16030v1 | http://arxiv.org/abs/2406.16030v1 | Jimin Sohn, Haeji Jung, Alex Cheng, Jooeon Kang, Yilin Du, David R. Mortensen | 2024-06-23T06:38:56Z | cs.CL, cs.AI | Existing zero-shot cross-lingual NER approaches require substantial prior
knowledge of the target language, which is impractical for low-resource
languages. In this paper, we propose a novel approach to NER using phonemic
representation based on the International Phonetic Alphabet (IPA) to bridge the
gap between representations of different languages. Our experiments show that
our method significantly outperforms baseline models in extremely low-resource
languages, with the highest average F-1 score (46.38%) and lowest standard
deviation (12.67), particularly demonstrating its robustness with non-Latin
scripts. | cs.CL | Zero-Shot Cross-Lingual NER Using Phonemic Representations for Low-Resource Languages |
2024-06-23T06:01:11Z | http://arxiv.org/pdf/2406.16021v1 | http://arxiv.org/abs/2406.16021v1 | Qiang Gao, Zixiang Meng, Bobo Li, Jun Zhou, Fei Li, Chong Teng, Donghong Ji | 2024-06-23T06:01:11Z | cs.CL, cs.AI | Document-level event extraction aims to extract structured event information
from unstructured text. However, a single document often contains limited event
information and the roles of different event arguments may be biased due to the
influence of the information source. This paper addresses the limitations of
traditional document-level event extraction by proposing the task of
cross-document event extraction (CDEE) to integrate event information from
multiple documents and provide a comprehensive perspective on events. We
construct a novel cross-document event extraction dataset, namely CLES, which
contains 20,059 documents and 37,688 mention-level events, where over 70% of
them are cross-document. To build a benchmark, we propose a CDEE pipeline that
includes 5 steps, namely event extraction, coreference resolution, entity
normalization, role normalization and entity-role resolution. Our CDEE pipeline
achieves about 72% F1 in end-to-end cross-document event extraction, suggesting
the challenge of this task. Our work builds a new line of information
extraction research and will attract new research attention. | cs.CL | Harvesting Events from Multiple Sources: Towards a Cross-Document Event Extraction Paradigm |
2024-06-23T05:40:26Z | http://arxiv.org/pdf/2406.16020v2 | http://arxiv.org/abs/2406.16020v2 | Bin Wang, Xunlong Zou, Geyu Lin, Shuo Sun, Zhuohan Liu, Wenyu Zhang, Zhengyuan Liu, AiTi Aw, Nancy F. Chen | 2024-06-25T12:27:06Z | cs.SD, cs.CL, eess.AS | We introduce AudioBench, a new benchmark designed to evaluate audio large
language models (AudioLLMs). AudioBench encompasses 8 distinct tasks and 26
carefully selected or newly curated datasets, focusing on speech understanding,
voice interpretation, and audio scene understanding. Despite the rapid
advancement of large language models, including multimodal versions, a
significant gap exists in comprehensive benchmarks for thoroughly evaluating
their capabilities. AudioBench addresses this gap by providing relevant
datasets and evaluation metrics. In our study, we evaluated the capabilities of
four models across various aspects and found that no single model excels
consistently across all tasks. We outline the research outlook for AudioLLMs
and anticipate that our open-source code, data, and leaderboard will offer a
robust testbed for future model developments. | cs.SD | AudioBench: A Universal Benchmark for Audio Large Language Models |
2024-06-23T05:02:21Z | http://arxiv.org/pdf/2406.16013v1 | http://arxiv.org/abs/2406.16013v1 | Soyeong Jeong, Jinheon Baek, Sukmin Cho, Sung Ju Hwang, Jong C. Park | 2024-06-23T05:02:21Z | cs.CL, cs.AI, cs.IR | Information retrieval models that aim to search for the documents relevant to
the given query have shown many successes, which have been applied to diverse
tasks. However, the query provided by the user is oftentimes very short, which
challenges the retrievers to correctly fetch relevant documents. To tackle
this, existing studies have proposed expanding the query with a couple of
additional (user-related) features related to the query. Yet, they may be
suboptimal to effectively augment the query, though there is plenty of
information available to augment it in a relational database. Motivated by
this, we present a novel retrieval framework called Database-Augmented Query
representation (DAQu), which augments the original query with various
(query-related) metadata across multiple tables. In addition, as the number of
features in the metadata can be very large and there is no order among them, we
encode them with our graph-based set encoding strategy, which considers
hierarchies of features in the database without order. We validate DAQu in
diverse retrieval scenarios that can incorporate metadata from the relational
database, demonstrating that ours significantly enhances overall retrieval
performance, compared to existing query augmentation methods. | cs.CL | Database-Augmented Query Representation for Information Retrieval |
2024-06-23T04:35:42Z | http://arxiv.org/pdf/2406.16008v2 | http://arxiv.org/abs/2406.16008v2 | Cheng-Yu Hsieh, Yung-Sung Chuang, Chun-Liang Li, Zifeng Wang, Long T. Le, Abhishek Kumar, James Glass, Alexander Ratner, Chen-Yu Lee, Ranjay Krishna, Tomas Pfister | 2024-07-03T17:40:00Z | cs.CL, cs.AI, cs.LG | Large language models (LLMs), even when specifically trained to process long
input contexts, struggle to capture relevant information located in the middle
of their input. This phenomenon has been known as the lost-in-the-middle
problem. In this work, we make three contributions. First, we set out to
understand the factors that cause this phenomenon. In doing so, we establish a
connection between lost-in-the-middle to LLMs' intrinsic attention bias: LLMs
exhibit a U-shaped attention bias where the tokens at the beginning and at the
end of its input receive higher attention, regardless of their relevance.
Second, we mitigate this positional bias through a calibration mechanism,
found-in-the-middle, that allows the model to attend to contexts faithfully
according to their relevance, even though when they are in the middle. Third,
we show found-in-the-middle not only achieves better performance in locating
relevant information within a long context, but also eventually leads to
improved retrieval-augmented generation (RAG) performance across various tasks,
outperforming existing methods by up to 15 percentage points. These findings
open up future directions in understanding LLM attention bias and its potential
consequences. | cs.CL | Found in the Middle: Calibrating Positional Attention Bias Improves Long Context Utilization |
2024-06-23T04:29:13Z | http://arxiv.org/pdf/2406.16007v1 | http://arxiv.org/abs/2406.16007v1 | Bowen Zheng, Ming Ma, Zhongqiao Lin, Tianming Yang | 2024-06-23T04:29:13Z | cs.CL | Large Language Models (LLMs) have demonstrated remarkable abilities, one of
the most important being In-Context Learning (ICL). With ICL, LLMs can derive
the underlying rule from a few demonstrations and provide answers that comply
with the rule. Previous work hypothesized that the network creates a "task
vector" in specific positions during ICL. Patching the "task vector" allows
LLMs to achieve zero-shot performance similar to few-shot learning. However, we
discover that such "task vectors" do not exist in tasks where the rule has to
be defined through multiple demonstrations. Instead, the rule information
provided by each demonstration is first transmitted to its answer position and
forms its own rule vector. Importantly, all the rule vectors contribute to the
output in a distributed manner. We further show that the rule vectors encode a
high-level abstraction of rules extracted from the demonstrations. These
results are further validated in a series of tasks that rely on rules dependent
on multiple demonstrations. Our study provides novel insights into the
mechanism underlying ICL in LLMs, demonstrating how ICL may be achieved through
an information aggregation mechanism. | cs.CL | Distributed Rule Vectors is A Key Mechanism in Large Language Models' In-Context Learning |
2024-06-23T03:12:03Z | http://arxiv.org/pdf/2406.15996v1 | http://arxiv.org/abs/2406.15996v1 | Bumjin Park, Jaesik Choi | 2024-06-23T03:12:03Z | cs.CL, cs.AI | Training data plays a pivotal role in AI models. Large language models (LLMs)
are trained with massive amounts of documents, and their parameters hold
document-related contents. Recently, several studies identified
content-specific locations in LLMs by examining the parameters. Instead of the
post hoc interpretation, we propose another approach. We propose document-wise
memory architecture to track document memories in training. The proposed
architecture maps document representations to memory entries, which softly mask
memories in the forward process of LLMs. Additionally, we propose document
guidance loss, which increases the likelihood of text with document memories
and reduces the likelihood of the text with the memories of other documents.
Experimental results on Wikitext-103-v1 with Pythia-1B show that the proposed
methods provide different memory entries for documents and high recall of
document-related content in generation with trained document-wise memories. | cs.CL | Memorizing Documents with Guidance in Large Language Models |
2024-06-23T02:59:15Z | http://arxiv.org/pdf/2406.15992v1 | http://arxiv.org/abs/2406.15992v1 | Yizhuo Zhang, Heng Wang, Shangbin Feng, Zhaoxuan Tan, Xiaochuang Han, Tianxing He, Yulia Tsvetkov | 2024-06-23T02:59:15Z | cs.CL, I.2.7 | Large language models (LLMs) demonstrate great potential for problems with
implicit graphical structures, while recent works seek to enhance the graph
reasoning capabilities of LLMs through specialized instruction tuning. The
resulting 'graph LLMs' are evaluated with in-distribution settings only, thus
it remains underexplored whether LLMs are learning generalizable graph
reasoning skills or merely memorizing patterns in the synthetic training data.
To this end, we propose the NLGift benchmark, an evaluation suite of LLM graph
reasoning generalization: whether LLMs could go beyond semantic, numeric,
structural, reasoning patterns in the synthetic training data and improve
utility on real-world graph-based tasks. Extensive experiments with two LLMs
across four graph reasoning tasks demonstrate that while generalization on
simple patterns (semantic, numeric) is somewhat satisfactory, LLMs struggle to
generalize across reasoning and real-world patterns, casting doubt on the
benefit of synthetic graph tuning for real-world tasks with underlying network
structures. We explore three strategies to improve LLM graph reasoning
generalization, and we find that while post-training alignment is most
promising for real-world tasks, empowering LLM graph reasoning to go beyond
pattern memorization remains an open research question. | cs.CL | Can LLM Graph Reasoning Generalize beyond Pattern Memorization? |
2024-06-23T02:54:48Z | http://arxiv.org/pdf/2406.15990v1 | http://arxiv.org/abs/2406.15990v1 | Qiang Gao, Bobo Li, Zixiang Meng, Yunlong Li, Jun Zhou, Fei Li, Chong Teng, Donghong Ji | 2024-06-23T02:54:48Z | cs.CL, cs.AI | Existing cross-document event coreference resolution models, which either
compute mention similarity directly or enhance mention representation by
extracting event arguments (such as location, time, agent, and patient),
lacking the ability to utilize document-level information. As a result, they
struggle to capture long-distance dependencies. This shortcoming leads to their
underwhelming performance in determining coreference for the events where their
argument information relies on long-distance dependencies. In light of these
limitations, we propose the construction of document-level Rhetorical Structure
Theory (RST) trees and cross-document Lexical Chains to model the structural
and semantic information of documents. Subsequently, cross-document
heterogeneous graphs are constructed and GAT is utilized to learn the
representations of events. Finally, a pair scorer calculates the similarity
between each pair of events and co-referred events can be recognized using
standard clustering algorithm. Additionally, as the existing cross-document
event coreference datasets are limited to English, we have developed a
large-scale Chinese cross-document event coreference dataset to fill this gap,
which comprises 53,066 event mentions and 4,476 clusters. After applying our
model on the English and Chinese datasets respectively, it outperforms all
baselines by large margins. | cs.CL | Enhancing Cross-Document Event Coreference Resolution by Discourse Structure and Semantic Information |
2024-06-23T02:02:52Z | http://arxiv.org/pdf/2406.15981v1 | http://arxiv.org/abs/2406.15981v1 | Xiaobo Guo, Soroush Vosoughi | 2024-06-23T02:02:52Z | cs.CL | Large Language Models (LLMs) have shown remarkable capabilities in zero-shot
learning applications, generating responses to queries using only pre-training
information without the need for additional fine-tuning. This represents a
significant departure from traditional machine learning approaches. Previous
research has indicated that LLMs may exhibit serial position effects, such as
primacy and recency biases, which are well-documented cognitive biases in human
psychology. Our extensive testing across various tasks and models confirms the
widespread occurrence of these effects, although their intensity varies. We
also discovered that while carefully designed prompts can somewhat mitigate
these biases, their effectiveness is inconsistent. These findings underscore
the significance of serial position effects during the inference process,
particularly in scenarios where there are no ground truth labels, highlighting
the need for greater focus on addressing these effects in LLM applications. | cs.CL | Serial Position Effects of Large Language Models |
2024-06-23T00:23:13Z | http://arxiv.org/pdf/2406.15968v1 | http://arxiv.org/abs/2406.15968v1 | Roy Xie, Junlin Wang, Ruomin Huang, Minxing Zhang, Rong Ge, Jian Pei, Neil Zhenqiang Gong, Bhuwan Dhingra | 2024-06-23T00:23:13Z | cs.CL, cs.LG | The rapid scaling of large language models (LLMs) has raised concerns about
the transparency and fair use of the pretraining data used for training them.
Detecting such content is challenging due to the scale of the data and limited
exposure of each instance during training. We propose ReCaLL (Relative
Conditional Log-Likelihood), a novel membership inference attack (MIA) to
detect LLMs' pretraining data by leveraging their conditional language modeling
capabilities. ReCaLL examines the relative change in conditional
log-likelihoods when prefixing target data points with non-member context. Our
empirical findings show that conditioning member data on non-member prefixes
induces a larger decrease in log-likelihood compared to non-member data. We
conduct comprehensive experiments and show that ReCaLL achieves
state-of-the-art performance on the WikiMIA dataset, even with random and
synthetic prefixes, and can be further improved using an ensemble approach.
Moreover, we conduct an in-depth analysis of LLMs' behavior with different
membership contexts, providing insights into how LLMs leverage membership
information for effective inference at both the sequence and token level. | cs.CL | ReCaLL: Membership Inference via Relative Conditional Log-Likelihoods |
2024-06-23T00:11:07Z | http://arxiv.org/pdf/2406.15966v1 | http://arxiv.org/abs/2406.15966v1 | Hassan Alhuzali, Ashwag Alasmari | 2024-06-23T00:11:07Z | cs.CL, cs.AI | Pre-trained Language Models (PLMs) have the potential to transform mental
health support by providing accessible and culturally sensitive resources.
However, despite this potential, their effectiveness in mental health care and
specifically for the Arabic language has not been extensively explored. To
bridge this gap, this study evaluates the effectiveness of foundational models
for classification of Questions and Answers (Q&A) in the domain of mental
health care. We leverage the MentalQA dataset, an Arabic collection featuring
Q&A interactions related to mental health. In this study, we conducted
experiments using four different types of learning approaches: traditional
feature extraction, PLMs as feature extractors, Fine-tuning PLMs and prompting
large language models (GPT-3.5 and GPT-4) in zero-shot and few-shot learning
settings. While traditional feature extractors combined with Support Vector
Machines (SVM) showed promising performance, PLMs exhibited even better results
due to their ability to capture semantic meaning. For example, MARBERT achieved
the highest performance with a Jaccard Score of 0.80 for question
classification and a Jaccard Score of 0.86 for answer classification. We
further conducted an in-depth analysis including examining the effects of
fine-tuning versus non-fine-tuning, the impact of varying data size, and
conducting error analysis. Our analysis demonstrates that fine-tuning proved to
be beneficial for enhancing the performance of PLMs, and the size of the
training data played a crucial role in achieving high performance. We also
explored prompting, where few-shot learning with GPT-3.5 yielded promising
results. There was an improvement of 12% for question and classification and
45% for answer classification. Based on our findings, it can be concluded that
PLMs and prompt-based approaches hold promise for mental health support in
Arabic. | cs.CL | Evaluating the Effectiveness of the Foundational Models for Q&A Classification in Mental Health care |
2024-06-23T00:04:07Z | http://arxiv.org/pdf/2406.15963v1 | http://arxiv.org/abs/2406.15963v1 | Mengxuan Sun, Ehud Reiter, Anne E Kiltie, George Ramsay, Lisa Duncan, Peter Murchie, Rosalind Adam | 2024-06-23T00:04:07Z | cs.HC, cs.CL, q-bio.OT | Electronic health records contain detailed information about the medical
condition of patients, but they are difficult for patients to understand even
if they have access to them. We explore whether ChatGPT (GPT 4) can help
explain multidisciplinary team (MDT) reports to colorectal and prostate cancer
patients. These reports are written in dense medical language and assume
clinical knowledge, so they are a good test of the ability of ChatGPT to
explain complex medical reports to patients. We asked clinicians and lay people
(not patients) to review explanations and responses of ChatGPT. We also ran
three focus groups (including cancer patients, caregivers, computer scientists,
and clinicians) to discuss output of ChatGPT. Our studies highlighted issues
with inaccurate information, inappropriate language, limited personalization,
AI distrust, and challenges integrating large language models (LLMs) into
clinical workflow. These issues will need to be resolved before LLMs can be
used to explain complex personal medical information to patients. | cs.HC | Effectiveness of ChatGPT in explaining complex medical reports to patients |
2024-06-22T23:26:07Z | http://arxiv.org/pdf/2406.17806v1 | http://arxiv.org/abs/2406.17806v1 | Xirui Li, Hengguang Zhou, Ruochen Wang, Tianyi Zhou, Minhao Cheng, Cho-Jui Hsieh | 2024-06-22T23:26:07Z | cs.CL, cs.AI, cs.CR, cs.CV, cs.LG | Humans are prone to cognitive distortions -- biased thinking patterns that
lead to exaggerated responses to specific stimuli, albeit in very different
contexts. This paper demonstrates that advanced Multimodal Large Language
Models (MLLMs) exhibit similar tendencies. While these models are designed to
respond queries under safety mechanism, they sometimes reject harmless queries
in the presence of certain visual stimuli, disregarding the benign nature of
their contexts. As the initial step in investigating this behavior, we identify
three types of stimuli that trigger the oversensitivity of existing MLLMs:
Exaggerated Risk, Negated Harm, and Counterintuitive Interpretation. To
systematically evaluate MLLMs' oversensitivity to these stimuli, we propose the
Multimodal OverSenSitivity Benchmark (MOSSBench). This toolkit consists of 300
manually collected benign multimodal queries, cross-verified by third-party
reviewers (AMT). Empirical studies using MOSSBench on 20 MLLMs reveal several
insights: (1). Oversensitivity is prevalent among SOTA MLLMs, with refusal
rates reaching up to 76% for harmless queries. (2). Safer models are more
oversensitive: increasing safety may inadvertently raise caution and
conservatism in the model's responses. (3). Different types of stimuli tend to
cause errors at specific stages -- perception, intent reasoning, and safety
judgement -- in the response process of MLLMs. These findings highlight the
need for refined safety mechanisms that balance caution with contextually
appropriate responses, improving the reliability of MLLMs in real-world
applications. We make our project available at
https://turningpoint-ai.github.io/MOSSBench/. | cs.CL | MOSSBench: Is Your Multimodal Language Model Oversensitive to Safe Queries? |
2024-06-22T22:59:09Z | http://arxiv.org/pdf/2406.17805v1 | http://arxiv.org/abs/2406.17805v1 | Darius Coelho, Harshit Barot, Naitik Rathod, Klaus Mueller | 2024-06-22T22:59:09Z | cs.CL, cs.AI, cs.HC | Recent advancements in large language models have revolutionized information
access, as these models harness data available on the web to address complex
queries, becoming the preferred information source for many users. In certain
cases, queries are about publicly available data, which can be effectively
answered with data visualizations. In this paper, we investigate the ability of
large language models to provide accurate data and relevant visualizations in
response to such queries. Specifically, we investigate the ability of GPT-3 and
GPT-4 to generate visualizations with dataless prompts, where no data
accompanies the query. We evaluate the results of the models by comparing them
to visualization cheat sheets created by visualization experts. | cs.CL | Can LLMs Generate Visualizations with Dataless Prompts? |
2024-06-22T22:07:40Z | http://arxiv.org/pdf/2406.15951v1 | http://arxiv.org/abs/2406.15951v1 | Shangbin Feng, Taylor Sorensen, Yuhan Liu, Jillian Fisher, Chan Young Park, Yejin Choi, Yulia Tsvetkov | 2024-06-22T22:07:40Z | cs.CL | While existing alignment paradigms have been integral in developing large
language models (LLMs), LLMs often learn an averaged human preference and
struggle to model diverse preferences across cultures, demographics, and
communities. We propose Modular Pluralism, a modular framework based on
multi-LLM collaboration for pluralistic alignment: it "plugs into" a base LLM a
pool of smaller but specialized community LMs, where models collaborate in
distinct modes to flexibility support three modes of pluralism: Overton,
steerable, and distributional. Modular Pluralism is uniquely compatible with
black-box LLMs and offers the modular control of adding new community LMs for
previously underrepresented communities. We evaluate Modular Pluralism with six
tasks and four datasets featuring questions/instructions with value-laden and
perspective-informed responses. Extensive experiments demonstrate that Modular
Pluralism advances the three pluralism objectives across six black-box and
open-source LLMs. Further analysis reveals that LLMs are generally faithful to
the inputs from smaller community LLMs, allowing seamless patching by adding a
new community LM to better cover previously underrepresented communities. | cs.CL | Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration |
2024-06-22T21:59:12Z | http://arxiv.org/pdf/2406.15948v1 | http://arxiv.org/abs/2406.15948v1 | Shangbin Feng, Weijia Shi, Yike Wang, Wenxuan Ding, Orevaoghene Ahia, Shuyue Stella Li, Vidhisha Balachandran, Sunayana Sitaram, Yulia Tsvetkov | 2024-06-22T21:59:12Z | cs.CL | Multilingual LLMs often have knowledge disparities across languages, with
larger gaps in under-resourced languages. Teaching LLMs to abstain in the face
of knowledge gaps is thus a promising strategy to mitigate hallucinations in
multilingual settings. However, previous studies on LLM abstention primarily
focus on English; we find that directly applying existing solutions beyond
English results in up to 20.5% performance gaps between high and low-resource
languages, potentially due to LLMs' drop in calibration and reasoning beyond a
few resource-rich languages. To this end, we propose strategies to enhance LLM
abstention by learning from multilingual feedback, where LLMs self-reflect on
proposed answers in one language by generating multiple feedback items in
related languages: we show that this helps identifying the knowledge gaps
across diverse languages, cultures, and communities. Extensive experiments
demonstrate that our multilingual feedback approach outperforms various strong
baselines, achieving up to 9.2% improvement for low-resource languages across
three black-box and open models on three datasets, featuring open-book,
closed-book, and commonsense QA. Further analysis reveals that multilingual
feedback is both an effective and a more equitable abstain strategy to serve
diverse language speakers, and cultural factors have great impact on language
selection and LLM abstention behavior, highlighting future directions for
multilingual and multi-cultural reliable language modeling. | cs.CL | Teaching LLMs to Abstain across Languages via Multilingual Feedback |
2024-06-22T21:12:57Z | http://arxiv.org/pdf/2406.15940v1 | http://arxiv.org/abs/2406.15940v1 | Christopher Burger, Yifan Hu, Thai Le | 2024-06-22T21:12:57Z | cs.LG, cs.AI, cs.CL | The location of knowledge within Generative Pre-trained Transformer
(GPT)-like models has seen extensive recent investigation. However, much of the
work is focused towards determining locations of individual facts, with the end
goal being the editing of facts that are outdated, erroneous, or otherwise
harmful, without the time and expense of retraining the entire model. In this
work, we investigate a broader view of knowledge location, that of concepts or
clusters of related information, instead of disparate individual facts. To do
this, we first curate a novel dataset, called DARC, that includes a total of 34
concepts of ~120K factual statements divided into two types of hierarchical
categories, namely taxonomy and meronomy. Next, we utilize existing causal
mediation analysis methods developed for determining regions of importance for
individual facts and apply them to a series of related categories to provide
detailed investigation into whether concepts are associated with distinct
regions within these models. We find that related categories exhibit similar
areas of importance in contrast to less similar categories. However,
fine-grained localization of individual category subsets to specific regions is
not apparent. | cs.LG | Beyond Individual Facts: Investigating Categorical Knowledge Locality of Taxonomy and Meronomy Concepts in GPT Models |
2024-06-22T20:57:12Z | http://arxiv.org/pdf/2406.15938v1 | http://arxiv.org/abs/2406.15938v1 | Ming Li, Han Chen, Chenguang Wang, Dang Nguyen, Dianqi Li, Tianyi Zhou | 2024-06-22T20:57:12Z | cs.CL, cs.AI, cs.LG | Large language models (LLMs) still lack delicate controllability over their
responses, which is critical to enhancing their performance and the user
experience. However, curating supervised fine-tuning (SFT) datasets to improve
LLM controllability usually relies on human experts or proprietary LLMs, which
requires additional costs. To bridge this gap, we propose Rule-based Data
Recycling (RuleR), a data augmentation method incorporating multiple
constraints into the original data samples according to predefined rules, which
creates new training tasks to consolidate the controllability of LLMs. Instead
of creating new data from scratch, RuleR ``recycles'' existing data by simply
applying rule-based edits to their responses and appending the
rule-instructions in their original instructions. Experimental results
demonstrate RuleR's effectiveness in improving LLM controllability while
maintaining general instruction-following capabilities. The code will be
released on https://github.com/MingLiiii/RuleR. | cs.CL | RuleR: Improving LLM Controllability by Rule-based Data Recycling |
2024-06-22T19:46:06Z | http://arxiv.org/pdf/2406.15927v1 | http://arxiv.org/abs/2406.15927v1 | Jannik Kossen, Jiatong Han, Muhammed Razzak, Lisa Schut, Shreshth Malik, Yarin Gal | 2024-06-22T19:46:06Z | cs.CL, cs.AI, cs.LG | We propose semantic entropy probes (SEPs), a cheap and reliable method for
uncertainty quantification in Large Language Models (LLMs). Hallucinations,
which are plausible-sounding but factually incorrect and arbitrary model
generations, present a major challenge to the practical adoption of LLMs.
Recent work by Farquhar et al. (2024) proposes semantic entropy (SE), which can
detect hallucinations by estimating uncertainty in the space semantic meaning
for a set of model generations. However, the 5-to-10-fold increase in
computation cost associated with SE computation hinders practical adoption. To
address this, we propose SEPs, which directly approximate SE from the hidden
states of a single generation. SEPs are simple to train and do not require
sampling multiple model generations at test time, reducing the overhead of
semantic uncertainty quantification to almost zero. We show that SEPs retain
high performance for hallucination detection and generalize better to
out-of-distribution data than previous probing methods that directly predict
model accuracy. Our results across models and tasks suggest that model hidden
states capture SE, and our ablation studies give further insights into the
token positions and model layers for which this is the case. | cs.CL | Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs |
2024-06-22T17:01:59Z | http://arxiv.org/pdf/2406.15891v1 | http://arxiv.org/abs/2406.15891v1 | Carlos Gómez-Rodríguez, Paul Williams | 2024-06-22T17:01:59Z | cs.CL, 68T50, I.2.7 | This is a summary of the paper "A Confederacy of Models: a Comprehensive
Evaluation of LLMs on Creative Writing", which was published in Findings of
EMNLP 2023. We evaluate a range of recent state-of-the-art, instruction-tuned
large language models (LLMs) on an English creative writing task, and compare
them to human writers. For this purpose, we use a specifically-tailored prompt
(based on an epic combat between Ignatius J. Reilly, main character of John
Kennedy Toole's "A Confederacy of Dunces", and a pterodactyl) to minimize the
risk of training data leakage and force the models to be creative rather than
reusing existing stories. The same prompt is presented to LLMs and human
writers, and evaluation is performed by humans using a detailed rubric
including various aspects like fluency, style, originality or humor. Results
show that some state-of-the-art commercial LLMs match or slightly outperform
our human writers in most of the evaluated dimensions. Open-source LLMs lag
behind. Humans keep a close lead in originality, and only the top three LLMs
can handle humor at human-like levels. | cs.CL | The Unlikely Duel: Evaluating Creative Writing in LLMs through a Unique Scenario |
2024-06-22T16:55:21Z | http://arxiv.org/pdf/2406.15890v1 | http://arxiv.org/abs/2406.15890v1 | Ari Azarafrooz, Farshid Faal | 2024-06-22T16:55:21Z | cs.LG, cs.AI, cs.CL, cs.GT | Recent research has shown the potential of Nash Learning via Human Feedback
for large language model alignment by incorporating the notion of a preference
model in a minimax game setup. We take this idea further by casting the
alignment as a mirror descent algorithm against the adaptive feedback of an
improved opponent, thereby removing the need for learning a preference model or
the existence of an annotated dataset altogether. The resulting algorithm,
which we refer to as Language Alignment via Nash-learning and Adaptive feedback
(LANA), is capable of self-alignment without the need for a human-annotated
preference dataset. We support this statement with various experiments and
mathematical discussion. | cs.LG | Language Alignment via Nash-learning and Adaptive feedback |
2024-06-22T16:37:51Z | http://arxiv.org/pdf/2406.15888v1 | http://arxiv.org/abs/2406.15888v1 | Khai Le-Duc, Khai-Nguyen Nguyen, Long Vo-Dang, Truong-Son Hy | 2024-06-22T16:37:51Z | cs.CL, cs.AI, cs.LG, cs.SD, eess.AS | In doctor-patient conversations, identifying medically relevant information
is crucial, posing the need for conversation summarization. In this work, we
propose the first deployable real-time speech summarization system for
real-world applications in industry, which generates a local summary after
every N speech utterances within a conversation and a global summary after the
end of a conversation. Our system could enhance user experience from a business
standpoint, while also reducing computational costs from a technical
perspective. Secondly, we present VietMed-Sum which, to our knowledge, is the
first speech summarization dataset for medical conversations. Thirdly, we are
the first to utilize LLM and human annotators collaboratively to create gold
standard and synthetic summaries for medical conversation summarization.
Finally, we present baseline results of state-of-the-art models on VietMed-Sum.
All code, data (English-translated and Vietnamese) and models are available
online: https://github.com/leduckhai/MultiMed | cs.CL | Real-time Speech Summarization for Medical Conversations |
2024-06-22T16:10:45Z | http://arxiv.org/pdf/2406.15883v1 | http://arxiv.org/abs/2406.15883v1 | Giang Do, Hung Le, Truyen Tran | 2024-06-22T16:10:45Z | cs.CL, cs.AI | Sparse mixture of experts (SMoE) have emerged as an effective approach for
scaling large language models while keeping a constant computational cost.
Regardless of several notable successes of SMoE, effective training such
architecture remains elusive due to the representation collapse problem, which
in turn harms model performance and causes parameter redundancy. In this work,
we present Similarity-based Sparse Mixture of Experts (SimSMoE), a novel
similarity of neural network algorithm, that guarantees a solution to address
the representation collapse issue between experts given a fixed FLOPs budget.
We conduct extensive empirical evaluations on three large language models for
both Pre-training and Fine-tuning tasks to illustrate the efficacy, robustness,
and scalability of our method. The results demonstrate that SimSMoE
significantly enhances existing routing policy and outperforms other SMoE
training methods in performance for the tasks. | cs.CL | SimSMoE: Solving Representational Collapse via Similarity Measure |
2024-06-22T15:52:04Z | http://arxiv.org/pdf/2406.15877v2 | http://arxiv.org/abs/2406.15877v2 | Terry Yue Zhuo, Minh Chien Vu, Jenny Chim, Han Hu, Wenhao Yu, Ratnadira Widyasari, Imam Nur Bani Yusuf, Haolan Zhan, Junda He, Indraneil Paul, Simon Brunner, Chen Gong, Thong Hoang, Armel Randy Zebaze, Xiaoheng Hong, Wen-Ding Li, Jean Kaddour, Ming Xu, Zhihan Zhang, Prateek Yadav, Naman Jain, Alex Gu, Zhoujun Cheng, Jiawei Liu, Qian Liu, Zijian Wang, David Lo, Binyuan Hui, Niklas Muennighoff, Daniel Fried, Xiaoning Du, Harm de Vries, Leandro Von Werra | 2024-06-26T17:05:14Z | cs.SE, cs.AI, cs.CL | Automated software engineering has been greatly empowered by the recent
advances in Large Language Models (LLMs) for programming. While current
benchmarks have shown that LLMs can perform various software engineering tasks
like human developers, the majority of their evaluations are limited to short
and self-contained algorithmic tasks. Solving challenging and practical
programming tasks requires the capability of utilizing diverse function calls
as tools to efficiently implement functionalities like data analysis and web
development. In addition, using multiple tools to solve a task needs
compositional reasoning by accurately understanding complex instructions.
Fulfilling both of these characteristics can pose a great challenge for LLMs.
To assess how well LLMs can solve challenging and practical programming tasks,
we introduce Bench, a benchmark that challenges LLMs to invoke multiple
function calls as tools from 139 libraries and 7 domains for 1,140 fine-grained
programming tasks. To evaluate LLMs rigorously, each programming task
encompasses 5.6 test cases with an average branch coverage of 99%. In addition,
we propose a natural-language-oriented variant of Bench, Benchi, that
automatically transforms the original docstrings into short instructions only
with essential information. Our extensive evaluation of 60 LLMs shows that LLMs
are not yet capable of following complex instructions to use function calls
precisely, with scores up to 60%, significantly lower than the human
performance of 97%. The results underscore the need for further advancements in
this area. | cs.SE | BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions |
2024-06-22T15:16:11Z | http://arxiv.org/pdf/2406.15871v1 | http://arxiv.org/abs/2406.15871v1 | Louis Give, Timo Zaoral, Maria Antonietta Bruno | 2024-06-22T15:16:11Z | cs.CL, cs.AI | Today, the detection of AI-generated content is receiving more and more
attention. Our idea is to go beyond detection and try to recover the prompt
used to generate a text. This paper, to the best of our knowledge, introduces
the first investigation in this particular domain without a closed set of
tasks. Our goal is to study if this approach is promising. We experiment with
zero-shot and few-shot in-context learning but also with LoRA fine-tuning.
After that, we evaluate the benefits of using a semi-synthetic dataset. For
this first study, we limit ourselves to text generated by a single model. The
results show that it is possible to recover the original prompt with a
reasonable degree of accuracy. | cs.CL | Uncovering Hidden Intentions: Exploring Prompt Recovery for Deeper Insights into Generated Texts |
2024-06-22T15:06:08Z | http://arxiv.org/pdf/2406.15869v1 | http://arxiv.org/abs/2406.15869v1 | Jason Angel, Segun Taofeek Aroyehun, Grigori Sidorov, Alexander Gelbukh | 2024-06-22T15:06:08Z | cs.CL | Identifying misogyny using artificial intelligence is a form of combating
online toxicity against women. However, the subjective nature of interpreting
misogyny poses a significant challenge to model the phenomenon. In this paper,
we propose a multitask learning approach that leverages the subjectivity of
this task to enhance the performance of the misogyny identification systems. We
incorporated diverse perspectives from annotators in our model design,
considering gender and age across six profile groups, and conducted extensive
experiments and error analysis using two language models to validate our four
alternative designs of the multitask learning technique to identify
misogynistic content in English tweets. The results demonstrate that
incorporating various viewpoints enhances the language models' ability to
interpret different forms of misogyny. This research advances content
moderation and highlights the importance of embracing diverse perspectives to
build effective online moderation systems. | cs.CL | A multitask learning framework for leveraging subjectivity of annotators to identify misogyny |
2024-06-22T14:32:35Z | http://arxiv.org/pdf/2406.17803v1 | http://arxiv.org/abs/2406.17803v1 | Bin Wu, Zhengyan Shi, Hossein A. Rahmani, Varsha Ramineni, Emine Yilmaz | 2024-06-22T14:32:35Z | cs.CL, cs.AI, cs.IR | Utilizing user profiles to personalize Large Language Models (LLMs) has been
shown to enhance the performance on a wide range of tasks. However, the precise
role of user profiles and their effect mechanism on LLMs remains unclear. This
study first confirms that the effectiveness of user profiles is primarily due
to personalization information rather than semantic information. Furthermore,
we investigate how user profiles affect the personalization of LLMs. Within the
user profile, we reveal that it is the historical personalized response
produced or approved by users that plays a pivotal role in personalizing LLMs.
This discovery unlocks the potential of LLMs to incorporate a greater number of
user profiles within the constraints of limited input length. As for the
position of user profiles, we observe that user profiles integrated into
different positions of the input context do not contribute equally to
personalization. Instead, where the user profile that is closer to the
beginning affects more on the personalization of LLMs. Our findings reveal the
role of user profiles for the personalization of LLMs, and showcase how
incorporating user profiles impacts performance providing insight to leverage
user profiles effectively. | cs.CL | Understanding the Role of User Profile in the Personalization of Large Language Models |
2024-06-22T14:19:51Z | http://arxiv.org/pdf/2406.15862v1 | http://arxiv.org/abs/2406.15862v1 | Moreno La Quatra, Alkis Koudounas, Elena Baralis, Sabato Marco Siniscalchi | 2024-06-22T14:19:51Z | cs.CL | Italy exhibits rich linguistic diversity across its territory due to the
distinct regional languages spoken in different areas. Recent advances in
self-supervised learning provide new opportunities to analyze Italy's
linguistic varieties using speech data alone. This includes the potential to
leverage representations learned from large amounts of data to better examine
nuances between closely related linguistic varieties. In this study, we focus
on automatically identifying the geographic region of origin of speech samples
drawn from Italy's diverse language varieties. We leverage self-supervised
learning models to tackle this task and analyze differences and similarities
between Italy's regional languages. In doing so, we also seek to uncover new
insights into the relationships among these diverse yet closely related
varieties, which may help linguists understand their interconnected evolution
and regional development over time and space. To improve the discriminative
ability of learned representations, we evaluate several supervised contrastive
learning objectives, both as pre-training steps and additional fine-tuning
objectives. Experimental evidence shows that pre-trained self-supervised models
can effectively identify regions from speech recording. Additionally,
incorporating contrastive objectives during fine-tuning improves classification
accuracy and yields embeddings that distinctly separate regional varieties,
demonstrating the value of combining self-supervised pre-training and
contrastive learning for this task. | cs.CL | Speech Analysis of Language Varieties in Italy |
2024-06-22T13:24:58Z | http://arxiv.org/pdf/2406.15846v1 | http://arxiv.org/abs/2406.15846v1 | Chen Xu, Jie Wang, Xiaoqian Liu, Qianqian Dong, Chunliang Zhang, Tong Xiao, Jingbo Zhu, Dapeng Man, Wu Yang | 2024-06-22T13:24:58Z | cs.CL, eess.AS | Speech-to-text (S2T) generation systems frequently face challenges in
low-resource scenarios, primarily due to the lack of extensive labeled
datasets. One emerging solution is constructing virtual training samples by
interpolating inputs and labels, which has notably enhanced system
generalization in other domains. Despite its potential, this technique's
application in S2T tasks has remained under-explored. In this paper, we delve
into the utility of interpolation augmentation, guided by several pivotal
questions. Our findings reveal that employing an appropriate strategy in
interpolation augmentation significantly enhances performance across diverse
tasks, architectures, and data scales, offering a promising avenue for more
robust S2T systems in resource-constrained settings. | cs.CL | Revisiting Interpolation Augmentation for Speech-to-Text Generation |
2024-06-22T12:50:41Z | http://arxiv.org/pdf/2407.02514v1 | http://arxiv.org/abs/2407.02514v1 | Shashank Kirtania, Priyanshu Gupta, Arjun Radhakirshna | 2024-06-22T12:50:41Z | cs.LO, cs.AI, cs.CL | In this paper we examine the limitations of Large Language Models (LLMs) for
complex reasoning tasks. While current approaches leverage formal languages as
intermediate representation of reasoning problems, they struggle with
generating intermediate formal specifications and refining these
representations. To address these issues, this paper proposes Logic-LM++, an
improvement on Logic-LM. It uses the ability of LLMs to do pairwise
comparisons, allowing the evaluation of the refinements suggested by the LLM.
The paper demonstrates that Logic-LM++ outperforms Logic-LM and LLM based
techniques on natural language reasoning tasks on two datasets, FOLIO and
AR-LSAT. Logic-LM++ show an average improvement of 13.5% on standard prompting,
11% on chain of thought prompting and 5% on Logic-LM. | cs.LO | LOGIC-LM++: Multi-Step Refinement for Symbolic Formulations |
2024-06-22T11:46:04Z | http://arxiv.org/pdf/2406.15823v1 | http://arxiv.org/abs/2406.15823v1 | Yash Kumar Lal, Vanya Cohen, Nathanael Chambers, Niranjan Balasubramanian, Raymond Mooney | 2024-06-22T11:46:04Z | cs.CL | Understanding the abilities of LLMs to reason about natural language plans,
such as instructional text and recipes, is critical to reliably using them in
decision-making systems. A fundamental aspect of plans is the temporal order in
which their steps needs to be executed, which reflects the underlying causal
dependencies between them. We introduce CaT-Bench, a benchmark of Step Order
Prediction questions, which test whether a step must necessarily occur before
or after another in cooking recipe plans. We use this to evaluate how well
frontier LLMs understand causal and temporal dependencies. We find that SOTA
LLMs are underwhelming (best zero-shot is only 0.59 in F1), and are biased
towards predicting dependence more often, perhaps relying on temporal order of
steps as a heuristic. While prompting for explanations and using few-shot
examples improve performance, the best F1 result is only 0.73. Further, human
evaluation of explanations along with answer correctness show that, on average,
humans do not agree with model reasoning. Surprisingly, we also find that
explaining after answering leads to better performance than normal
chain-of-thought prompting, and LLM answers are not consistent across questions
about the same step pairs. Overall, results show that LLMs' ability to detect
dependence between steps has significant room for improvement. | cs.CL | CaT-BENCH: Benchmarking Language Model Understanding of Causal and Temporal Dependencies in Plans |
2024-06-22T10:49:36Z | http://arxiv.org/pdf/2406.17801v1 | http://arxiv.org/abs/2406.17801v1 | Xiaopeng Wang, Yi Lu, Xin Qi, Zhiyong Wang, Yuankun Xie, Shuchen Shi, Ruibo Fu | 2024-06-22T10:49:36Z | cs.SD, cs.CL, eess.AS | This paper presents the development of a speech synthesis system for the
LIMMITS'24 Challenge, focusing primarily on Track 2. The objective of the
challenge is to establish a multi-speaker, multi-lingual Indic Text-to-Speech
system with voice cloning capabilities, covering seven Indian languages with
both male and female speakers. The system was trained using challenge data and
fine-tuned for few-shot voice cloning on target speakers. Evaluation included
both mono-lingual and cross-lingual synthesis across all seven languages, with
subjective tests assessing naturalness and speaker similarity. Our system uses
the VITS2 architecture, augmented with a multi-lingual ID and a BERT model to
enhance contextual language comprehension. In Track 1, where no additional data
usage was permitted, our model achieved a Speaker Similarity score of 4.02. In
Track 2, which allowed the use of extra data, it attained a Speaker Similarity
score of 4.17. | cs.SD | A multi-speaker multi-lingual voice cloning system based on vits2 for limmits 2024 challenge |
2024-06-22T10:36:04Z | http://arxiv.org/pdf/2406.15812v1 | http://arxiv.org/abs/2406.15812v1 | Lorenzo Basile, Santiago Acevedo, Luca Bortolussi, Fabio Anselmi, Alex Rodriguez | 2024-06-22T10:36:04Z | cs.LG, cs.CL, cs.CV | To gain insight into the mechanisms behind machine learning methods, it is
crucial to establish connections among the features describing data points.
However, these correlations often exhibit a high-dimensional and strongly
nonlinear nature, which makes them challenging to detect using standard
methods. This paper exploits the entanglement between intrinsic dimensionality
and correlation to propose a metric that quantifies the (potentially nonlinear)
correlation between high-dimensional manifolds. We first validate our method on
synthetic data in controlled environments, showcasing its advantages and
drawbacks compared to existing techniques. Subsequently, we extend our analysis
to large-scale applications in neural network representations. Specifically, we
focus on latent representations of multimodal data, uncovering clear
correlations between paired visual and textual embeddings, whereas existing
methods struggle significantly in detecting similarity. Our results indicate
the presence of highly nonlinear correlation patterns between latent manifolds. | cs.LG | Intrinsic Dimension Correlation: uncovering nonlinear connections in multimodal representations |
2024-06-22T10:25:55Z | http://arxiv.org/pdf/2406.15809v1 | http://arxiv.org/abs/2406.15809v1 | Garima Chhikara, Anurag Sharma, V. Gurucharan, Kripabandhu Ghosh, Abhijnan Chakraborty | 2024-06-22T10:25:55Z | cs.CL, cs.LG | Large Language Models (LLMs) have demonstrated impressive performance across
a wide range of NLP tasks, including summarization. Inherently LLMs produce
abstractive summaries, and the task of achieving extractive summaries through
LLMs still remains largely unexplored. To bridge this gap, in this work, we
propose a novel framework LaMSUM to generate extractive summaries through LLMs
for large user-generated text by leveraging voting algorithms. Our evaluation
on three popular open-source LLMs (Llama 3, Mixtral and Gemini) reveal that the
LaMSUM outperforms state-of-the-art extractive summarization methods. We
further attempt to provide the rationale behind the output summary produced by
LLMs. Overall, this is one of the early attempts to achieve extractive
summarization for large user-generated text by utilizing LLMs, and likely to
generate further interest in the community. | cs.CL | LaMSUM: A Novel Framework for Extractive Summarization of User Generated Content using LLMs |
Subsets and Splits