publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-10-25T23:32:12Z | http://arxiv.org/pdf/2310.17054v1 | http://arxiv.org/abs/2310.17054v1 | Yufei Tian, Felix Zhang, Nanyun Peng | 2023-10-25T23:32:12Z | cs.CL, cs.LG | Large language models (LLMs) such as GPT-3 have demonstrated a strong
capability to generate coherent and contextually relevant text. However, amidst
their successes, a crucial issue persists: their generated outputs still lack
commonsense at times. Moreover, fine-tuning the entire LLM towards more
commonsensical outputs is computationally expensive if not infeasible. In this
paper, we present a computation-efficient framework that steers a frozen
Pre-Trained Language Model (PTLM) towards more commonsensical generation (i.e.,
producing a plausible output that incorporates a list of concepts in a
meaningful way). Specifically, we first construct a reference-free evaluator
that assigns a sentence with a commonsensical score by grounding the sentence
to a dynamic commonsense knowledge base from four different relational aspects.
We then use the scorer as the oracle for commonsense knowledge, and extend the
controllable generation method called NADO to train an auxiliary head that
guides a fixed PTLM to better satisfy the oracle. We test our framework on a
series of GPT-2-, Flan-T5-, and Alpaca-based language models (LMs) on two
constrained concept-to-sentence benchmarks. Human evaluation results
demonstrate that our method consistently leads to the most commonsensical
outputs. | cs.CL | BOOST: Harnessing Black-Box Control to Boost Commonsense in LMs' Generation |
2023-10-25T22:42:30Z | http://arxiv.org/pdf/2310.17041v1 | http://arxiv.org/abs/2310.17041v1 | Abhilasha Lodha, Gayatri Belapurkar, Saloni Chalkapurkar, Yuanming Tao, Reshmi Ghosh, Samyadeep Basu, Dmitrii Petrov, Soundararajan Srinivasan | 2023-10-25T22:42:30Z | cs.CL, cs.AI, cs.IR | Fine-tuning all the layers of a pre-trained neural language encoder (either
using all the parameters or using parameter-efficient methods) is often the
de-facto way of adapting it to a new task. We show evidence that for different
downstream language tasks, fine-tuning only a subset of layers is sufficient to
obtain performance that is close to and often better than fine-tuning all the
layers in the language encoder. We propose an efficient metric based on the
diagonal of the Fisher information matrix (FIM score), to select the candidate
layers for selective fine-tuning. We show, empirically on GLUE and SuperGLUE
tasks and across distinct language encoders, that this metric can effectively
select layers leading to a strong downstream performance. Our work highlights
that task-specific information corresponding to a given downstream task is
often localized within a few layers, and tuning only those is sufficient for
strong performance. Additionally, we demonstrate the robustness of the FIM
score to rank layers in a manner that remains constant during the optimization
process. | cs.CL | On Surgical Fine-tuning for Language Encoders |
2023-10-25T22:36:40Z | http://arxiv.org/pdf/2310.18369v1 | http://arxiv.org/abs/2310.18369v1 | Daniela Ben-David, Tzuf Paz-Argaman, Reut Tsarfaty | 2023-10-25T22:36:40Z | cs.CL, cs.AI, cs.CV, I.2.7; I.5.4 | We propose a modular framework that leverages the expertise of different
foundation models over different modalities and domains in order to perform a
single, complex, multi-modal task, without relying on prompt engineering or
otherwise tailor-made multi-modal training. Our approach enables decentralized
command execution and allows each model to both contribute and benefit from the
expertise of the other models. Our method can be extended to a variety of
foundation models (including audio and vision), above and beyond only language
models, as it does not depend on prompts. We demonstrate our approach on two
tasks. On the well-known task of stylized image captioning, our experiments
show that our approach outperforms semi-supervised state-of-the-art models,
while being zero-shot and avoiding costly training, data collection, and prompt
engineering. We further demonstrate this method on a novel task, audio-aware
image captioning, in which an image and audio are given and the task is to
generate text that describes the image within the context of the provided
audio. Our code is available on GitHub. | cs.CL | Apollo: Zero-shot MultiModal Reasoning with Multiple Experts |
2023-10-25T22:22:18Z | http://arxiv.org/pdf/2310.17034v1 | http://arxiv.org/abs/2310.17034v1 | Besnik Fetahu, Pedro Faustini, Giuseppe Castellucci, Anjie Fang, Oleg Rokhlenko, Shervin Malmasi | 2023-10-25T22:22:18Z | cs.CL | The adoption of voice assistants like Alexa or Siri has grown rapidly,
allowing users to instantly access information via voice search. Query
suggestion is a standard feature of screen-based search experiences, allowing
users to explore additional topics. However, this is not trivial to implement
in voice-based settings. To enable this, we tackle the novel task of suggesting
questions with compact and natural voice hints to allow users to ask follow-up
questions.
We define the task, ground it in syntactic theory and outline linguistic
desiderata for spoken hints. We propose baselines and an approach using
sequence-to-sequence Transformers to generate spoken hints from a list of
questions. Using a new dataset of 6681 input questions and human written hints,
we evaluated the models with automatic metrics and human evaluation. Results
show that a naive approach of concatenating suggested questions creates poor
voice hints. Our approach, which applies a linguistically-motivated pretraining
task was strongly preferred by humans for producing the most natural hints. | cs.CL | Follow-on Question Suggestion via Voice Hints for Voice Assistants |
2023-10-25T22:00:05Z | http://arxiv.org/pdf/2310.17022v3 | http://arxiv.org/abs/2310.17022v3 | Sidharth Mudgal, Jong Lee, Harish Ganapathy, YaGuang Li, Tao Wang, Yanping Huang, Zhifeng Chen, Heng-Tze Cheng, Michael Collins, Trevor Strohman, Jilin Chen, Alex Beutel, Ahmad Beirami | 2024-06-03T20:50:26Z | cs.LG, cs.AI, cs.CL | KL-regularized reinforcement learning (RL) is a popular alignment framework
to control the language model responses towards high reward outcomes. We pose a
tokenwise RL objective and propose a modular solver for it, called controlled
decoding (CD). CD exerts control through a separate prefix scorer module, which
is trained to learn a value function for the reward. The prefix scorer is used
at inference time to control the generation from a frozen base model, provably
sampling from a solution to the RL objective. We empirically demonstrate that
CD is effective as a control mechanism on popular benchmarks. We also show that
prefix scorers for multiple rewards may be combined at inference time,
effectively solving a multi-objective RL problem with no additional training.
We show that the benefits of applying CD transfer to an unseen base model with
no further tuning as well. Finally, we show that CD can be applied in a
blockwise decoding fashion at inference-time, essentially bridging the gap
between the popular best-of-K strategy and tokenwise control through
reinforcement learning. This makes CD a promising approach for alignment of
language models. | cs.LG | Controlled Decoding from Language Models |
2023-10-25T21:46:34Z | http://arxiv.org/pdf/2310.17019v1 | http://arxiv.org/abs/2310.17019v1 | K. R. Zentner, Ryan Julian, Brian Ichter, Gaurav S. Sukhatme | 2023-10-25T21:46:34Z | cs.LG, cs.CL, cs.RO | This paper combines two contributions. First, we introduce an extension of
the Meta-World benchmark, which we call "Language-World," which allows a large
language model to operate in a simulated robotic environment using
semi-structured natural language queries and scripted skills described using
natural language. By using the same set of tasks as Meta-World, Language-World
results can be easily compared to Meta-World results, allowing for a point of
comparison between recent methods using Large Language Models (LLMs) and those
using Deep Reinforcement Learning. Second, we introduce a method we call Plan
Conditioned Behavioral Cloning (PCBC), that allows finetuning the behavior of
high-level plans using end-to-end demonstrations. Using Language-World, we show
that PCBC is able to achieve strong performance in a variety of few-shot
regimes, often achieving task generalization with as little as a single
demonstration. We have made Language-World available as open-source software at
https://github.com/krzentner/language-world/. | cs.LG | Conditionally Combining Robot Skills using Large Language Models |
2023-10-25T21:37:57Z | http://arxiv.org/pdf/2310.17017v1 | http://arxiv.org/abs/2310.17017v1 | Young Min Cho, Sunny Rai, Lyle Ungar, João Sedoc, Sharath Chandra Guntuku | 2023-10-25T21:37:57Z | cs.CL, cs.AI | Mental health conversational agents (a.k.a. chatbots) are widely studied for
their potential to offer accessible support to those experiencing mental health
challenges. Previous surveys on the topic primarily consider papers published
in either computer science or medicine, leading to a divide in understanding
and hindering the sharing of beneficial knowledge between both domains. To
bridge this gap, we conduct a comprehensive literature review using the PRISMA
framework, reviewing 534 papers published in both computer science and
medicine. Our systematic review reveals 136 key papers on building mental
health-related conversational agents with diverse characteristics of modeling
and experimental design techniques. We find that computer science papers focus
on LLM techniques and evaluating response quality using automated metrics with
little attention to the application while medical papers use rule-based
conversational agents and outcome metrics to measure the health outcomes of
participants. Based on our findings on transparency, ethics, and cultural
heterogeneity in this review, we provide a few recommendations to help bridge
the disciplinary divide and enable the cross-disciplinary development of mental
health conversational agents. | cs.CL | An Integrative Survey on Mental Health Conversational Agents to Bridge Computer Science and Medical Perspectives |
2023-10-25T21:29:36Z | http://arxiv.org/pdf/2310.17015v3 | http://arxiv.org/abs/2310.17015v3 | Anna Koufakou, Diego Grisales, Ragy Costa de jesus, Oscar Fox | 2023-10-30T13:33:16Z | cs.CL | Emotion recognition in text, the task of identifying emotions such as joy or
anger, is a challenging problem in NLP with many applications. One of the
challenges is the shortage of available datasets that have been annotated with
emotions. Certain existing datasets are small, follow different emotion
taxonomies and display imbalance in their emotion distribution. In this work,
we studied the impact of data augmentation techniques precisely when applied to
small imbalanced datasets, for which current state-of-the-art models (such as
RoBERTa) under-perform. Specifically, we utilized four data augmentation
methods (Easy Data Augmentation EDA, static and contextual Embedding-based, and
ProtAugment) on three datasets that come from different sources and vary in
size, emotion categories and distributions. Our experimental results show that
using the augmented data when training the classifier model leads to
significant improvements. Finally, we conducted two case studies: a) directly
using the popular chat-GPT API to paraphrase text using different prompts, and
b) using external data to augment the training set. Results show the promising
potential of these methods. | cs.CL | Data Augmentation for Emotion Detection in Small Imbalanced Text Data |
2023-10-25T21:18:35Z | http://arxiv.org/pdf/2310.17010v1 | http://arxiv.org/abs/2310.17010v1 | Claudio Fanconi, Moritz Vandenhirtz, Severin Husmann, Julia E. Vogt | 2023-10-25T21:18:35Z | cs.CL, cs.AI | Prototype learning, a popular machine learning method designed for inherently
interpretable decisions, leverages similarities to learned prototypes for
classifying new data. While it is mainly applied in computer vision, in this
work, we build upon prior research and further explore the extension of
prototypical networks to natural language processing. We introduce a learned
weighted similarity measure that enhances the similarity computation by
focusing on informative dimensions of pre-trained sentence embeddings.
Additionally, we propose a post-hoc explainability mechanism that extracts
prediction-relevant words from both the prototype and input sentences. Finally,
we empirically demonstrate that our proposed method not only improves
predictive performance on the AG News and RT Polarity datasets over a previous
prototype-based approach, but also improves the faithfulness of explanations
compared to rationale-based recurrent convolutions. | cs.CL | This Reads Like That: Deep Learning for Interpretable Natural Language Processing |
2023-10-25T20:48:16Z | http://arxiv.org/pdf/2310.16995v1 | http://arxiv.org/abs/2310.16995v1 | Saptarshi Sengupta, Connor Heaton, Shreya Ghosh, Preslav Nakov, Prasenjit Mitra | 2023-10-25T20:48:16Z | cs.CL | Domain adaptation, the process of training a model in one domain and applying
it to another, has been extensively explored in machine learning. While
training a domain-specific foundation model (FM) from scratch is an option,
recent methods have focused on adapting pre-trained FMs for domain-specific
tasks. However, our experiments reveal that either approach does not
consistently achieve state-of-the-art (SOTA) results in the target domain. In
this work, we study extractive question answering within closed domains and
introduce the concept of targeted pre-training. This involves determining and
generating relevant data to further pre-train our models, as opposed to the
conventional philosophy of utilizing domain-specific FMs trained on a wide
range of data. Our proposed framework uses Galactica to generate synthetic,
``targeted'' corpora that align with specific writing styles and topics, such
as research papers and radiology reports. This process can be viewed as a form
of knowledge distillation. We apply our method to two biomedical extractive
question answering datasets, COVID-QA and RadQA, achieving a new benchmark on
the former and demonstrating overall improvements on the latter. Code available
at https://github.com/saptarshi059/CDQA-v1-Targetted-PreTraining/tree/main. | cs.CL | Quality > Quantity: Synthetic Corpora from Foundation Models for Closed-Domain Extractive Question Answering |
2023-10-25T20:43:07Z | http://arxiv.org/pdf/2310.16992v1 | http://arxiv.org/abs/2310.16992v1 | Sinclair Schneider, Florian Steuber, Joao A. G. Schneider, Gabi Dreo Rodosek | 2023-10-25T20:43:07Z | cs.CL | With the rise of generative pre-trained transformer models such as GPT-3,
GPT-NeoX, or OPT, distinguishing human-generated texts from machine-generated
ones has become important. We refined five separate language models to generate
synthetic tweets, uncovering that shallow learning classification algorithms,
like Naive Bayes, achieve detection accuracy between 0.6 and 0.8.
Shallow learning classifiers differ from human-based detection, especially
when using higher temperature values during text generation, resulting in a
lower detection rate. Humans prioritize linguistic acceptability, which tends
to be higher at lower temperature values. In contrast, transformer-based
classifiers have an accuracy of 0.9 and above. We found that using a
reinforcement learning approach to refine our generative models can
successfully evade BERT-based classifiers with a detection accuracy of 0.15 or
less. | cs.CL | How well can machine-generated texts be identified and can language models be trained to avoid identification? |
2023-10-25T20:41:30Z | http://arxiv.org/pdf/2310.16990v1 | http://arxiv.org/abs/2310.16990v1 | Leon Liyang Zhang, Jiarui Lu, Joel Ruben Antony Moniz, Aditya Kulkarni, Dhivya Piraviperumal, Tien Dung Tran, Nicholas Tzou, Hong Yu | 2023-10-25T20:41:30Z | cs.CL, cs.AI, cs.LG | In the context of a voice assistant system, steering refers to the phenomenon
in which a user issues a follow-up command attempting to direct or clarify a
previous turn. We propose STEER, a steering detection model that predicts
whether a follow-up turn is a user's attempt to steer the previous command.
Constructing a training dataset for steering use cases poses challenges due to
the cold-start problem. To overcome this, we developed heuristic rules to
sample opt-in usage data, approximating positive and negative samples without
any annotation. Our experimental results show promising performance in
identifying steering intent, with over 95% accuracy on our sampled data.
Moreover, STEER, in conjunction with our sampling strategy, aligns effectively
with real-world steering scenarios, as evidenced by its strong zero-shot
performance on a human-graded evaluation set. In addition to relying solely on
user transcripts as input, we introduce STEER+, an enhanced version of the
model. STEER+ utilizes a semantic parse tree to provide more context on
out-of-vocabulary words, such as named entities that often occur at the
sentence boundary. This further improves model performance, reducing error rate
in domains where entities frequently appear, such as messaging. Lastly, we
present a data analysis that highlights the improvement in user experience when
voice assistants support steering use cases. | cs.CL | STEER: Semantic Turn Extension-Expansion Recognition for Voice Assistants |
2023-10-25T20:09:14Z | http://arxiv.org/pdf/2310.16968v1 | http://arxiv.org/abs/2310.16968v1 | Nafis Irtiza Tripto, Mohammed Eunus Ali | 2023-10-25T20:09:14Z | cs.CL, cs.CY | Social structures and real-world incidents often influence contemporary
literary fiction. Existing research in literary fiction analysis explains these
real-world phenomena through the manual critical analysis of stories.
Conventional Natural Language Processing (NLP) methodologies, including
sentiment analysis, narrative summarization, and topic modeling, have
demonstrated substantial efficacy in analyzing and identifying similarities
within fictional works. However, the intricate dynamics of character
interactions within fiction necessitate a more nuanced approach that
incorporates visualization techniques. Character interaction graphs (or
networks) emerge as a highly suitable means for visualization and information
retrieval from the realm of fiction. Therefore, we leverage character
interaction graphs with NLP-derived features to explore a diverse spectrum of
societal inquiries about contemporary culture's impact on the landscape of
literary fiction. Our study involves constructing character interaction graphs
from fiction, extracting relevant graph features, and exploiting these features
to resolve various real-life queries. Experimental evaluation of influential
Bengali fiction over half a century demonstrates that character interaction
graphs can be highly effective in specific assessments and information
retrieval from literary fiction. Our data and codebase are available at
https://cutt.ly/fbMgGEM | cs.CL | Understanding Social Structures from Contemporary Literary Fiction using Character Interaction Graph -- Half Century Chronology of Influential Bengali Writers |
2023-10-25T20:05:07Z | http://arxiv.org/pdf/2310.16964v1 | http://arxiv.org/abs/2310.16964v1 | Mateusz Lango, Ondřej Dušek | 2023-10-25T20:05:07Z | cs.CL, I.2.7 | Hallucination of text ungrounded in the input is a well-known problem in
neural data-to-text generation. Many methods have been proposed to mitigate it,
but they typically require altering model architecture or collecting additional
data, and thus cannot be easily applied to an existing model. In this paper, we
explore a new way to mitigate hallucinations by combining the probabilistic
output of a generator language model (LM) with the output of a special "text
critic" classifier, which guides the generation by assessing the match between
the input data and the text generated so far. Our method does not need any
changes to the underlying LM's architecture or training procedure and can thus
be combined with any model and decoding operating on word probabilities. The
critic does not need any additional training data, using the base LM's training
data and synthetic negative examples. Our experimental results show that our
method improves over the baseline on the WebNLG and OpenDialKG benchmarks. | cs.CL | Critic-Driven Decoding for Mitigating Hallucinations in Data-to-text Generation |
2023-10-25T19:39:58Z | http://arxiv.org/pdf/2310.18368v2 | http://arxiv.org/abs/2310.18368v2 | Babak Hemmatian, Razan Baltaji, Lav R. Varshney | 2023-12-09T18:11:06Z | cs.CL, I.2.7 | Abid et al. (2021) showed a tendency in GPT-3 to generate mostly violent
completions when prompted about Muslims, compared with other religions. Two
pre-registered replication attempts found few violent completions and only a
weak anti-Muslim bias in the more recent InstructGPT, fine-tuned to eliminate
biased and toxic outputs. However, more pre-registered experiments showed that
using common names associated with the religions in prompts increases
several-fold the rate of violent completions, revealing a significant
second-order anti-Muslim bias. ChatGPT showed a bias many times stronger
regardless of prompt format, suggesting that the effects of debiasing were
reduced with continued model development. Our content analysis revealed
religion-specific themes containing offensive stereotypes across all
experiments. Our results show the need for continual de-biasing of models in
ways that address both explicit and higher-order associations. | cs.CL | Muslim-Violence Bias Persists in Debiased GPT Models |
2023-10-25T19:25:16Z | http://arxiv.org/pdf/2310.16944v1 | http://arxiv.org/abs/2310.16944v1 | Lewis Tunstall, Edward Beeching, Nathan Lambert, Nazneen Rajani, Kashif Rasul, Younes Belkada, Shengyi Huang, Leandro von Werra, Clémentine Fourrier, Nathan Habib, Nathan Sarrazin, Omar Sanseviero, Alexander M. Rush, Thomas Wolf | 2023-10-25T19:25:16Z | cs.LG, cs.CL | We aim to produce a smaller language model that is aligned to user intent.
Previous research has shown that applying distilled supervised fine-tuning
(dSFT) on larger models significantly improves task accuracy; however, these
models are unaligned, i.e. they do not respond well to natural prompts. To
distill this property, we experiment with the use of preference data from AI
Feedback (AIF). Starting from a dataset of outputs ranked by a teacher model,
we apply distilled direct preference optimization (dDPO) to learn a chat model
with significantly improved intent alignment. The approach requires only a few
hours of training without any additional sampling during fine-tuning. The final
result, Zephyr-7B, sets the state-of-the-art on chat benchmarks for 7B
parameter models, and requires no human annotation. In particular, results on
MT-Bench show that Zephyr-7B surpasses Llama2-Chat-70B, the best open-access
RLHF-based model. Code, models, data, and tutorials for the system are
available at https://github.com/huggingface/alignment-handbook. | cs.LG | Zephyr: Direct Distillation of LM Alignment |
2023-10-25T19:04:33Z | http://arxiv.org/pdf/2310.16937v2 | http://arxiv.org/abs/2310.16937v2 | Razan Baltaji, Saurabh Pujar, Louis Mandel, Martin Hirzel, Luca Buratti, Lav Varshney | 2024-03-25T20:14:07Z | cs.CL, I.2.7; I.2.5 | Large language models (LLMs) have become remarkably good at improving
developer productivity for high-resource programming languages. These models
use two kinds of data: large amounts of unlabeled code samples for pre-training
and relatively smaller amounts of labeled code samples for fine-tuning or
in-context learning. Unfortunately, many programming languages are
low-resource, lacking labeled samples for most tasks and often even lacking
unlabeled samples. Therefore, users of low-resource languages (e.g., legacy or
new languages) miss out on the benefits of LLMs. Cross-lingual transfer uses
data from a source language to improve model performance on a target language.
It has been well-studied for natural languages, but has received little
attention for programming languages. This paper reports extensive experiments
on four tasks using a transformer-based LLM and 11 to 41 programming languages
to explore the following questions. First, how well does cross-lingual transfer
work for a given task across different language pairs. Second, given a task and
target language, how should one choose a source language. Third, which
characteristics of a language pair are predictive of transfer performance, and
how does that depend on the given task. Our empirical study with 1,808
experiments reveals practical and scientific insights, such as Kotlin and
JavaScript being the most transferable source languages and different tasks
relying on substantially different features. Overall, we find that learning
transfers well across several programming languages. | cs.CL | Learning Transfers over Several Programming Languages |
2023-10-25T18:55:40Z | http://arxiv.org/pdf/2310.16931v1 | http://arxiv.org/abs/2310.16931v1 | Luca Della Libera, Pooneh Mousavi, Salah Zaiem, Cem Subakan, Mirco Ravanelli | 2023-10-25T18:55:40Z | cs.CL, cs.AI | Modern multilingual automatic speech recognition (ASR) systems like Whisper
have made it possible to transcribe audio in multiple languages with a single
model. However, current state-of-the-art ASR models are typically evaluated on
individual languages or in a multi-task setting, overlooking the challenge of
continually learning new languages. There is insufficient research on how to
add new languages without losing valuable information from previous data.
Furthermore, existing continual learning benchmarks focus mostly on vision and
language tasks, leaving continual learning for multilingual ASR largely
unexplored. To bridge this gap, we propose CL-MASR, a benchmark designed for
studying multilingual ASR in a continual learning setting. CL-MASR provides a
diverse set of continual learning methods implemented on top of large-scale
pretrained ASR models, along with common metrics to assess the effectiveness of
learning new languages while addressing the issue of catastrophic forgetting.
To the best of our knowledge, CL-MASR is the first continual learning benchmark
for the multilingual ASR task. The code is available at
https://github.com/speechbrain/benchmarks. | cs.CL | CL-MASR: A Continual Learning Benchmark for Multilingual ASR |
2023-10-25T18:44:14Z | http://arxiv.org/pdf/2310.16924v1 | http://arxiv.org/abs/2310.16924v1 | Nikita Mehandru, Sweta Agrawal, Yimin Xiao, Elaine C Khoong, Ge Gao, Marine Carpuat, Niloufar Salehi | 2023-10-25T18:44:14Z | cs.CL, cs.HC | A major challenge in the practical use of Machine Translation (MT) is that
users lack guidance to make informed decisions about when to rely on outputs.
Progress in quality estimation research provides techniques to automatically
assess MT quality, but these techniques have primarily been evaluated in vitro
by comparison against human judgments outside of a specific context of use.
This paper evaluates quality estimation feedback in vivo with a human study
simulating decision-making in high-stakes medical settings. Using Emergency
Department discharge instructions, we study how interventions based on quality
estimation versus backtranslation assist physicians in deciding whether to show
MT outputs to a patient. We find that quality estimation improves appropriate
reliance on MT, but backtranslation helps physicians detect more clinically
harmful errors that QE alone often misses. | cs.CL | Physician Detection of Clinical Harm in Machine Translation: Quality Estimation Aids in Reliance and Backtranslation Identifies Critical Errors |
2023-10-25T18:00:15Z | http://arxiv.org/pdf/2310.16897v1 | http://arxiv.org/abs/2310.16897v1 | Solveig Helland, Elena Gavagnin, Alexandre de Spindler | 2023-10-25T18:00:15Z | cs.CL | The growing capabilities of transformer models pave the way for solving
increasingly complex NLP tasks. A key to supporting application-specific
requirements is the ability to fine-tune. However, compiling a fine-tuning
dataset tailored to complex tasks is tedious and results in large datasets,
limiting the ability to control transformer output. We present an approach in
which complex tasks are divided into simpler subtasks. Multiple transformer
models are fine-tuned to one subtask each, and lined up to accomplish the
complex task. This simplifies the compilation of fine-tuning datasets and
increases overall controllability. Using the example of reducing gender bias as
a complex task, we demonstrate our approach and show that it performs better
than using a single model. | cs.CL | Divide et Impera: Multi-Transformer Architectures for Complex NLP-Tasks |
2023-10-25T17:59:32Z | http://arxiv.org/pdf/2310.16836v1 | http://arxiv.org/abs/2310.16836v1 | Shih-yang Liu, Zechun Liu, Xijie Huang, Pingcheng Dong, Kwang-Ting Cheng | 2023-10-25T17:59:32Z | cs.CL, cs.AI, cs.AR, cs.CV | We propose LLM-FP4 for quantizing both weights and activations in large
language models (LLMs) down to 4-bit floating-point values, in a post-training
manner. Existing post-training quantization (PTQ) solutions are primarily
integer-based and struggle with bit widths below 8 bits. Compared to integer
quantization, floating-point (FP) quantization is more flexible and can better
handle long-tail or bell-shaped distributions, and it has emerged as a default
choice in many hardware platforms. One characteristic of FP quantization is
that its performance largely depends on the choice of exponent bits and
clipping range. In this regard, we construct a strong FP-PTQ baseline by
searching for the optimal quantization parameters. Furthermore, we observe a
high inter-channel variance and low intra-channel variance pattern in
activation distributions, which adds activation quantization difficulty. We
recognize this pattern to be consistent across a spectrum of transformer models
designed for diverse tasks, such as LLMs, BERT, and Vision Transformer models.
To tackle this, we propose per-channel activation quantization and show that
these additional scaling factors can be reparameterized as exponential biases
of weights, incurring a negligible cost. Our method, for the first time, can
quantize both weights and activations in the LLaMA-13B to only 4-bit and
achieves an average score of 63.1 on the common sense zero-shot reasoning
tasks, which is only 5.8 lower than the full-precision model, significantly
outperforming the previous state-of-the-art by 12.7 points. Code is available
at: https://github.com/nbasyl/LLM-FP4. | cs.CL | LLM-FP4: 4-Bit Floating-Point Quantized Transformers |
2023-10-25T17:59:12Z | http://arxiv.org/pdf/2310.16834v3 | http://arxiv.org/abs/2310.16834v3 | Aaron Lou, Chenlin Meng, Stefano Ermon | 2024-06-06T21:06:44Z | stat.ML, cs.CL, cs.LG | Despite their groundbreaking performance for many generative modeling tasks,
diffusion models have fallen short on discrete data domains such as natural
language. Crucially, standard diffusion models rely on the well-established
theory of score matching, but efforts to generalize this to discrete structures
have not yielded the same empirical gains. In this work, we bridge this gap by
proposing score entropy, a novel loss that naturally extends score matching to
discrete spaces, integrates seamlessly to build discrete diffusion models, and
significantly boosts performance. Experimentally, we test our Score Entropy
Discrete Diffusion models (SEDD) on standard language modeling tasks. For
comparable model sizes, SEDD beats existing language diffusion paradigms
(reducing perplexity by $25$-$75$\%) and is competitive with autoregressive
models, in particular outperforming GPT-2. Furthermore, compared to
autoregressive mdoels, SEDD generates faithful text without requiring
distribution annealing techniques like temperature scaling (around
$6$-$8\times$ better generative perplexity than un-annealed GPT-2), can trade
compute and quality (similar quality with $32\times$ fewer network
evaluations), and enables controllable infilling (matching nucleus sampling
quality while enabling other strategies besides left to right prompting). | stat.ML | Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution |
2023-10-25T17:51:56Z | http://arxiv.org/pdf/2310.16822v1 | http://arxiv.org/abs/2310.16822v1 | Xuming Hu, Junzhe Chen, Aiwei Liu, Shiao Meng, Lijie Wen, Philip S. Yu | 2023-10-25T17:51:56Z | cs.CL, cs.AI, cs.MM | How can we better extract entities and relations from text? Using multimodal
extraction with images and text obtains more signals for entities and
relations, and aligns them through graphs or hierarchical fusion, aiding in
extraction. Despite attempts at various fusions, previous works have overlooked
many unlabeled image-caption pairs, such as NewsCLIPing. This paper proposes
innovative pre-training objectives for entity-object and relation-image
alignment, extracting objects from images and aligning them with entity and
relation prompts for soft pseudo-labels. These labels are used as
self-supervised signals for pre-training, enhancing the ability to extract
entities and relations. Experiments on three datasets show an average 3.41% F1
improvement over prior SOTA. Additionally, our method is orthogonal to previous
multimodal fusions, and using it on prior SOTA fusions further improves 5.47%
F1. | cs.CL | Prompt Me Up: Unleashing the Power of Alignments for Multimodal Entity and Relation Extraction |
2023-10-25T17:39:07Z | http://arxiv.org/pdf/2310.16810v1 | http://arxiv.org/abs/2310.16810v1 | Yongxin Zhou, Fabien Ringeval, François Portet | 2023-10-25T17:39:07Z | cs.CL, cs.AI | This study explores the capabilities of prompt-driven Large Language Models
(LLMs) like ChatGPT and GPT-4 in adhering to human guidelines for dialogue
summarization. Experiments employed DialogSum (English social conversations)
and DECODA (French call center interactions), testing various prompts:
including prompts from existing literature and those from human summarization
guidelines, as well as a two-step prompt approach. Our findings indicate that
GPT models often produce lengthy summaries and deviate from human summarization
guidelines. However, using human guidelines as an intermediate step shows
promise, outperforming direct word-length constraint prompts in some cases. The
results reveal that GPT models exhibit unique stylistic tendencies in their
summaries. While BERTScores did not dramatically decrease for GPT outputs
suggesting semantic similarity to human references and specialised pre-trained
models, ROUGE scores reveal grammatical and lexical disparities between
GPT-generated and human-written summaries. These findings shed light on the
capabilities and limitations of GPT models in following human instructions for
dialogue summarization. | cs.CL | Can GPT models Follow Human Summarization Guidelines? Evaluating ChatGPT and GPT-4 for Dialogue Summarization |
2023-10-25T17:34:52Z | http://arxiv.org/pdf/2310.16803v1 | http://arxiv.org/abs/2310.16803v1 | Saiteja Utpala, Alex Gu, Pin Yu Chen | 2023-10-25T17:34:52Z | cs.CL, cs.LG | Recently, code language models have achieved notable advancements in
addressing a diverse array of essential code comprehension and generation
tasks. Yet, the field lacks a comprehensive deep dive and understanding of the
code embeddings of multilingual code models. In this paper, we present a
comprehensive study on multilingual code embeddings, focusing on the
cross-lingual capabilities of these embeddings across different programming
languages. Through probing experiments, we demonstrate that code embeddings
comprise two distinct components: one deeply tied to the nuances and syntax of
a specific language, and the other remaining agnostic to these details,
primarily focusing on semantics. Further, we show that when we isolate and
eliminate this language-specific component, we witness significant improvements
in downstream code retrieval tasks, leading to an absolute increase of up to
+17 in the Mean Reciprocal Rank (MRR). | cs.CL | Language Agnostic Code Embeddings |
2023-10-25T17:23:37Z | http://arxiv.org/pdf/2310.16790v1 | http://arxiv.org/abs/2310.16790v1 | Zhendong Chu, Ruiyi Zhang, Tong Yu, Rajiv Jain, Vlad I Morariu, Jiuxiang Gu, Ani Nenkova | 2023-10-25T17:23:37Z | cs.CL, cs.AI, cs.LG | To achieve state-of-the-art performance, one still needs to train NER models
on large-scale, high-quality annotated data, an asset that is both costly and
time-intensive to accumulate. In contrast, real-world applications often resort
to massive low-quality labeled data through non-expert annotators via
crowdsourcing and external knowledge bases via distant supervision as a
cost-effective alternative. However, these annotation methods result in noisy
labels, which in turn lead to a notable decline in performance. Hence, we
propose to denoise the noisy NER data with guidance from a small set of clean
instances. Along with the main NER model we train a discriminator model and use
its outputs to recalibrate the sample weights. The discriminator is capable of
detecting both span and category errors with different discriminative prompts.
Results on public crowdsourcing and distant supervision datasets show that the
proposed method can consistently improve performance with a small guidance set. | cs.CL | Improving a Named Entity Recognizer Trained on Noisy Data with a Few Clean Instances |
2023-10-25T17:21:23Z | http://arxiv.org/pdf/2310.16789v3 | http://arxiv.org/abs/2310.16789v3 | Weijia Shi, Anirudh Ajith, Mengzhou Xia, Yangsibo Huang, Daogao Liu, Terra Blevins, Danqi Chen, Luke Zettlemoyer | 2024-03-09T22:26:06Z | cs.CL, cs.CR, cs.LG | Although large language models (LLMs) are widely deployed, the data used to
train them is rarely disclosed. Given the incredible scale of this data, up to
trillions of tokens, it is all but certain that it includes potentially
problematic text such as copyrighted materials, personally identifiable
information, and test data for widely reported reference benchmarks. However,
we currently have no way to know which data of these types is included or in
what proportions. In this paper, we study the pretraining data detection
problem: given a piece of text and black-box access to an LLM without knowing
the pretraining data, can we determine if the model was trained on the provided
text? To facilitate this study, we introduce a dynamic benchmark WIKIMIA that
uses data created before and after model training to support gold truth
detection. We also introduce a new detection method Min-K% Prob based on a
simple hypothesis: an unseen example is likely to contain a few outlier words
with low probabilities under the LLM, while a seen example is less likely to
have words with such low probabilities. Min-K% Prob can be applied without any
knowledge about the pretraining corpus or any additional training, departing
from previous detection methods that require training a reference model on data
that is similar to the pretraining data. Moreover, our experiments demonstrate
that Min-K% Prob achieves a 7.4% improvement on WIKIMIA over these previous
methods. We apply Min-K% Prob to three real-world scenarios, copyrighted book
detection, contaminated downstream example detection and privacy auditing of
machine unlearning, and find it a consistently effective solution. | cs.CL | Detecting Pretraining Data from Large Language Models |
2023-10-25T17:20:26Z | http://arxiv.org/pdf/2310.16787v3 | http://arxiv.org/abs/2310.16787v3 | Shayne Longpre, Robert Mahari, Anthony Chen, Naana Obeng-Marnu, Damien Sileo, William Brannon, Niklas Muennighoff, Nathan Khazam, Jad Kabbara, Kartik Perisetla, Xinyi Wu, Enrico Shippole, Kurt Bollacker, Tongshuang Wu, Luis Villa, Sandy Pentland, Sara Hooker | 2023-11-04T19:10:06Z | cs.CL, cs.AI, cs.LG | The race to train language models on vast, diverse, and inconsistently
documented datasets has raised pressing concerns about the legal and ethical
risks for practitioners. To remedy these practices threatening data
transparency and understanding, we convene a multi-disciplinary effort between
legal and machine learning experts to systematically audit and trace 1800+ text
datasets. We develop tools and standards to trace the lineage of these
datasets, from their source, creators, series of license conditions,
properties, and subsequent use. Our landscape analysis highlights the sharp
divides in composition and focus of commercially open vs closed datasets, with
closed datasets monopolizing important categories: lower resource languages,
more creative tasks, richer topic variety, newer and more synthetic training
data. This points to a deepening divide in the types of data that are made
available under different license conditions, and heightened implications for
jurisdictional legal interpretations of copyright and fair use. We also observe
frequent miscategorization of licenses on widely used dataset hosting sites,
with license omission of 70%+ and error rates of 50%+. This points to a crisis
in misattribution and informed use of the most popular datasets driving many
recent breakthroughs. As a contribution to ongoing improvements in dataset
transparency and responsible use, we release our entire audit, with an
interactive UI, the Data Provenance Explorer, which allows practitioners to
trace and filter on data provenance for the most popular open source finetuning
data collections: www.dataprovenance.org. | cs.CL | The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI |
2023-10-25T17:15:55Z | http://arxiv.org/pdf/2310.16781v3 | http://arxiv.org/abs/2310.16781v3 | Morris Alper, Hadar Averbuch-Elor | 2024-04-02T05:50:21Z | cs.CV, cs.CL, cs.LG | Although the mapping between sound and meaning in human language is assumed
to be largely arbitrary, research in cognitive science has shown that there are
non-trivial correlations between particular sounds and meanings across
languages and demographic groups, a phenomenon known as sound symbolism. Among
the many dimensions of meaning, sound symbolism is particularly salient and
well-demonstrated with regards to cross-modal associations between language and
the visual domain. In this work, we address the question of whether sound
symbolism is reflected in vision-and-language models such as CLIP and Stable
Diffusion. Using zero-shot knowledge probing to investigate the inherent
knowledge of these models, we find strong evidence that they do show this
pattern, paralleling the well-known kiki-bouba effect in psycholinguistics. Our
work provides a novel method for demonstrating sound symbolism and
understanding its nature using computational tools. Our code will be made
publicly available. | cs.CV | Kiki or Bouba? Sound Symbolism in Vision-and-Language Models |
2023-10-25T17:06:42Z | http://arxiv.org/pdf/2310.16776v5 | http://arxiv.org/abs/2310.16776v5 | Devleena Das, Vivek Khetan | 2024-06-13T02:31:28Z | cs.CL, cs.AI | Recent advances have led to the availability of many pre-trained language
models (PLMs); however, a question that remains is how much data is truly
needed to fine-tune PLMs for downstream tasks? In this work, we introduce
DEFT-UCS, a data-efficient fine-tuning framework that leverages unsupervised
core-set selection to identify a smaller, representative dataset that reduces
the amount of data needed to fine-tune PLMs for downstream tasks. We examine
the efficacy of DEFT-UCS in the context of text-editing LMs, and compare to the
state-of-the art text-editing model, CoEDIT. Our results demonstrate that
DEFT-UCS models are just as accurate as CoEDIT, across eight different datasets
consisting of six different editing tasks, while finetuned on 70% less data. | cs.CL | DEFT: Data Efficient Fine-Tuning for Pre-Trained Language Models via Unsupervised Core-Set Selection |
2023-10-25T16:52:00Z | http://arxiv.org/pdf/2310.16763v1 | http://arxiv.org/abs/2310.16763v1 | Gabriel Mukobi, Peter Chatain, Su Fong, Robert Windesheim, Gitta Kutyniok, Kush Bhatia, Silas Alberti | 2023-10-25T16:52:00Z | cs.CL, cs.AI, cs.LG | While large language models demonstrate remarkable capabilities, they often
present challenges in terms of safety, alignment with human values, and
stability during training. Here, we focus on two prevalent methods used to
align these models, Supervised Fine-Tuning (SFT) and Reinforcement Learning
from Human Feedback (RLHF). SFT is simple and robust, powering a host of
open-source models, while RLHF is a more sophisticated method used in top-tier
models like ChatGPT but also suffers from instability and susceptibility to
reward hacking. We propose a novel approach, Supervised Iterative Learning from
Human Feedback (SuperHF), which seeks to leverage the strengths of both
methods. Our hypothesis is two-fold: that the reward model used in RLHF is
critical for efficient data use and model generalization and that the use of
Proximal Policy Optimization (PPO) in RLHF may not be necessary and could
contribute to instability issues. SuperHF replaces PPO with a simple supervised
loss and a Kullback-Leibler (KL) divergence prior. It creates its own training
data by repeatedly sampling a batch of model outputs and filtering them through
the reward model in an online learning regime. We then break down the reward
optimization problem into three components: robustly optimizing the training
rewards themselves, preventing reward hacking-exploitation of the reward model
that degrades model performance-as measured by a novel METEOR similarity
metric, and maintaining good performance on downstream evaluations. Our
experimental results show SuperHF exceeds PPO-based RLHF on the training
objective, easily and favorably trades off high reward with low reward hacking,
improves downstream calibration, and performs the same on our GPT-4 based
qualitative evaluation scheme all the while being significantly simpler to
implement, highlighting SuperHF's potential as a competitive language model
alignment technique. | cs.CL | SuperHF: Supervised Iterative Learning from Human Feedback |
2023-10-25T16:50:24Z | http://arxiv.org/pdf/2310.16761v1 | http://arxiv.org/abs/2310.16761v1 | Bhavuk Singhal, Ashim Gupta, Shivasankaran V P, Amrith Krishna | 2023-10-25T16:50:24Z | cs.CL | Identifying intents from dialogue utterances forms an integral component of
task-oriented dialogue systems. Intent-related tasks are typically formulated
either as a classification task, where the utterances are classified into
predefined categories or as a clustering task when new and previously unknown
intent categories need to be discovered from these utterances. Further, the
intent classification may be modeled in a multiclass (MC) or multilabel (ML)
setup. While typically these tasks are modeled as separate tasks, we propose
IntenDD, a unified approach leveraging a shared utterance encoding backbone.
IntenDD uses an entirely unsupervised contrastive learning strategy for
representation learning, where pseudo-labels for the unlabeled utterances are
generated based on their lexical features. Additionally, we introduce a
two-step post-processing setup for the classification tasks using modified
adsorption. Here, first, the residuals in the training data are propagated
followed by smoothing the labels both modeled in a transductive setting.
Through extensive evaluations on various benchmark datasets, we find that our
approach consistently outperforms competitive baselines across all three tasks.
On average, IntenDD reports percentage improvements of 2.32%, 1.26%, and 1.52%
in their respective metrics for few-shot MC, few-shot ML, and the intent
discovery tasks respectively. | cs.CL | IntenDD: A Unified Contrastive Learning Approach for Intent Detection and Discovery |
2023-10-25T16:41:15Z | http://arxiv.org/pdf/2310.16755v1 | http://arxiv.org/abs/2310.16755v1 | Yinghui He, Yufan Wu, Yilin Jia, Rada Mihalcea, Yulong Chen, Naihao Deng | 2023-10-25T16:41:15Z | cs.CL, cs.AI | Theory of Mind (ToM) is the ability to reason about one's own and others'
mental states. ToM plays a critical role in the development of intelligence,
language understanding, and cognitive processes. While previous work has
primarily focused on first and second-order ToM, we explore higher-order ToM,
which involves recursive reasoning on others' beliefs. We introduce HI-TOM, a
Higher Order Theory of Mind benchmark. Our experimental evaluation using
various Large Language Models (LLMs) indicates a decline in performance on
higher-order ToM tasks, demonstrating the limitations of current LLMs. We
conduct a thorough analysis of different failure cases of LLMs, and share our
thoughts on the implications of our findings on the future of NLP. | cs.CL | HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models |
2023-10-25T16:39:00Z | http://arxiv.org/pdf/2310.16753v1 | http://arxiv.org/abs/2310.16753v1 | Yuqing Wang, Prashanth Vijayaraghavan, Ehsan Degan | 2023-10-25T16:39:00Z | cs.CL, cs.AI, cs.LG | Email is a widely used tool for business communication, and email marketing
has emerged as a cost-effective strategy for enterprises. While previous
studies have examined factors affecting email marketing performance, limited
research has focused on understanding email response behavior by considering
email content and metadata. This study proposes a Prototype-based Multi-view
Network (PROMINET) that incorporates semantic and structural information from
email data. By utilizing prototype learning, the PROMINET model generates
latent exemplars, enabling interpretable email response prediction. The model
maps learned semantic and structural exemplars to observed samples in the
training data at different levels of granularity, such as document, sentence,
or phrase. The approach is evaluated on two real-world email datasets: the
Enron corpus and an in-house Email Marketing corpus. Experimental results
demonstrate that the PROMINET model outperforms baseline models, achieving a
~3% improvement in F1 score on both datasets. Additionally, the model provides
interpretability through prototypes at different granularity levels while
maintaining comparable performance to non-interpretable models. The learned
prototypes also show potential for generating suggestions to enhance email text
editing and improve the likelihood of effective email responses. This research
contributes to enhancing sender-receiver communication and customer engagement
in email interactions. | cs.CL | PROMINET: Prototype-based Multi-View Network for Interpretable Email Response Prediction |
2023-10-25T16:32:02Z | http://arxiv.org/pdf/2310.16749v1 | http://arxiv.org/abs/2310.16749v1 | Vineet Bhat, Preethi Jyothi, Pushpak Bhattacharyya | 2023-10-25T16:32:02Z | cs.CL, cs.HC | Disfluency correction (DC) is the process of removing disfluent elements like
fillers, repetitions and corrections from spoken utterances to create readable
and interpretable text. DC is a vital post-processing step applied to Automatic
Speech Recognition (ASR) outputs, before subsequent processing by downstream
language understanding tasks. Existing DC research has primarily focused on
English due to the unavailability of large-scale open-source datasets. Towards
the goal of multilingual disfluency correction, we present a high-quality
human-annotated DC corpus covering four important Indo-European languages:
English, Hindi, German and French. We provide extensive analysis of results of
state-of-the-art DC models across all four languages obtaining F1 scores of
97.55 (English), 94.29 (Hindi), 95.89 (German) and 92.97 (French). To
demonstrate the benefits of DC on downstream tasks, we show that DC leads to
5.65 points increase in BLEU scores on average when used in conjunction with a
state-of-the-art Machine Translation (MT) system. We release code to run our
experiments along with our annotated dataset here. | cs.CL | DISCO: A Large Scale Human Annotated Corpus for Disfluency Correction in Indo-European Languages |
2023-10-25T16:23:17Z | http://arxiv.org/pdf/2310.16746v1 | http://arxiv.org/abs/2310.16746v1 | Nafis Irtiza Tripto, Adaku Uchendu, Thai Le, Mattia Setzu, Fosca Giannotti, Dongwon Lee | 2023-10-25T16:23:17Z | cs.CL | Authorship Analysis, also known as stylometry, has been an essential aspect
of Natural Language Processing (NLP) for a long time. Likewise, the recent
advancement of Large Language Models (LLMs) has made authorship analysis
increasingly crucial for distinguishing between human-written and AI-generated
texts. However, these authorship analysis tasks have primarily been focused on
written texts, not considering spoken texts. Thus, we introduce the largest
benchmark for spoken texts - HANSEN (Human ANd ai Spoken tExt beNchmark).
HANSEN encompasses meticulous curation of existing speech datasets accompanied
by transcripts, alongside the creation of novel AI-generated spoken text
datasets. Together, it comprises 17 human datasets, and AI-generated spoken
texts created using 3 prominent LLMs: ChatGPT, PaLM2, and Vicuna13B. To
evaluate and demonstrate the utility of HANSEN, we perform Authorship
Attribution (AA) & Author Verification (AV) on human-spoken datasets and
conducted Human vs. AI spoken text detection using state-of-the-art (SOTA)
models. While SOTA methods, such as, character ngram or Transformer-based
model, exhibit similar AA & AV performance in human-spoken datasets compared to
written ones, there is much room for improvement in AI-generated spoken text
detection. The HANSEN benchmark is available at:
https://huggingface.co/datasets/HANSEN-REPO/HANSEN. | cs.CL | HANSEN: Human and AI Spoken Text Benchmark for Authorship Analysis |
2023-10-25T16:11:55Z | http://arxiv.org/pdf/2310.16738v1 | http://arxiv.org/abs/2310.16738v1 | Xi Wang, Hossein A. Rahmani, Jiqun Liu, Emine Yilmaz | 2023-10-25T16:11:55Z | cs.CL, cs.IR | Conversational Recommendation System (CRS) is a rapidly growing research area
that has gained significant attention alongside advancements in language
modelling techniques. However, the current state of conversational
recommendation faces numerous challenges due to its relative novelty and
limited existing contributions. In this study, we delve into benchmark datasets
for developing CRS models and address potential biases arising from the
feedback loop inherent in multi-turn interactions, including selection bias and
multiple popularity bias variants. Drawing inspiration from the success of
generative data via using language models and data augmentation techniques, we
present two novel strategies, 'Once-Aug' and 'PopNudge', to enhance model
performance while mitigating biases. Through extensive experiments on ReDial
and TG-ReDial benchmark datasets, we show a consistent improvement of CRS
techniques with our data augmentation approaches and offer additional insights
on addressing multiple newly formulated biases. | cs.CL | Improving Conversational Recommendation Systems via Bias Analysis and Language-Model-Enhanced Data Augmentation |
2023-10-25T16:00:47Z | http://arxiv.org/pdf/2310.16731v1 | http://arxiv.org/abs/2310.16731v1 | Roshanak Mirzaee, Parisa Kordjamshidi | 2023-10-25T16:00:47Z | cs.CL | Spatial reasoning over text is challenging as the models not only need to
extract the direct spatial information from the text but also reason over those
and infer implicit spatial relations. Recent studies highlight the struggles
even large language models encounter when it comes to performing spatial
reasoning over text. In this paper, we explore the potential benefits of
disentangling the processes of information extraction and reasoning in models
to address this challenge. To explore this, we design various models that
disentangle extraction and reasoning(either symbolic or neural) and compare
them with state-of-the-art(SOTA) baselines with no explicit design for these
parts. Our experimental results consistently demonstrate the efficacy of
disentangling, showcasing its ability to enhance models' generalizability
within realistic data domains. | cs.CL | Disentangling Extraction and Reasoning in Multi-hop Spatial Reasoning |
2023-10-25T15:34:55Z | http://arxiv.org/pdf/2310.16713v2 | http://arxiv.org/abs/2310.16713v2 | Liu Yang, Haihua Yang, Wenjun Cheng, Lei Lin, Chenxia Li, Yifu Chen, Lunan Liu, Jianfei Pan, Tianwen Wei, Biye Li, Liang Zhao, Lijie Wang, Bo Zhu, Guoliang Li, Xuejie Wu, Xilin Luo, Rui Hu | 2023-10-26T09:01:16Z | cs.CL, cs.AI | Large language models (LLMs) have shown great potential to solve varieties of
natural language processing (NLP) tasks, including mathematical reasoning. In
this work, we present SkyMath, a large language model for mathematics with 13
billion parameters. By applying self-compare fine-tuning, we have enhanced
mathematical reasoning abilities of Skywork-13B-Base remarkably. On GSM8K,
SkyMath outperforms all known open-source models of similar size and has
established a new SOTA performance. | cs.CL | SkyMath: Technical Report |
2023-10-25T15:34:30Z | http://arxiv.org/pdf/2310.16712v1 | http://arxiv.org/abs/2310.16712v1 | Ganesh Jawahar, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Dujian Ding | 2023-10-25T15:34:30Z | cs.CL | Large language models (LLMs) have become an integral component in solving a
wide range of NLP tasks. In this work, we explore a novel use case of using
LLMs to build performance predictors (PP): models that, given a specific deep
neural network architecture, predict its performance on a downstream task. We
design PP prompts for LLMs consisting of: (i) role: description of the role
assigned to the LLM, (ii) instructions: set of instructions to be followed by
the LLM to carry out performance prediction, (iii) hyperparameters: a
definition of each architecture-specific hyperparameter and (iv)
demonstrations: sample architectures along with their efficiency metrics and
'training from scratch' performance. For machine translation (MT) tasks, we
discover that GPT-4 with our PP prompts (LLM-PP) can predict the performance of
architecture with a mean absolute error matching the SOTA and a marginal
degradation in rank correlation coefficient compared to SOTA performance
predictors. Further, we show that the predictions from LLM-PP can be distilled
to a small regression model (LLM-Distill-PP). LLM-Distill-PP models
surprisingly retain the performance of LLM-PP largely and can be a
cost-effective alternative for heavy use cases of performance estimation.
Specifically, for neural architecture search (NAS), we propose a Hybrid-Search
algorithm for NAS (HS-NAS), which uses LLM-Distill-PP for the initial part of
search, resorting to the baseline predictor for rest of the search. We show
that HS-NAS performs very similar to SOTA NAS across benchmarks, reduces search
hours by 50% roughly, and in some cases, improves latency, GFLOPs, and model
size. | cs.CL | LLM Performance Predictors are good initializers for Architecture Search |
2023-10-25T14:48:58Z | http://arxiv.org/pdf/2310.16685v1 | http://arxiv.org/abs/2310.16685v1 | Amanda Ferrari Iaquinta, Gustavo Voltani von Atzingen | 2023-10-25T14:48:58Z | cs.CL, cs.AI | The large language based-model chatbot ChatGPT gained a lot of popularity
since its launch and has been used in a wide range of situations. This research
centers around a particular situation, when the ChatGPT is used to produce news
that will be consumed by the population, causing the facilitation in the
production of fake news, spread of misinformation and lack of trust in news
sources. Aware of these problems, this research aims to build an artificial
intelligence model capable of performing authorship attribution on news
articles, identifying the ones written by the ChatGPT. To achieve this goal, a
dataset containing equal amounts of human and ChatGPT written news was
assembled and different natural processing language techniques were used to
extract features from it that were used to train, validate and test three
models built with different techniques. The best performance was produced by
the Bidirectional Long Short Term Memory (LSTM) Neural Network model, achiving
91.57\% accuracy when tested against the data from the testing set. | cs.CL | Detection of news written by the ChatGPT through authorship attribution performed by a Bidirectional LSTM model |
2023-10-25T14:45:48Z | http://arxiv.org/pdf/2310.16681v1 | http://arxiv.org/abs/2310.16681v1 | Xingmeng Zhao, Tongnian Wang, Sheri Osborn, Anthony Rios | 2023-10-25T14:45:48Z | cs.CL | Language models have seen significant growth in the size of their corpus,
leading to notable performance improvements. Yet, there has been limited
progress in developing models that handle smaller, more human-like datasets. As
part of the BabyLM shared task, this study explores the impact of reinforcement
learning from human feedback (RLHF) on language models pretrained from scratch
with a limited training corpus. Comparing two GPT-2 variants, the larger model
performs better in storytelling tasks after RLHF fine-tuning. These findings
suggest that RLHF techniques may be more advantageous for larger models due to
their higher learning and adaptation capacity, though more experiments are
needed to confirm this finding. These insights highlight the potential benefits
of RLHF fine-tuning for language models within limited data, enhancing their
ability to maintain narrative focus and coherence while adhering better to
initial instructions in storytelling tasks. The code for this work is publicly
at https://github.com/Zephyr1022/BabyStories-UTSA. | cs.CL | BabyStories: Can Reinforcement Learning Teach Baby Language Models to Write Better Stories? |
2023-10-25T14:41:14Z | http://arxiv.org/pdf/2310.16676v3 | http://arxiv.org/abs/2310.16676v3 | Tao Shi, Xiao Liang, Yaoyuan Liang, Xinyi Tong, Shao-Lun Huang | 2023-12-11T04:51:49Z | cs.CL | Emotion recognition in conversations (ERC) is a rapidly evolving task within
the natural language processing community, which aims to detect the emotions
expressed by speakers during a conversation. Recently, a growing number of ERC
methods have focused on leveraging supervised contrastive learning (SCL) to
enhance the robustness and generalizability of learned features. However,
current SCL-based approaches in ERC are impeded by the constraint of large
batch sizes and the lack of compatibility with most existing ERC models. To
address these challenges, we propose an efficient and model-agnostic SCL
framework named Supervised Sample-Label Contrastive Learning with Soft-HGR
Maximal Correlation (SSLCL), which eliminates the need for a large batch size
and can be seamlessly integrated with existing ERC models without introducing
any model-specific assumptions. Specifically, we introduce a novel perspective
on utilizing label representations by projecting discrete labels into dense
embeddings through a shallow multilayer perceptron, and formulate the training
objective to maximize the similarity between sample features and their
corresponding ground-truth label embeddings, while minimizing the similarity
between sample features and label embeddings of disparate classes. Moreover, we
innovatively adopt the Soft-HGR maximal correlation as a measure of similarity
between sample features and label embeddings, leading to significant
performance improvements over conventional similarity measures. Additionally,
multimodal cues of utterances are effectively leveraged by SSLCL as data
augmentations to boost model performances. Extensive experiments on two ERC
benchmark datasets, IEMOCAP and MELD, demonstrate the compatibility and
superiority of our proposed SSLCL framework compared to existing
state-of-the-art SCL methods. Our code is available at
\url{https://github.com/TaoShi1998/SSLCL}. | cs.CL | SSLCL: An Efficient Model-Agnostic Supervised Contrastive Learning Framework for Emotion Recognition in Conversations |
2023-10-25T14:08:39Z | http://arxiv.org/pdf/2310.16654v1 | http://arxiv.org/abs/2310.16654v1 | Boda Lin, Xinyi Zhou, Binghao Tang, Xiaocheng Gong, Si Li | 2023-10-25T14:08:39Z | cs.CL | Pre-trained language models have been widely used in dependency parsing task
and have achieved significant improvements in parser performance. However, it
remains an understudied question whether pre-trained language models can
spontaneously exhibit the ability of dependency parsing without introducing
additional parser structure in the zero-shot scenario. In this paper, we
propose to explore the dependency parsing ability of large language models such
as ChatGPT and conduct linguistic analysis. The experimental results
demonstrate that ChatGPT is a potential zero-shot dependency parser, and the
linguistic analysis also shows some unique preferences in parsing outputs. | cs.CL | ChatGPT is a Potential Zero-Shot Dependency Parser |
2023-10-25T13:20:54Z | http://arxiv.org/pdf/2310.16621v1 | http://arxiv.org/abs/2310.16621v1 | Hawau Olamide Toyin, Amirbek Djanibekov, Ajinkya Kulkarni, Hanan Aldarmaki | 2023-10-25T13:20:54Z | cs.CL, cs.AI, cs.SD, eess.AS | We present ArTST, a pre-trained Arabic text and speech transformer for
supporting open-source speech technologies for the Arabic language. The model
architecture follows the unified-modal framework, SpeechT5, that was recently
released for English, and is focused on Modern Standard Arabic (MSA), with
plans to extend the model for dialectal and code-switched Arabic in future
editions. We pre-trained the model from scratch on MSA speech and text data,
and fine-tuned it for the following tasks: Automatic Speech Recognition (ASR),
Text-To-Speech synthesis (TTS), and spoken dialect identification. In our
experiments comparing ArTST with SpeechT5, as well as with previously reported
results in these tasks, ArTST performs on a par with or exceeding the current
state-of-the-art in all three tasks. Moreover, we find that our pre-training is
conducive for generalization, which is particularly evident in the low-resource
TTS task. The pre-trained model as well as the fine-tuned ASR and TTS models
are released for research use. | cs.CL | ArTST: Arabic Text and Speech Transformer |
2023-10-25T13:12:39Z | http://arxiv.org/pdf/2310.16616v1 | http://arxiv.org/abs/2310.16616v1 | Yiming Lin, Xiao-Bo Jin, Qiufeng Wang, Kaizhu Huang | 2023-10-25T13:12:39Z | cs.CV, cs.CL | Panoramic Narrative Grounding (PNG) is an emerging visual grounding task that
aims to segment visual objects in images based on dense narrative captions. The
current state-of-the-art methods first refine the representation of phrase by
aggregating the most similar $k$ image pixels, and then match the refined text
representations with the pixels of the image feature map to generate
segmentation results. However, simply aggregating sampled image features
ignores the contextual information, which can lead to phrase-to-pixel
mis-match. In this paper, we propose a novel learning framework called
Deformable Attention Refined Matching Network (DRMN), whose main idea is to
bring deformable attention in the iterative process of feature learning to
incorporate essential context information of different scales of pixels. DRMN
iteratively re-encodes pixels with the deformable attention network after
updating the feature representation of the top-$k$ most similar pixels. As
such, DRMN can lead to accurate yet discriminative pixel representations,
purify the top-$k$ most similar pixels, and consequently alleviate the
phrase-to-pixel mis-match substantially.Experimental results show that our
novel design significantly improves the matching results between text phrases
and image pixels. Concretely, DRMN achieves new state-of-the-art performance on
the PNG benchmark with an average recall improvement 3.5%. The codes are
available in: https://github.com/JaMesLiMers/DRMN. | cs.CV | Context Does Matter: End-to-end Panoptic Narrative Grounding with Deformable Attention Refined Matching Network |
2023-10-25T13:07:07Z | http://arxiv.org/pdf/2310.16609v1 | http://arxiv.org/abs/2310.16609v1 | Marek Kubis, Paweł Skórzewski, Marcin Sowański, Tomasz Ziętkiewicz | 2023-10-25T13:07:07Z | cs.CL, cs.AI, cs.SD, eess.AS | In a spoken dialogue system, an NLU model is preceded by a speech recognition
system that can deteriorate the performance of natural language understanding.
This paper proposes a method for investigating the impact of speech recognition
errors on the performance of natural language understanding models. The
proposed method combines the back transcription procedure with a fine-grained
technique for categorizing the errors that affect the performance of NLU
models. The method relies on the usage of synthesized speech for NLU
evaluation. We show that the use of synthesized speech in place of audio
recording does not change the outcomes of the presented technique in a
significant way. | cs.CL | Back Transcription as a Method for Evaluating Robustness of Natural Language Understanding Models to Speech Recognition Errors |
2023-10-25T12:59:51Z | http://arxiv.org/pdf/2310.16607v2 | http://arxiv.org/abs/2310.16607v2 | Stephanie Brandl, Emanuele Bugliarello, Ilias Chalkidis | 2023-11-13T15:20:43Z | cs.CL | In order to build reliable and trustworthy NLP applications, models need to
be both fair across different demographics and explainable. Usually these two
objectives, fairness and explainability, are optimized and/or examined
independently of each other. Instead, we argue that forthcoming, trustworthy
NLP systems should consider both. In this work, we perform a first study to
understand how they influence each other: do fair(er) models rely on more
plausible rationales? and vice versa. To this end, we conduct experiments on
two English multi-class text classification datasets, BIOS and ECtHR, that
provide information on gender and nationality, respectively, as well as
human-annotated rationales. We fine-tune pre-trained language models with
several methods for (i) bias mitigation, which aims to improve fairness; (ii)
rationale extraction, which aims to produce plausible explanations. We find
that bias mitigation algorithms do not always lead to fairer models. Moreover,
we discover that empirical fairness and explainability are orthogonal. | cs.CL | On the Interplay between Fairness and Explainability |
2023-10-25T12:16:33Z | http://arxiv.org/pdf/2310.16582v2 | http://arxiv.org/abs/2310.16582v2 | Tianlong Li, Shihan Dou, Changze Lv, Wenhao Liu, Jianhan Xu, Muling Wu, Zixuan Ling, Xiaoqing Zheng, Xuanjing Huang | 2024-01-06T14:17:40Z | cs.CL | Personality plays a pivotal role in shaping human expression patterns, thus
regulating the personality of large language models (LLMs) holds significant
potential in enhancing the user experience of LLMs. Previous methods either
relied on fine-tuning LLMs on specific corpora or necessitated manually crafted
prompts to elicit specific personalities from LLMs. However, the former
approach is inefficient and costly, while the latter cannot precisely
manipulate personality traits at a fine-grained level. To address the above
challenges, we have employed a novel Unsupervisedly-Built Personalized Lexicons
(UBPL) in a pluggable manner during the decoding phase of LLMs to manipulate
their personality traits. UBPL is a lexicon built through an unsupervised
approach from a situational judgment test dataset (SJTs4LLM). Users can utilize
UBPL to adjust the probability vectors of predicted words in the decoding phase
of LLMs, thus influencing the personality expression of LLMs. Extensive
experimentation demonstrates the remarkable effectiveness and pluggability of
our method for fine-grained manipulation of LLM's personality. | cs.CL | Tailoring Personality Traits in Large Language Models via Unsupervisedly-Built Personalized Lexicons |
2023-10-25T12:06:55Z | http://arxiv.org/pdf/2310.16579v1 | http://arxiv.org/abs/2310.16579v1 | Ruichao Yang, Wei Gao, Jing Ma, Hongzhan Lin, Zhiwei Yang | 2023-10-25T12:06:55Z | cs.CL | In recent years, we witness the explosion of false and unconfirmed
information (i.e., rumors) that went viral on social media and shocked the
public. Rumors can trigger versatile, mostly controversial stance expressions
among social media users. Rumor verification and stance detection are different
yet relevant tasks. Fake news debunking primarily focuses on determining the
truthfulness of news articles, which oversimplifies the issue as fake news
often combines elements of both truth and falsehood. Thus, it becomes crucial
to identify specific instances of misinformation within the articles. In this
research, we investigate a novel task in the field of fake news debunking,
which involves detecting sentence-level misinformation. One of the major
challenges in this task is the absence of a training dataset with
sentence-level annotations regarding veracity. Inspired by the Multiple
Instance Learning (MIL) approach, we propose a model called Weakly Supervised
Detection of Misinforming Sentences (WSDMS). This model only requires bag-level
labels for training but is capable of inferring both sentence-level
misinformation and article-level veracity, aided by relevant social media
conversations that are attentively contextualized with news sentences. We
evaluate WSDMS on three real-world benchmarks and demonstrate that it
outperforms existing state-of-the-art baselines in debunking fake news at both
the sentence and article levels. | cs.CL | WSDMS: Debunk Fake News via Weakly Supervised Detection of Misinforming Sentences with Contextualized Social Wisdom |
2023-10-25T11:57:13Z | http://arxiv.org/pdf/2310.16570v2 | http://arxiv.org/abs/2310.16570v2 | Paul Youssef, Osman Alperen Koraş, Meijie Li, Jörg Schlötterer, Christin Seifert | 2023-12-04T19:23:33Z | cs.CL | Pre-trained Language Models (PLMs) are trained on vast unlabeled data, rich
in world knowledge. This fact has sparked the interest of the community in
quantifying the amount of factual knowledge present in PLMs, as this explains
their performance on downstream tasks, and potentially justifies their use as
knowledge bases. In this work, we survey methods and datasets that are used to
probe PLMs for factual knowledge. Our contributions are: (1) We propose a
categorization scheme for factual probing methods that is based on how their
inputs, outputs and the probed PLMs are adapted; (2) We provide an overview of
the datasets used for factual probing; (3) We synthesize insights about
knowledge retention and prompt optimization in PLMs, analyze obstacles to
adopting PLMs as knowledge bases and outline directions for future work. | cs.CL | Give Me the Facts! A Survey on Factual Knowledge Probing in Pre-trained Language Models |
2023-10-25T11:51:22Z | http://arxiv.org/pdf/2310.16568v1 | http://arxiv.org/abs/2310.16568v1 | Palak Jain, Livio Baldini Soares, Tom Kwiatkowski | 2023-10-25T11:51:22Z | cs.CL | We present 1-Pager the first system that answers a question and retrieves
evidence using a single Transformer-based model and decoding process. 1-Pager
incrementally partitions the retrieval corpus using constrained decoding to
select a document and answer string, and we show that this is competitive with
comparable retrieve-and-read alternatives according to both retrieval and
answer accuracy metrics. 1-Pager also outperforms the equivalent closed-book
question answering model, by grounding predictions in an evidence corpus. While
1-Pager is not yet on-par with more expensive systems that read many more
documents before generating an answer, we argue that it provides an important
step toward attributed generation by folding retrieval into the
sequence-to-sequence paradigm that is currently dominant in NLP. We also show
that the search paths used to partition the corpus are easy to read and
understand, paving a way forward for interpretable neural retrieval. | cs.CL | 1-PAGER: One Pass Answer Generation and Evidence Retrieval |
2023-10-25T10:50:18Z | http://arxiv.org/pdf/2310.18366v1 | http://arxiv.org/abs/2310.18366v1 | Alicia Jiayun Law, Ruoyu Hu, Lisa Alazraki, Anandha Gopalan, Neophytos Polydorou, Abbas Edalat | 2023-10-25T10:50:18Z | cs.CL, cs.AI, cs.LG | In this work, we propose a computational framework that leverages existing
out-of-language data to create a conversational agent for the delivery of
Self-Attachment Technique (SAT) in Mandarin. Our framework does not require
large-scale human translations, yet it achieves a comparable performance whilst
also maintaining safety and reliability. We propose two different methods of
augmenting available response data through empathetic rewriting. We evaluate
our chatbot against a previous, English-only SAT chatbot through non-clinical
human trials (N=42), each lasting five days, and quantitatively show that we
are able to attain a comparable level of performance to the English SAT
chatbot. We provide qualitative analysis on the limitations of our study and
suggestions with the aim of guiding future improvements. | cs.CL | A Multilingual Virtual Guide for Self-Attachment Technique |
2023-10-25T10:35:09Z | http://arxiv.org/pdf/2310.16538v1 | http://arxiv.org/abs/2310.16538v1 | Jaemin Shin, Hyungjun Yoon, Seungjoo Lee, Sungjoon Park, Yunxin Liu, Jinho D. Choi, Sung-Ju Lee | 2023-10-25T10:35:09Z | cs.CL, cs.AI, cs.LG | Psychiatrists diagnose mental disorders via the linguistic use of patients.
Still, due to data privacy, existing passive mental health monitoring systems
use alternative features such as activity, app usage, and location via mobile
devices. We propose FedTherapist, a mobile mental health monitoring system that
utilizes continuous speech and keyboard input in a privacy-preserving way via
federated learning. We explore multiple model designs by comparing their
performance and overhead for FedTherapist to overcome the complex nature of
on-device language model training on smartphones. We further propose a
Context-Aware Language Learning (CALL) methodology to effectively utilize
smartphones' large and noisy text for mental health signal sensing. Our
IRB-approved evaluation of the prediction of self-reported depression, stress,
anxiety, and mood from 46 participants shows higher accuracy of FedTherapist
compared with the performance with non-language features, achieving 0.15 AUROC
improvement and 8.21% MAE reduction. | cs.CL | FedTherapist: Mental Health Monitoring with User-Generated Linguistic Expressions on Smartphones via Federated Learning |
2023-10-25T10:34:02Z | http://arxiv.org/pdf/2310.16535v1 | http://arxiv.org/abs/2310.16535v1 | Qingyuan Tian, Hanlun Zhu, Lei Wang, Yang Li, Yunshi Lan | 2023-10-25T10:34:02Z | cs.CL, cs.AI | With the help of Chain-of-Thought (CoT) prompting, Large Language Models
(LLMs) have achieved remarkable performance on various reasoning tasks.
However, most of them have been evaluated under noise-free context and the
dilemma for LLMs to produce inaccurate results under the noisy context has not
been fully investigated. Existing studies utilize trigger sentences to
encourage LLMs to concentrate on the relevant information but the trigger has
limited effect on final answer prediction. Inspired by interactive CoT method,
where intermediate reasoning steps are promoted by multiple rounds of
interaction between users and LLMs, we propose a novel prompting method, namely
R$^3$ prompting, for CoT reasoning under noisy context. Specifically, R$^3$
prompting interacts with LLMs to perform key sentence extraction, variable
declaration and answer prediction, which corresponds to a thought process of
reviewing, rephrasing and resolving. The responses generated at the last
interaction will perform as hints to guide toward the responses of the next
interaction. Our experiments show that R$^3$ prompting significantly
outperforms existing CoT prompting methods on five reasoning tasks under noisy
context. With GPT-3.5-turbo, we observe 3.7% accuracy improvement on average on
the reasoning tasks under noisy context compared to the most competitive
prompting baseline. More analyses and ablation studies show the robustness and
generalization of R$^3$ prompting method in solving reasoning tasks in LLMs
under noisy context. | cs.CL | R$^3$ Prompting: Review, Rephrase and Resolve for Chain-of-Thought Reasoning in Large Language Models under Noisy Context |
2023-10-25T10:33:17Z | http://arxiv.org/pdf/2310.16534v1 | http://arxiv.org/abs/2310.16534v1 | Yang Wu, Shilong Wang, Hao Yang, Tian Zheng, Hongbo Zhang, Yanyan Zhao, Bing Qin | 2023-10-25T10:33:17Z | cs.CL, cs.CV | In this paper, we evaluate different abilities of GPT-4V including visual
understanding, language understanding, visual puzzle solving, and understanding
of other modalities such as depth, thermal, video, and audio. To estimate
GPT-4V's performance, we manually construct 656 test instances and carefully
evaluate the results of GPT-4V. The highlights of our findings are as follows:
(1) GPT-4V exhibits impressive performance on English visual-centric benchmarks
but fails to recognize simple Chinese texts in the images; (2) GPT-4V shows
inconsistent refusal behavior when answering questions related to sensitive
traits such as gender, race, and age; (3) GPT-4V obtains worse results than
GPT-4 (API) on language understanding tasks including general language
understanding benchmarks and visual commonsense knowledge evaluation
benchmarks; (4) Few-shot prompting can improve GPT-4V's performance on both
visual understanding and language understanding; (5) GPT-4V struggles to find
the nuances between two similar images and solve the easy math picture puzzles;
(6) GPT-4V shows non-trivial performance on the tasks of similar modalities to
image, such as video and thermal. Our experimental results reveal the ability
and limitations of GPT-4V and we hope our paper can provide some insights into
the application and research of GPT-4V. | cs.CL | An Early Evaluation of GPT-4V(ision) |
2023-10-25T10:22:49Z | http://arxiv.org/pdf/2310.16528v1 | http://arxiv.org/abs/2310.16528v1 | Jindřich Helcl, Jindřich Libovický | 2023-10-25T10:22:49Z | cs.CL | We present the Charles University system for the MRL~2023 Shared Task on
Multi-lingual Multi-task Information Retrieval. The goal of the shared task was
to develop systems for named entity recognition and question answering in
several under-represented languages. Our solutions to both subtasks rely on the
translate-test approach. We first translate the unlabeled examples into English
using a multilingual machine translation model. Then, we run inference on the
translated data using a strong task-specific model. Finally, we project the
labeled data back into the original language. To keep the inferred tags on the
correct positions in the original language, we propose a method based on
scoring the candidate positions using a label-sensitive translation model. In
both settings, we experiment with finetuning the classification models on the
translated data. However, due to a domain mismatch between the development data
and the shared task validation and test sets, the finetuned models could not
outperform our baselines. | cs.CL | CUNI Submission to MRL 2023 Shared Task on Multi-lingual Multi-task Information Retrieval |
2023-10-25T10:17:17Z | http://arxiv.org/pdf/2310.16523v1 | http://arxiv.org/abs/2310.16523v1 | Preethi Lahoti, Nicholas Blumm, Xiao Ma, Raghavendra Kotikalapudi, Sahitya Potluri, Qijun Tan, Hansa Srinivasan, Ben Packer, Ahmad Beirami, Alex Beutel, Jilin Chen | 2023-10-25T10:17:17Z | cs.CL, cs.AI | A crucial challenge for generative large language models (LLMs) is diversity:
when a user's prompt is under-specified, models may follow implicit assumptions
while generating a response, which may result in homogenization of the
responses, as well as certain demographic groups being under-represented or
even erased from the generated responses. In this paper, we formalize diversity
of representation in generative LLMs. We present evaluation datasets and
propose metrics to measure diversity in generated responses along people and
culture axes. We find that LLMs understand the notion of diversity, and that
they can reason and critique their own responses for that goal. This finding
motivated a new prompting technique called collective-critique and self-voting
(CCSV) to self-improve people diversity of LLMs by tapping into its diversity
reasoning capabilities, without relying on handcrafted examples or prompt
tuning. Extensive empirical experiments with both human and automated
evaluations show that our proposed approach is effective at improving people
and culture diversity, and outperforms all baseline methods by a large margin. | cs.CL | Improving Diversity of Demographic Representation in Large Language Models via Collective-Critiques and Self-Voting |
2023-10-25T10:06:17Z | http://arxiv.org/pdf/2310.16517v1 | http://arxiv.org/abs/2310.16517v1 | Mingfeng Xue, Dayiheng Liu, Kexin Yang, Guanting Dong, Wenqiang Lei, Zheng Yuan, Chang Zhou, Jingren Zhou | 2023-10-25T10:06:17Z | cs.CL | The emergence of large language models (LLMs) has revolutionized natural
language processing tasks. However, existing instruction-tuning datasets suffer
from occupational bias: the majority of data relates to only a few occupations,
which hampers the instruction-tuned LLMs to generate helpful responses to
professional queries from practitioners in specific fields. To mitigate this
issue and promote occupation-inclusive LLMs, we create an instruction-tuning
dataset named \emph{OccuQuest}, which contains 110,000+ prompt-completion pairs
and 30,000+ dialogues covering over 1,000 occupations in 26 occupational
categories. We systematically request ChatGPT, organizing queries
hierarchically based on Occupation, Responsibility, Topic, and Question, to
ensure a comprehensive coverage of occupational specialty inquiries. By
comparing with three commonly used datasets (Dolly, ShareGPT, and WizardLM), we
observe that OccuQuest exhibits a more balanced distribution across
occupations. Furthermore, we assemble three test sets for comprehensive
evaluation, an occu-test set covering 25 occupational categories, an estate set
focusing on real estate, and an occu-quora set containing real-world questions
from Quora. We then fine-tune LLaMA on OccuQuest to obtain OccuLLaMA, which
significantly outperforms state-of-the-art LLaMA variants (Vicuna, Tulu, and
WizardLM) on professional questions in GPT-4 and human evaluations. Notably, on
the occu-quora set, OccuLLaMA reaches a high win rate of 86.4\% against
WizardLM. | cs.CL | OccuQuest: Mitigating Occupational Bias for Inclusive Large Language Models |
2023-10-25T09:09:55Z | http://arxiv.org/pdf/2310.16484v1 | http://arxiv.org/abs/2310.16484v1 | Max Müller-Eberstein, Rob van der Goot, Barbara Plank, Ivan Titov | 2023-10-25T09:09:55Z | cs.CL | Representational spaces learned via language modeling are fundamental to
Natural Language Processing (NLP), however there has been limited understanding
regarding how and when during training various types of linguistic information
emerge and interact. Leveraging a novel information theoretic probing suite,
which enables direct comparisons of not just task performance, but their
representational subspaces, we analyze nine tasks covering syntax, semantics
and reasoning, across 2M pre-training steps and five seeds. We identify
critical learning phases across tasks and time, during which subspaces emerge,
share information, and later disentangle to specialize. Across these phases,
syntactic knowledge is acquired rapidly after 0.5% of full training. Continued
performance improvements primarily stem from the acquisition of open-domain
knowledge, while semantics and reasoning tasks benefit from later boosts to
long-range contextualization and higher specialization. Measuring cross-task
similarity further reveals that linguistically related tasks share information
throughout training, and do so more during the critical phase of learning than
before or after. Our findings have implications for model interpretability,
multi-task learning, and learning from limited data. | cs.CL | Subspace Chronicles: How Linguistic Information Emerges, Shifts and Interacts during Language Model Training |
2023-10-25T08:13:02Z | http://arxiv.org/pdf/2310.16450v3 | http://arxiv.org/abs/2310.16450v3 | Guanzheng Chen, Xin Li, Zaiqiao Meng, Shangsong Liang, Lidong Bing | 2024-03-24T17:14:11Z | cs.CL | Transformer-based Large Language Models (LLMs) are pioneering advances in
many natural language processing tasks, however, their exceptional capabilities
are restricted within the preset context window of Transformer. Position
Embedding (PE) scaling methods, while effective in extending the context window
to a specific length, demonstrate either notable limitations in their
extrapolation abilities or sacrificing partial performance within the context
window. Length extrapolation methods, although theoretically capable of
extending the context window beyond the training sequence length, often
underperform in practical long-context applications. To address these
challenges, we propose Continuous Length EXtrapolation (CLEX) for LLMs. We
generalise the PE scaling approaches to model the continuous dynamics by
ordinary differential equations over the length scaling factor, thereby
overcoming the constraints of current PE scaling methods designed for specific
lengths. Moreover, by extending the dynamics to desired context lengths beyond
the training sequence length, CLEX facilitates the length extrapolation with
impressive performance in practical tasks. We demonstrate that CLEX can be
seamlessly incorporated into LLMs equipped with Rotary Position Embedding, such
as LLaMA and GPT-NeoX, with negligible impact on training and inference
latency. Experimental results reveal that CLEX can effectively extend the
context window to over 4x or almost 8x training length, with no deterioration
in performance. Furthermore, when evaluated on the practical LongBench
benchmark, our model trained on a 4k length exhibits competitive performance
against state-of-the-art open-source models trained on context lengths up to
32k. Our code is available at https://github.com/DAMO-NLP-SG/CLEX. | cs.CL | CLEX: Continuous Length Extrapolation for Large Language Models |
2023-10-25T08:10:04Z | http://arxiv.org/pdf/2310.16446v1 | http://arxiv.org/abs/2310.16446v1 | Hokeun Yoon, JinYeong Bak | 2023-10-25T08:10:04Z | cs.CL, cs.AI | Question generation (QG) from a given context can enhance comprehension,
engagement, assessment, and overall efficacy in learning or conversational
environments. Despite recent advancements in QG, the challenge of enhancing or
measuring the diversity of generated questions often remains unaddressed. In
this paper, we introduce a multi-question generation model (mQG), which is
capable of generating multiple, diverse, and answerable questions by focusing
on context and questions. To validate the answerability of the generated
questions, we employ a SQuAD2.0 fine-tuned question answering model,
classifying the questions as answerable or not. We train and evaluate mQG on
the FairytaleQA dataset, a well-structured QA dataset based on storybooks, with
narrative questions. We further apply a zero-shot adaptation on the TellMeWhy
and SQuAD1.1 datasets. mQG shows promising results across various evaluation
metrics, among strong baselines. | cs.CL | Diversity Enhanced Narrative Question Generation for Storybooks |
2023-10-25T08:03:10Z | http://arxiv.org/pdf/2310.16436v2 | http://arxiv.org/abs/2310.16436v2 | Ge Zheng, Bin Yang, Jiajin Tang, Hong-Yu Zhou, Sibei Yang | 2023-10-26T04:16:52Z | cs.CV, cs.CL | A long-standing goal of AI systems is to perform complex multimodal reasoning
like humans. Recently, large language models (LLMs) have made remarkable
strides in such multi-step reasoning on the language modality solely by
leveraging the chain of thought (CoT) to mimic human thinking. However, the
transfer of these advancements to multimodal contexts introduces heightened
challenges, including but not limited to the impractical need for
labor-intensive annotation and the limitations in terms of flexibility,
generalizability, and explainability. To evoke CoT reasoning in multimodality,
this work first conducts an in-depth analysis of these challenges posed by
multimodality and presents two key insights: "keeping critical thinking" and
"letting everyone do their jobs" in multimodal CoT reasoning. Furthermore, this
study proposes a novel DDCoT prompting that maintains a critical attitude
through negative-space prompting and incorporates multimodality into reasoning
by first dividing the reasoning responsibility of LLMs into reasoning and
recognition and then integrating the visual recognition capability of visual
models into the joint reasoning process. The rationales generated by DDCoT not
only improve the reasoning abilities of both large and small language models in
zero-shot prompting and fine-tuning learning, significantly outperforming
state-of-the-art methods but also exhibit impressive generalizability and
explainability. | cs.CV | DDCoT: Duty-Distinct Chain-of-Thought Prompting for Multimodal Reasoning in Language Models |
2023-10-25T07:47:01Z | http://arxiv.org/pdf/2310.16427v2 | http://arxiv.org/abs/2310.16427v2 | Xinyuan Wang, Chenxi Li, Zhen Wang, Fan Bai, Haotian Luo, Jiayou Zhang, Nebojsa Jojic, Eric P. Xing, Zhiting Hu | 2023-12-07T14:39:22Z | cs.CL | Highly effective, task-specific prompts are often heavily engineered by
experts to integrate detailed instructions and domain insights based on a deep
understanding of both instincts of large language models (LLMs) and the
intricacies of the target task. However, automating the generation of such
expert-level prompts remains elusive. Existing prompt optimization methods tend
to overlook the depth of domain knowledge and struggle to efficiently explore
the vast space of expert-level prompts. Addressing this, we present
PromptAgent, an optimization method that autonomously crafts prompts equivalent
in quality to those handcrafted by experts. At its core, PromptAgent views
prompt optimization as a strategic planning problem and employs a principled
planning algorithm, rooted in Monte Carlo tree search, to strategically
navigate the expert-level prompt space. Inspired by human-like trial-and-error
exploration, PromptAgent induces precise expert-level insights and in-depth
instructions by reflecting on model errors and generating constructive error
feedback. Such a novel framework allows the agent to iteratively examine
intermediate prompts (states), refine them based on error feedbacks (actions),
simulate future rewards, and search for high-reward paths leading to expert
prompts. We apply PromptAgent to 12 tasks spanning three practical domains:
BIG-Bench Hard (BBH), as well as domain-specific and general NLP tasks, showing
it significantly outperforms strong Chain-of-Thought and recent prompt
optimization baselines. Extensive analyses emphasize its capability to craft
expert-level, detailed, and domain-insightful prompts with great efficiency and
generalizability. | cs.CL | PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimization |
2023-10-25T07:10:42Z | http://arxiv.org/pdf/2310.16417v1 | http://arxiv.org/abs/2310.16417v1 | Kang Kim, Hankyu Cho | 2023-10-25T07:10:42Z | cs.CL | Recent years have seen remarkable advances in the field of Simultaneous
Machine Translation (SiMT) due to the introduction of innovative policies that
dictate whether to READ or WRITE at each step of the translation process.
However, a common assumption in many existing studies is that operations are
carried out at the subword level, even though the standard unit for input and
output in most practical scenarios is typically at the word level. This paper
demonstrates that policies devised and validated at the subword level are
surpassed by those operating at the word level, which process multiple subwords
to form a complete word in a single step. Additionally, we suggest a method to
boost SiMT models using language models (LMs), wherein the proposed word-level
policy plays a vital role in addressing the subword disparity between LMs and
SiMT models. Code is available at https://github.com/xl8-ai/WordSiMT. | cs.CL | Enhanced Simultaneous Machine Translation with Word-level Policies |
2023-10-25T06:54:39Z | http://arxiv.org/pdf/2310.16411v1 | http://arxiv.org/abs/2310.16411v1 | Alon Goldstein, Miriam Havin, Roi Reichart, Ariel Goldstein | 2023-10-25T06:54:39Z | cs.CL, cs.HC | This paper investigates the problem-solving capabilities of Large Language
Models (LLMs) by evaluating their performance on stumpers, unique single-step
intuition problems that pose challenges for human solvers but are easily
verifiable. We compare the performance of four state-of-the-art LLMs
(Davinci-2, Davinci-3, GPT-3.5-Turbo, GPT-4) to human participants. Our
findings reveal that the new-generation LLMs excel in solving stumpers and
surpass human performance. However, humans exhibit superior skills in verifying
solutions to the same problems. This research enhances our understanding of
LLMs' cognitive abilities and provides insights for enhancing their
problem-solving potential across various domains. | cs.CL | Decoding Stumpers: Large Language Models vs. Human Problem-Solvers |
2023-10-25T06:38:42Z | http://arxiv.org/pdf/2310.16402v1 | http://arxiv.org/abs/2310.16402v1 | Ji Jiang, Meng Cao, Tengtao Song, Long Chen, Yi Wang, Yuexian Zou | 2023-10-25T06:38:42Z | cs.CV, cs.CL | Video Referring Expression Comprehension (REC) aims to localize a target
object in videos based on the queried natural language. Recent improvements in
video REC have been made using Transformer-based methods with learnable
queries. However, we contend that this naive query design is not ideal given
the open-world nature of video REC brought by text supervision. With numerous
potential semantic categories, relying on only a few slow-updated queries is
insufficient to characterize them. Our solution to this problem is to create
dynamic queries that are conditioned on both the input video and language to
model the diverse objects referred to. Specifically, we place a fixed number of
learnable bounding boxes throughout the frame and use corresponding region
features to provide prior information. Also, we noticed that current query
features overlook the importance of cross-modal alignment. To address this, we
align specific phrases in the sentence with semantically relevant visual areas,
annotating them in existing video datasets (VID-Sentence and VidSTG). By
incorporating these two designs, our proposed model (called ConFormer)
outperforms other models on widely benchmarked datasets. For example, in the
testing split of VID-Sentence dataset, ConFormer achieves 8.75% absolute
improvement on [email protected] compared to the previous state-of-the-art model. | cs.CV | Video Referring Expression Comprehension via Transformer with Content-conditioned Query |
2023-10-25T06:23:48Z | http://arxiv.org/pdf/2311.02089v1 | http://arxiv.org/abs/2311.02089v1 | Zhenrui Yue, Sara Rabhi, Gabriel de Souza Pereira Moreira, Dong Wang, Even Oldridge | 2023-10-25T06:23:48Z | cs.IR, cs.AI, cs.CL | Recently, large language models (LLMs) have exhibited significant progress in
language understanding and generation. By leveraging textual features,
customized LLMs are also applied for recommendation and demonstrate
improvements across diverse recommendation scenarios. Yet the majority of
existing methods perform training-free recommendation that heavily relies on
pretrained knowledge (e.g., movie recommendation). In addition, inference on
LLMs is slow due to autoregressive generation, rendering existing methods less
effective for real-time recommendation. As such, we propose a two-stage
framework using large language models for ranking-based recommendation
(LlamaRec). In particular, we use small-scale sequential recommenders to
retrieve candidates based on the user interaction history. Then, both history
and retrieved items are fed to the LLM in text via a carefully designed prompt
template. Instead of generating next-item titles, we adopt a verbalizer-based
approach that transforms output logits into probability distributions over the
candidate items. Therefore, the proposed LlamaRec can efficiently rank items
without generating long text. To validate the effectiveness of the proposed
framework, we compare against state-of-the-art baseline methods on benchmark
datasets. Our experimental results demonstrate the performance of LlamaRec,
which consistently achieves superior performance in both recommendation
performance and efficiency. | cs.IR | LlamaRec: Two-Stage Recommendation using Large Language Models for Ranking |
2023-10-25T06:22:29Z | http://arxiv.org/pdf/2310.16393v1 | http://arxiv.org/abs/2310.16393v1 | Vipul Rathore, Rajdeep Dhingra, Parag Singla, Mausam | 2023-10-25T06:22:29Z | cs.CL | We tackle the problem of zero-shot cross-lingual transfer in NLP tasks via
the use of language adapters (LAs). Most of the earlier works have explored
training with adapter of a single source (often English), and testing either
using the target LA or LA of another related language. Training target LA
requires unlabeled data, which may not be readily available for low resource
unseen languages: those that are neither seen by the underlying multilingual
language model (e.g., mBERT), nor do we have any (labeled or unlabeled) data
for them. We posit that for more effective cross-lingual transfer, instead of
just one source LA, we need to leverage LAs of multiple (linguistically or
geographically related) source languages, both at train and test-time - which
we investigate via our novel neural architecture, ZGUL. Extensive
experimentation across four language groups, covering 15 unseen target
languages, demonstrates improvements of up to 3.2 average F1 points over
standard fine-tuning and other strong baselines on POS tagging and NER tasks.
We also extend ZGUL to settings where either (1) some unlabeled data or (2)
few-shot training examples are available for the target language. We find that
ZGUL continues to outperform baselines in these settings too. | cs.CL | ZGUL: Zero-shot Generalization to Unseen Languages using Multi-source Ensembling of Language Adapters |
2023-10-25T05:12:35Z | http://arxiv.org/pdf/2310.16368v1 | http://arxiv.org/abs/2310.16368v1 | Masashi Oshika, Kosuke Yamada, Ryohei Sasano, Koichi Takeda | 2023-10-25T05:12:35Z | cs.CL | It has been known to be difficult to generate adequate sports updates from a
sequence of vast amounts of diverse live tweets, although the live sports
viewing experience with tweets is gaining the popularity. In this paper, we
focus on soccer matches and work on building a system to generate live updates
for soccer matches from tweets so that users can instantly grasp a match's
progress and enjoy the excitement of the match from raw tweets. Our proposed
system is based on a large pre-trained language model and incorporates a
mechanism to control the number of updates and a mechanism to reduce the
redundancy of duplicate and similar updates. | cs.CL | Transformer-based Live Update Generation for Soccer Matches from Microblog Posts |
2023-10-25T04:56:07Z | http://arxiv.org/pdf/2310.16361v1 | http://arxiv.org/abs/2310.16361v1 | Besnik Fetahu, Zhiyu Chen, Oleg Rokhlenko, Shervin Malmasi | 2023-10-25T04:56:07Z | cs.CL, cs.AI | E-commerce product catalogs contain billions of items. Most products have
lengthy titles, as sellers pack them with product attributes to improve
retrieval, and highlight key product aspects. This results in a gap between
such unnatural products titles, and how customers refer to them. It also limits
how e-commerce stores can use these seller-provided titles for recommendation,
QA, or review summarization.
Inspired by recent work on instruction-tuned LLMs, we present InstructPTS, a
controllable approach for the task of Product Title Summarization (PTS).
Trained using a novel instruction fine-tuning strategy, our approach is able to
summarize product titles according to various criteria (e.g. number of words in
a summary, inclusion of specific phrases, etc.). Extensive evaluation on a
real-world e-commerce catalog shows that compared to simple fine-tuning of
LLMs, our proposed approach can generate more accurate product name summaries,
with an improvement of over 14 and 8 BLEU and ROUGE points, respectively. | cs.CL | InstructPTS: Instruction-Tuning LLMs for Product Title Summarization |
2023-10-25T04:38:02Z | http://arxiv.org/pdf/2310.16358v1 | http://arxiv.org/abs/2310.16358v1 | Quzhe Huang, Yanxi Zhang, Dongyan Zhao | 2023-10-25T04:38:02Z | cs.CL | Document-level Event Argument Extraction (EAE) requires the model to extract
arguments of multiple events from a single document. Considering the underlying
dependencies between these events, recent efforts leverage the idea of
"memory", where the results of already predicted events are cached and can be
retrieved to help the prediction of upcoming events. These methods extract
events according to their appearance order in the document, however, the event
that appears in the first sentence does not mean that it is the easiest to
extract. Existing methods might introduce noise to the extraction of upcoming
events if they rely on an incorrect prediction of previous events. In order to
provide more reliable memory, we propose a simple-to-complex progressive
framework for document-level EAE. Specifically, we first calculate the
difficulty of each event and then, we conduct the extraction following a
simple-to-complex order. In this way, the memory will store the most certain
results, and the model could use these reliable sources to help the prediction
of more difficult events. Experiments on WikiEvents show that our model
outperforms SOTA by 1.4% in F1, indicating the proposed simple-to-complex
framework is useful in the EAE task. | cs.CL | From Simple to Complex: A Progressive Framework for Document-level Informative Argument Extraction |
2023-10-25T04:35:06Z | http://arxiv.org/pdf/2310.16356v1 | http://arxiv.org/abs/2310.16356v1 | Yoshinari Fujinuma, Siddharth Varia, Nishant Sankaran, Srikar Appalaraju, Bonan Min, Yogarshi Vyas | 2023-10-25T04:35:06Z | cs.CL | Document image classification is different from plain-text document
classification and consists of classifying a document by understanding the
content and structure of documents such as forms, emails, and other such
documents. We show that the only existing dataset for this task (Lewis et al.,
2006) has several limitations and we introduce two newly curated multilingual
datasets WIKI-DOC and MULTIEURLEX-DOC that overcome these limitations. We
further undertake a comprehensive study of popular visually-rich document
understanding or Document AI models in previously untested setting in document
image classification such as 1) multi-label classification, and 2) zero-shot
cross-lingual transfer setup. Experimental results show limitations of
multilingual Document AI models on cross-lingual transfer across typologically
distant languages. Our datasets and findings open the door for future research
into improving Document AI models. | cs.CL | A Multi-Modal Multilingual Benchmark for Document Image Classification |
2023-10-25T04:22:40Z | http://arxiv.org/pdf/2310.16350v2 | http://arxiv.org/abs/2310.16350v2 | Xiaobing Sun, Jiaxi Li, Wei Lu | 2023-10-26T03:26:30Z | cs.CL | The underlying mechanism of neural networks in capturing precise knowledge
has been the subject of consistent research efforts. In this work, we propose a
theoretical approach based on Neural Tangent Kernels (NTKs) to investigate such
mechanisms. Specifically, considering the infinite network width, we
hypothesize the learning dynamics of target models may intuitively unravel the
features they acquire from training data, deepening our insights into their
internal mechanisms. We apply our approach to several fundamental models and
reveal how these models leverage statistical features during gradient descent
and how they are integrated into final decisions. We also discovered that the
choice of activation function can affect feature extraction. For instance, the
use of the \textit{ReLU} activation function could potentially introduce a bias
in features, providing a plausible explanation for its replacement with
alternative functions in recent pre-trained language models. Additionally, we
find that while self-attention and CNN models may exhibit limitations in
learning n-grams, multiplication-based models seem to excel in this area. We
verify these theoretical findings through experiments and find that they can be
applied to analyze language modeling tasks, which can be regarded as a special
variant of classification. Our contributions offer insights into the roles and
capacities of fundamental components within large language models, thereby
aiding the broader understanding of these complex systems. | cs.CL | Unraveling Feature Extraction Mechanisms in Neural Networks |
2023-10-25T03:58:49Z | http://arxiv.org/pdf/2310.16343v2 | http://arxiv.org/abs/2310.16343v2 | Xiang Chen, Xiaojun Wan | 2024-03-21T08:29:35Z | cs.CL | Advancements in natural language generation (NLG) and large language models
(LLMs) have led to proficient text generation in various tasks. However,
integrating intricate constraints into neural text generation, due to LLMs'
opacity, remains challenging. This study investigates constrained text
generation for LLMs, where predefined constraints are applied during LLM's
generation process. Our research mainly focuses on mainstream open-source LLMs,
categorizing constraints into lexical, structural, and relation-based types. We
also present various benchmarks to facilitate fair evaluation. The study
addresses some key research questions, including evaluating, understanding and
improving constrained text generation for LLMs. Results illuminate LLMs'
capacity and deficiency to incorporate constraints and provide insights for
future developments in constrained text generation. Codes and datasets will be
released upon acceptance. | cs.CL | Evaluating, Understanding, and Improving Constrained Text Generation for Large Language Models |
2023-10-25T03:53:31Z | http://arxiv.org/pdf/2310.16340v1 | http://arxiv.org/abs/2310.16340v1 | Zefan Wang, Zichuan Liu, Yingying Zhang, Aoxiao Zhong, Lunting Fan, Lingfei Wu, Qingsong Wen | 2023-10-25T03:53:31Z | cs.SE, cs.CL | Large language model (LLM) applications in cloud root cause analysis (RCA)
have been actively explored recently. However, current methods are still
reliant on manual workflow settings and do not unleash LLMs' decision-making
and environment interaction capabilities. We present RCAgent, a tool-augmented
LLM autonomous agent framework for practical and privacy-aware industrial RCA
usage. Running on an internally deployed model rather than GPT families,
RCAgent is capable of free-form data collection and comprehensive analysis with
tools. Our framework combines a variety of enhancements, including a unique
Self-Consistency for action trajectories, and a suite of methods for context
management, stabilization, and importing domain knowledge. Our experiments show
RCAgent's evident and consistent superiority over ReAct across all aspects of
RCA -- predicting root causes, solutions, evidence, and responsibilities -- and
tasks covered or uncovered by current rules, as validated by both automated
metrics and human evaluations. Furthermore, RCAgent has already been integrated
into the diagnosis and issue discovery workflow of the Real-time Compute
Platform for Apache Flink of Alibaba Cloud. | cs.SE | RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models |
2023-10-25T03:40:50Z | http://arxiv.org/pdf/2310.16338v2 | http://arxiv.org/abs/2310.16338v2 | Alexander H. Liu, Matt Le, Apoorv Vyas, Bowen Shi, Andros Tjandra, Wei-Ning Hsu | 2024-03-25T18:18:40Z | eess.AS, cs.CL, cs.LG, cs.SD | Generative models have gained more and more attention in recent years for
their remarkable success in tasks that required estimating and sampling data
distribution to generate high-fidelity synthetic data. In speech,
text-to-speech synthesis and neural vocoder are good examples where generative
models have shined. While generative models have been applied to different
applications in speech, there exists no general-purpose generative model that
models speech directly. In this work, we take a step toward this direction by
showing a single pre-trained generative model can be adapted to different
downstream tasks with strong performance. Specifically, we pre-trained a
generative model, named SpeechFlow, on 60k hours of untranscribed speech with
Flow Matching and masked conditions. Experiment results show the pre-trained
generative model can be fine-tuned with task-specific data to match or surpass
existing expert models on speech enhancement, separation, and synthesis. Our
work suggested a foundational model for generation tasks in speech can be built
with generative pre-training. | eess.AS | Generative Pre-training for Speech with Flow Matching |
2023-10-25T03:21:20Z | http://arxiv.org/pdf/2310.16329v1 | http://arxiv.org/abs/2310.16329v1 | Aviya Maimon, Reut Tsarfaty | 2023-10-25T03:21:20Z | cs.CL, cs.AI, cs.DB | Coherence is a linguistic term that refers to the relations between small
textual units (sentences, propositions), which make the text logically
consistent and meaningful to the reader. With the advances of generative
foundational models in NLP, there is a pressing need to automatically assess
the human-perceived coherence of automatically generated texts. Up until now,
little work has been done on explicitly assessing the coherence of generated
texts and analyzing the factors contributing to (in)coherence. Previous work on
the topic used other tasks, e.g., sentence reordering, as proxies of coherence,
rather than approaching coherence detection heads on. In this paper, we
introduce {\sc CoheSentia}, a novel benchmark of human-perceived coherence of
automatically generated texts. Our annotation protocol reflects two
perspectives; one is global, assigning a single coherence score, and the other
is incremental, scoring sentence by sentence. The incremental method produces
an (in)coherence score for each text fragment and also pinpoints reasons for
incoherence at that point. Our benchmark contains 500 automatically-generated
and human-annotated paragraphs, each annotated in both methods, by multiple
raters. Our analysis shows that the inter-annotator agreement in the
incremental mode is higher than in the holistic alternative, and our
experiments show that standard LMs fine-tuned for coherence detection show
varied performance on the different factors contributing to (in)coherence. All
in all, these models yield unsatisfactory performance, emphasizing the need for
developing more reliable methods for coherence assessment. | cs.CL | CoheSentia: A Novel Benchmark of Incremental versus Holistic Assessment of Coherence in Generated Texts |
2023-10-25T03:10:52Z | http://arxiv.org/pdf/2310.16322v1 | http://arxiv.org/abs/2310.16322v1 | Jan Christian Blaise Cruz | 2023-10-25T03:10:52Z | cs.CL | In this paper, we describe the constrained MT systems submitted by Samsung
R&D Institute Philippines to the WMT 2023 General Translation Task for two
directions: en$\rightarrow$he and he$\rightarrow$en. Our systems comprise of
Transformer-based sequence-to-sequence models that are trained with a mix of
best practices: comprehensive data preprocessing pipelines, synthetic
backtranslated data, and the use of noisy channel reranking during online
decoding. Our models perform comparably to, and sometimes outperform, strong
baseline unconstrained systems such as mBART50 M2M and NLLB 200 MoE despite
having significantly fewer parameters on two public benchmarks: FLORES-200 and
NTREX-128. | cs.CL | Samsung R&D Institute Philippines at WMT 2023 |
2023-10-25T03:04:57Z | http://arxiv.org/pdf/2310.16319v1 | http://arxiv.org/abs/2310.16319v1 | Yukun Zhao, Lingyong Yan, Weiwei Sun, Chong Meng, Shuaiqiang Wang, Zhicong Cheng, Zhaochun Ren, Dawei Yin | 2023-10-25T03:04:57Z | cs.CL | Dialogue assessment plays a critical role in the development of open-domain
dialogue systems. Existing work are uncapable of providing an end-to-end and
human-epistemic assessment dataset, while they only provide sub-metrics like
coherence or the dialogues are conversed between annotators far from real user
settings. In this paper, we release a large-scale dialogue quality assessment
dataset (DiQAD), for automatically assessing open-domain dialogue quality.
Specifically, we (1) establish the assessment criteria based on the dimensions
conforming to human judgements on dialogue qualities, and (2) annotate
large-scale dialogues that conversed between real users based on these
annotation criteria, which contains around 100,000 dialogues. We conduct
several experiments and report the performances of the baselines as the
benchmark on DiQAD. The dataset is openly accessible at
https://github.com/yukunZhao/Dataset_Dialogue_quality_evaluation. | cs.CL | DiQAD: A Benchmark Dataset for End-to-End Open-domain Dialogue Assessment |
2023-10-25T02:22:50Z | http://arxiv.org/pdf/2310.16303v1 | http://arxiv.org/abs/2310.16303v1 | Ayesha Qamar, Chetan Verma, Ahmed El-Kishky, Sumit Binnani, Sneha Mehta, Taylor Berg-Kirkpatrick | 2023-10-25T02:22:50Z | cs.CL, cs.IR | Understanding and representing webpages is crucial to online social networks
where users may share and engage with URLs. Common language model (LM) encoders
such as BERT can be used to understand and represent the textual content of
webpages. However, these representations may not model thematic information of
web domains and URLs or accurately capture their appeal to social media users.
In this work, we introduce a new pre-training objective that can be used to
adapt LMs to understand URLs and webpages. Our proposed framework consists of
two steps: (1) scalable graph embeddings to learn shallow representations of
URLs based on user engagement on social media and (2) a contrastive objective
that aligns LM representations with the aforementioned graph-based
representation. We apply our framework to the multilingual version of BERT to
obtain the model URL-BERT. We experimentally demonstrate that our continued
pre-training approach improves webpage understanding on a variety of tasks and
Twitter internal and external benchmarks. | cs.CL | URL-BERT: Training Webpage Representations via Social Media Engagements |
2023-10-25T02:18:40Z | http://arxiv.org/pdf/2310.16301v1 | http://arxiv.org/abs/2310.16301v1 | Chao-Hong Tan, Jia-Chen Gu, Zhen-Hua Ling | 2023-10-25T02:18:40Z | cs.CL | Large Language Models (LLMs) have emerged as influential instruments within
the realm of natural language processing; nevertheless, their capacity to
handle multi-party conversations (MPCs) -- a scenario marked by the presence of
multiple interlocutors involved in intricate information exchanges -- remains
uncharted. In this paper, we delve into the potential of generative LLMs such
as ChatGPT and GPT-4 within the context of MPCs. An empirical analysis is
conducted to assess the zero-shot learning capabilities of ChatGPT and GPT-4 by
subjecting them to evaluation across three MPC datasets that encompass five
representative tasks. The findings reveal that ChatGPT's performance on a
number of evaluated MPC tasks leaves much to be desired, whilst GPT-4's results
portend a promising future. Additionally, we endeavor to bolster performance
through the incorporation of MPC structures, encompassing both speaker and
addressee architecture. This study provides an exhaustive evaluation and
analysis of applying generative LLMs to MPCs, casting a light upon the
conception and creation of increasingly effective and robust MPC agents.
Concurrently, this work underscores the challenges implicit in the utilization
of LLMs for MPCs, such as deciphering graphical information flows and
generating stylistically consistent responses. | cs.CL | Is ChatGPT a Good Multi-Party Conversation Solver? |
2023-10-25T01:30:26Z | http://arxiv.org/pdf/2312.03707v2 | http://arxiv.org/abs/2312.03707v2 | Hongren Wang | 2024-05-21T09:14:04Z | cs.CL | This study addresses the challenges of multi-label text classification. The
difficulties arise from imbalanced data sets, varied text lengths, and numerous
subjective feature labels. Existing solutions include traditional machine
learning and deep neural networks for predictions. However, both approaches
have their limitations. Traditional machine learning often overlooks the
associations between words, while deep neural networks, despite their better
classification performance, come with increased training complexity and time.
This paper proposes a method utilizing the bag-of-words model approach based on
the GloVe model and the CNN-BiLSTM network. The principle is to use the word
vector matrix trained by the GloVe model as the input for the text embedding
layer. Given that the GloVe model requires no further training, the neural
network model can be trained more efficiently. The method achieves an accuracy
rate of 87.26% on the test set and an F1 score of 0.8737, showcasing promising
results. | cs.CL | Multi-label Text Classification using GloVe and Neural Network Models |
2023-10-25T01:20:17Z | http://arxiv.org/pdf/2310.16278v1 | http://arxiv.org/abs/2310.16278v1 | Yi-Chen Chang, Canasai Kruengkrai, Junichi Yamagishi | 2023-10-25T01:20:17Z | cs.CL, cs.AI | This paper introduces the Cross-lingual Fact Extraction and VERification
(XFEVER) dataset designed for benchmarking the fact verification models across
different languages. We constructed it by translating the claim and evidence
texts of the Fact Extraction and VERification (FEVER) dataset into six
languages. The training and development sets were translated using machine
translation, whereas the test set includes texts translated by professional
translators and machine-translated texts. Using the XFEVER dataset, two
cross-lingual fact verification scenarios, zero-shot learning and
translate-train learning, are defined, and baseline models for each scenario
are also proposed in this paper. Experimental results show that the
multilingual language model can be used to build fact verification models in
different languages efficiently. However, the performance varies by language
and is somewhat inferior to the English case. We also found that we can
effectively mitigate model miscalibration by considering the prediction
similarity between the English and target languages. The XFEVER dataset, code,
and model checkpoints are available at
https://github.com/nii-yamagishilab/xfever. | cs.CL | XFEVER: Exploring Fact Verification across Languages |
2023-10-25T01:07:50Z | http://arxiv.org/pdf/2310.18365v2 | http://arxiv.org/abs/2310.18365v2 | Luyang Fang, Gyeong-Geon Lee, Xiaoming Zhai | 2023-11-18T02:05:27Z | cs.CL, cs.AI, cs.CY | Machine learning-based automatic scoring can be challenging if students'
responses are unbalanced across scoring categories, as it introduces
uncertainty in the machine training process. To meet this challenge, we
introduce a novel text data augmentation framework using GPT-4, a generative
large language model, specifically tailored for unbalanced datasets in
automatic scoring. Our experimental dataset comprised student-written responses
to two science items. We crafted prompts for GPT-4 to generate responses
resembling student-written answers, particularly for the minority scoring
classes, to augment the data. We then finetuned DistillBERT for automatic
scoring based on the augmented and original datasets. Model performance was
assessed using accuracy, precision, recall, and F1 score. We incorporate varied
amounts of augmented data to examine scoring performance, and our findings
revealed remarkedly improved model performance. The average maximum increase
observed across two items is: 3.5% for accuracy, 30.6% for precision, 21.1% for
recall, and 24.2% for F1 score. Notably, using just 5% of the augmented data
led to substantial improvements: 2.6%, 29.2%, 15.1%, and 19.6%. Interestingly,
the extent of improvement varied depending on specific datasets. Moreover, we
found that a varying amount of augmented data (5%-40%) was needed to obtain a
stable improvement. We also compare models trained with GPT-4 augmented data
and those trained with additional student-written responses. The findings
indicate that former ones match or even exceed the performance of the latter.
Specifically, there is an average difference of 1.7%, 1.9%, 11.0%, and 7.8% for
four metrics separately. This research underscores the potential and
effectiveness of data augmentation techniques utilizing GPT-4 in addressing
unbalanced datasets within automated assessment. | cs.CL | Using GPT-4 to Augment Unbalanced Data for Automatic Scoring |
2023-10-25T01:05:03Z | http://arxiv.org/pdf/2310.16271v1 | http://arxiv.org/abs/2310.16271v1 | Jixiang Hong, Quan Tu, Changyu Chen, Xing Gao, Ji Zhang, Rui Yan | 2023-10-25T01:05:03Z | cs.CL, cs.AI | Language models trained on large-scale corpus often generate content that is
harmful, toxic, or contrary to human preferences, making their alignment with
human values a critical concern. Reinforcement learning from human feedback
(RLHF) with algorithms like PPO is a prevalent approach for alignment but is
often complex, unstable, and resource-intensive. Recently, ranking-based
alignment methods have emerged, offering stability and effectiveness by
replacing the RL framework with supervised fine-tuning, but they are costly due
to the need for annotated data. Considering that existing large language models
(LLMs) like ChatGPT are already relatively well-aligned and cost-friendly,
researchers have begun to align the language model with human preference from
AI feedback. The common practices, which unidirectionally distill the
instruction-following responses from LLMs, are constrained by their bottleneck.
Thus we introduce CycleAlign to distill alignment capabilities from
parameter-invisible LLMs (black-box) to a parameter-visible model (white-box)
in an iterative manner. With in-context learning (ICL) as the core of the
cycle, the black-box models are able to rank the model-generated responses
guided by human-craft instruction and demonstrations about their preferences.
During iterative interaction, the white-box models also have a judgment about
responses generated by them. Consequently, the agreement ranking could be
viewed as a pseudo label to dynamically update the in-context demonstrations
and improve the preference ranking ability of black-box models. Through
multiple interactions, the CycleAlign framework could align the white-box model
with the black-box model effectively in a low-resource way. Empirical results
illustrate that the model fine-tuned by CycleAlign remarkably exceeds existing
methods, and achieves the state-of-the-art performance in alignment with human
value. | cs.CL | CycleAlign: Iterative Distillation from Black-box LLM to White-box Models for Better Human Alignment |
2023-10-25T01:03:35Z | http://arxiv.org/pdf/2310.16270v1 | http://arxiv.org/abs/2310.16270v1 | Mansi Sakarvadia, Arham Khan, Aswathy Ajith, Daniel Grzenda, Nathaniel Hudson, André Bauer, Kyle Chard, Ian Foster | 2023-10-25T01:03:35Z | cs.CL, cs.AI, cs.LG | Transformer-based Large Language Models (LLMs) are the state-of-the-art for
natural language tasks. Recent work has attempted to decode, by reverse
engineering the role of linear layers, the internal mechanisms by which LLMs
arrive at their final predictions for text completion tasks. Yet little is
known about the specific role of attention heads in producing the final token
prediction. We propose Attention Lens, a tool that enables researchers to
translate the outputs of attention heads into vocabulary tokens via learned
attention-head-specific transformations called lenses. Preliminary findings
from our trained lenses indicate that attention heads play highly specialized
roles in language models. The code for Attention Lens is available at
github.com/msakarvadia/AttentionLens. | cs.CL | Attention Lens: A Tool for Mechanistically Interpreting the Attention Head Information Retrieval Mechanism |
2023-10-25T01:01:28Z | http://arxiv.org/pdf/2310.16269v1 | http://arxiv.org/abs/2310.16269v1 | Cristina España-Bonet | 2023-10-25T01:01:28Z | cs.CL, cs.AI, cs.CY | Neutrality is difficult to achieve and, in politics, subjective. Traditional
media typically adopt an editorial line that can be used by their potential
readers as an indicator of the media bias. Several platforms currently rate
news outlets according to their political bias. The editorial line and the
ratings help readers in gathering a balanced view of news. But in the advent of
instruction-following language models, tasks such as writing a newspaper
article can be delegated to computers. Without imposing a biased persona, where
would an AI-based news outlet lie within the bias ratings? In this work, we use
the ratings of authentic news outlets to create a multilingual corpus of news
with coarse stance annotations (Left and Right) along with automatically
extracted topic annotations. We show that classifiers trained on this data are
able to identify the editorial line of most unseen newspapers in English,
German, Spanish and Catalan. We then apply the classifiers to 101
newspaper-like articles written by ChatGPT and Bard in the 4 languages at
different time periods. We observe that, similarly to traditional newspapers,
ChatGPT editorial line evolves with time and, being a data-driven system, the
stance of the generated articles differs among languages. | cs.CL | Multilingual Coarse Political Stance Classification of Media. The Editorial Line of a ChatGPT and Bard Newspaper |
2023-10-25T00:32:56Z | http://arxiv.org/pdf/2310.16263v1 | http://arxiv.org/abs/2310.16263v1 | Jiexin Wang, Liuwen Cao, Xitong Luo, Zhiping Zhou, Jiayuan Xie, Adam Jatowt, Yi Cai | 2023-10-25T00:32:56Z | cs.SE, cs.AI, cs.CL, cs.CR | Large language models (LLMs) have brought significant advancements to code
generation, benefiting both novice and experienced developers. However, their
training using unsanitized data from open-source repositories, like GitHub,
introduces the risk of inadvertently propagating security vulnerabilities. To
effectively mitigate this concern, this paper presents a comprehensive study
focused on evaluating and enhancing code LLMs from a software security
perspective. We introduce SecuCoGen\footnote{SecuCoGen has been uploaded as
supplemental material and will be made publicly available after publication.},
a meticulously curated dataset targeting 21 critical vulnerability types.
SecuCoGen comprises 180 samples and serves as the foundation for conducting
experiments on three crucial code-related tasks: code generation, code repair
and vulnerability classification, with a strong emphasis on security. Our
experimental results reveal that existing models often overlook security
concerns during code generation, leading to the generation of vulnerable code.
To address this, we propose effective approaches to mitigate the security
vulnerabilities and enhance the overall robustness of code generated by LLMs.
Moreover, our study identifies weaknesses in existing models' ability to repair
vulnerable code, even when provided with vulnerability information.
Additionally, certain vulnerability types pose challenges for the models,
hindering their performance in vulnerability classification. Based on these
findings, we believe our study will have a positive impact on the software
engineering community, inspiring the development of improved methods for
training and utilizing LLMs, thereby leading to safer and more trustworthy
model deployment. | cs.SE | Enhancing Large Language Models for Secure Code Generation: A Dataset-driven Study on Vulnerability Mitigation |
2023-10-25T00:31:29Z | http://arxiv.org/pdf/2310.16261v1 | http://arxiv.org/abs/2310.16261v1 | Ting-Rui Chiang, Dani Yogatama | 2023-10-25T00:31:29Z | cs.CL | We analyze the masked language modeling pretraining objective function from
the perspective of the distributional hypothesis. We investigate whether better
sample efficiency and the better generalization capability of models pretrained
with masked language modeling can be attributed to the semantic similarity
encoded in the pretraining data's distributional property. Via a synthetic
dataset, our analysis suggests that distributional property indeed leads to the
better sample efficiency of pretrained masked language models, but does not
fully explain the generalization capability. We also conduct analyses over two
real-world datasets and demonstrate that the distributional property does not
explain the generalization ability of pretrained natural language models
either. Our results illustrate our limited understanding of model pretraining
and provide future research directions. | cs.CL | The Distributional Hypothesis Does Not Fully Explain the Benefits of Masked Language Model Pretraining |
2023-10-24T23:53:15Z | http://arxiv.org/pdf/2310.16251v1 | http://arxiv.org/abs/2310.16251v1 | Dhruv Kumar, Vipul Raheja, Alice Kaiser-Schatzlein, Robyn Perry, Apurva Joshi, Justin Hugues-Nuger, Samuel Lou, Navid Chowdhury | 2023-10-24T23:53:15Z | cs.CL, cs.AI | We present Speakerly, a new real-time voice-based writing assistance system
that helps users with text composition across various use cases such as emails,
instant messages, and notes. The user can interact with the system through
instructions or dictation, and the system generates a well-formatted and
coherent document. We describe the system architecture and detail how we
address the various challenges while building and deploying such a system at
scale. More specifically, our system uses a combination of small, task-specific
models as well as pre-trained language models for fast and effective text
composition while supporting a variety of input modes for better usability. | cs.CL | Speakerly: A Voice-based Writing Assistant for Text Composition |
2023-10-24T23:45:57Z | http://arxiv.org/pdf/2310.16248v3 | http://arxiv.org/abs/2310.16248v3 | Amir Hossein Kargaran, Ayyoob Imani, François Yvon, Hinrich Schütze | 2024-07-02T23:34:35Z | cs.CL | Several recent papers have published good solutions for language
identification (LID) for about 300 high-resource and medium-resource languages.
However, there is no LID available that (i) covers a wide range of low-resource
languages, (ii) is rigorously evaluated and reliable and (iii) efficient and
easy to use. Here, we publish GlotLID-M, an LID model that satisfies the
desiderata of wide coverage, reliability and efficiency. It identifies 1665
languages, a large increase in coverage compared to prior work. In our
experiments, GlotLID-M outperforms four baselines (CLD3, FT176, OpenLID and
NLLB) when balancing F1 and false positive rate (FPR). We analyze the unique
challenges that low-resource LID poses: incorrect corpus metadata, leakage from
high-resource languages, difficulty separating closely related languages,
handling of macrolanguage vs varieties and in general noisy data. We hope that
integrating GlotLID-M into dataset creation pipelines will improve quality and
enhance accessibility of NLP technology for low-resource languages and
cultures. GlotLID-M model (including future versions), code, and list of data
sources are available: https://github.com/cisnlp/GlotLID. | cs.CL | GlotLID: Language Identification for Low-Resource Languages |
2023-10-24T23:30:17Z | http://arxiv.org/pdf/2310.16242v2 | http://arxiv.org/abs/2310.16242v2 | Yonchanok Khaokaew, Kaixin Ji, Thuc Hanh Nguyen, Hiruni Kegalle, Marwah Alaofi, Hao Xue, Flora D. Salim | 2024-05-07T00:20:30Z | cs.LG, cs.CL | This paper explores the intersection of technology and sleep pattern
comprehension, presenting a cutting-edge two-stage framework that harnesses the
power of Large Language Models (LLMs). The primary objective is to deliver
precise sleep predictions paired with actionable feedback, addressing the
limitations of existing solutions. This innovative approach involves leveraging
the GLOBEM dataset alongside synthetic data generated by LLMs. The results
highlight significant improvements, underlining the efficacy of merging
advanced machine-learning techniques with a user-centric design ethos. Through
this exploration, we bridge the gap between technological sophistication and
user-friendly design, ensuring that our framework yields accurate predictions
and translates them into actionable insights. | cs.LG | ZzzGPT: An Interactive GPT Approach to Enhance Sleep Quality |
2023-10-24T23:29:06Z | http://arxiv.org/pdf/2310.16240v1 | http://arxiv.org/abs/2310.16240v1 | Raymond Li, Gabriel Murray, Giuseppe Carenini | 2023-10-24T23:29:06Z | cs.CL | In this work, we propose a method that combines two popular research areas by
injecting linguistic structures into pre-trained language models in the
parameter-efficient fine-tuning (PEFT) setting. In our approach, parallel
adapter modules encoding different linguistic structures are combined using a
novel Mixture-of-Linguistic-Experts architecture, where Gumbel-Softmax gates
are used to determine the importance of these modules at each layer of the
model. To reduce the number of parameters, we first train the model for a fixed
small number of steps before pruning the experts based on their importance
scores. Our experiment results with three different pre-trained models show
that our approach can outperform state-of-the-art PEFT methods with a
comparable number of parameters. In addition, we provide additional analysis to
examine the experts selected by each model at each layer to provide insights
for future studies. | cs.CL | Mixture-of-Linguistic-Experts Adapters for Improving and Interpreting Pre-trained Language Models |
2023-10-24T22:41:14Z | http://arxiv.org/pdf/2310.16226v3 | http://arxiv.org/abs/2310.16226v3 | Saurabh Garg, Mehrdad Farajtabar, Hadi Pouransari, Raviteja Vemulapalli, Sachin Mehta, Oncel Tuzel, Vaishaal Shankar, Fartash Faghri | 2024-03-21T04:47:27Z | cs.CV, cs.CL, cs.LG | Keeping large foundation models up to date on latest data is inherently
expensive. To avoid the prohibitive costs of constantly retraining, it is
imperative to continually train these models. This problem is exacerbated by
the lack of any large scale continual learning benchmarks or baselines. We
introduce the first set of web-scale Time-Continual (TiC) benchmarks for
training vision-language models: TiC-DataComp, TiC-YFCC, and TiC-Redcaps.
TiC-DataComp, our largest dataset, contains over 12.7B timestamped image-text
pairs spanning 9 years (2014-2022). We first use our benchmarks to curate
various dynamic evaluations to measure temporal robustness of existing models.
We show OpenAI's CLIP (trained on data up to 2020) loses $\approx 8\%$
zero-shot accuracy on our curated retrieval task from 2021-2022 compared with
more recently trained models in OpenCLIP repository. We then study how to
efficiently train models on time-continuous data. We demonstrate that a simple
rehearsal-based approach that continues training from the last checkpoint and
replays old data reduces compute by $2.5\times$ when compared to the standard
practice of retraining from scratch. Code is available at
https://github.com/apple/ml-tic-clip. | cs.CV | TiC-CLIP: Continual Training of CLIP Models |
2023-10-24T22:34:43Z | http://arxiv.org/pdf/2310.16225v1 | http://arxiv.org/abs/2310.16225v1 | Susanna Rücker, Alan Akbik | 2023-10-24T22:34:43Z | cs.CL, cs.AI, cs.LG | The CoNLL-03 corpus is arguably the most well-known and utilized benchmark
dataset for named entity recognition (NER). However, prior works found
significant numbers of annotation errors, incompleteness, and inconsistencies
in the data. This poses challenges to objectively comparing NER approaches and
analyzing their errors, as current state-of-the-art models achieve F1-scores
that are comparable to or even exceed the estimated noise level in CoNLL-03. To
address this issue, we present a comprehensive relabeling effort assisted by
automatic consistency checking that corrects 7.0% of all labels in the English
CoNLL-03. Our effort adds a layer of entity linking annotation both for better
explainability of NER labels and as additional safeguard of annotation quality.
Our experimental evaluation finds not only that state-of-the-art approaches
reach significantly higher F1-scores (97.1%) on our data, but crucially that
the share of correct predictions falsely counted as errors due to annotation
noise drops from 47% to 6%. This indicates that our resource is well suited to
analyze the remaining errors made by state-of-the-art models, and that the
theoretical upper bound even on high resource, coarse-grained NER is not yet
reached. To facilitate such analysis, we make CleanCoNLL publicly available to
the research community. | cs.CL | CleanCoNLL: A Nearly Noise-Free Named Entity Recognition Dataset |
2023-10-24T22:18:13Z | http://arxiv.org/pdf/2310.16218v3 | http://arxiv.org/abs/2310.16218v3 | Song Wang, Yaochen Zhu, Haochen Liu, Zaiyi Zheng, Chen Chen, Jundong Li | 2023-12-14T21:49:59Z | cs.CL, cs.AI | Large language models (LLMs) have recently transformed both the academic and
industrial landscapes due to their remarkable capacity to understand, analyze,
and generate texts based on their vast knowledge and reasoning ability.
Nevertheless, one major drawback of LLMs is their substantial computational
cost for pre-training due to their unprecedented amounts of parameters. The
disadvantage is exacerbated when new knowledge frequently needs to be
introduced into the pre-trained model. Therefore, it is imperative to develop
effective and efficient techniques to update pre-trained LLMs. Traditional
methods encode new knowledge in pre-trained LLMs through direct fine-tuning.
However, naively re-training LLMs can be computationally intensive and risks
degenerating valuable pre-trained knowledge irrelevant to the update in the
model. Recently, Knowledge-based Model Editing (KME) has attracted increasing
attention, which aims to precisely modify the LLMs to incorporate specific
knowledge, without negatively influencing other irrelevant knowledge. In this
survey, we aim to provide a comprehensive and in-depth overview of recent
advances in the field of KME. We first introduce a general formulation of KME
to encompass different KME strategies. Afterward, we provide an innovative
taxonomy of KME techniques based on how the new knowledge is introduced into
pre-trained LLMs, and investigate existing KME strategies while analyzing key
insights, advantages, and limitations of methods from each category. Moreover,
representative metrics, datasets, and applications of KME are introduced
accordingly. Finally, we provide an in-depth analysis regarding the
practicality and remaining challenges of KME and suggest promising research
directions for further advancement in this field. | cs.CL | Knowledge Editing for Large Language Models: A Survey |
2023-10-24T21:30:15Z | http://arxiv.org/pdf/2310.16197v1 | http://arxiv.org/abs/2310.16197v1 | Adithya Pratapa, Kevin Small, Markus Dreyer | 2023-10-24T21:30:15Z | cs.CL | Generating concise summaries of news events is a challenging natural language
processing task. While journalists often curate timelines to highlight key
sub-events, newcomers to a news event face challenges in catching up on its
historical context. In this paper, we address this need by introducing the task
of background news summarization, which complements each timeline update with a
background summary of relevant preceding events. We construct a dataset by
merging existing timeline datasets and asking human annotators to write a
background summary for each timestep of each news event. We establish strong
baseline performance using state-of-the-art summarization systems and propose a
query-focused variant to generate background summaries. To evaluate background
summary quality, we present a question-answering-based evaluation metric,
Background Utility Score (BUS), which measures the percentage of questions
about a current event timestep that a background summary answers. Our
experiments show the effectiveness of instruction fine-tuned systems such as
Flan-T5, in addition to strong zero-shot performance using GPT-3.5. | cs.CL | Background Summarization of Event Timelines |
2023-10-24T21:23:53Z | http://arxiv.org/pdf/2310.16193v1 | http://arxiv.org/abs/2310.16193v1 | Chenghao Xiao, Yizhi Li, G Thomas Hudson, Chenghua Lin, Noura Al Moubayed | 2023-10-24T21:23:53Z | cs.CL, cs.AI | In recent years, contrastive learning (CL) has been extensively utilized to
recover sentence and document-level encoding capability from pre-trained
language models. In this work, we question the length generalizability of
CL-based models, i.e., their vulnerability towards length-induced semantic
shift. We verify not only that length vulnerability is a significant yet
overlooked research gap, but we can devise unsupervised CL methods solely
depending on the semantic signal provided by document length. We first derive
the theoretical foundations underlying length attacks, showing that elongating
a document would intensify the high intra-document similarity that is already
brought by CL. Moreover, we found that isotropy promised by CL is highly
dependent on the length range of text exposed in training. Inspired by these
findings, we introduce a simple yet universal document representation learning
framework, LA(SER)$^{3}$: length-agnostic self-reference for semantically
robust sentence representation learning, achieving state-of-the-art
unsupervised performance on the standard information retrieval benchmark. | cs.CL | Length is a Curse and a Blessing for Document-level Semantics |
2023-10-24T21:00:41Z | http://arxiv.org/pdf/2310.16183v2 | http://arxiv.org/abs/2310.16183v2 | Md. Arid Hasan, Firoj Alam, Anika Anjum, Shudipta Das, Afiyat Anjum | 2024-02-22T02:32:06Z | cs.CL, cs.LG, I.2.7 | We present an overview of the BLP Sentiment Shared Task, organized as part of
the inaugural BLP 2023 workshop, co-located with EMNLP 2023. The task is
defined as the detection of sentiment in a given piece of social media text.
This task attracted interest from 71 participants, among whom 29 and 30 teams
submitted systems during the development and evaluation phases, respectively.
In total, participants submitted 597 runs. However, a total of 15 teams
submitted system description papers. The range of approaches in the submitted
systems spans from classical machine learning models, fine-tuning pre-trained
models, to leveraging Large Language Model (LLMs) in zero- and few-shot
settings. In this paper, we provide a detailed account of the task setup,
including dataset development and evaluation setup. Additionally, we provide a
brief overview of the systems submitted by the participants. All datasets and
evaluation scripts from the shared task have been made publicly available for
the research community, to foster further research in this domain. | cs.CL | BLP-2023 Task 2: Sentiment Analysis |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.