publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2024-03-01T10:03:07Z | http://arxiv.org/pdf/2403.00417v1 | http://arxiv.org/abs/2403.00417v1 | Jinbiao Yang | 2024-03-01T10:03:07Z | cs.CL | Tokenization significantly influences language models(LMs)' performance. This
paper traces the evolution of tokenizers from word-level to subword-level,
analyzing how they balance tokens and types to enhance model adaptability while
controlling complexity. Despite subword tokenizers like Byte Pair Encoding
(BPE) overcoming many word tokenizer limitations, they encounter difficulties
in handling non-Latin languages and depend heavily on extensive training data
and computational resources to grasp the nuances of multiword expressions
(MWEs). This article argues that tokenizers, more than mere technical tools,
should drawing inspiration from the cognitive science about human language
processing. This study then introduces the "Principle of Least Effort" from
cognitive science, that humans naturally seek to reduce cognitive effort, and
discusses the benefits of this principle for tokenizer development. Based on
this principle, the paper proposes that the Less-is-Better (LiB) model could be
a new approach for LLM tokenizer. The LiB model can autonomously learn an
integrated vocabulary consisting of subwords, words, and MWEs, which
effectively reduces both the numbers of tokens and types. Comparative
evaluations show that the LiB tokenizer outperforms existing word and BPE
tokenizers, presenting an innovative method for tokenizer development, and
hinting at the possibility of future cognitive science-based tokenizers being
more efficient. | cs.CL | Rethinking Tokenization: Crafting Better Tokenizers for Large Language Models |
2024-03-01T09:57:46Z | http://arxiv.org/pdf/2403.00411v2 | http://arxiv.org/abs/2403.00411v2 | Recep Firat Cekinel, Pinar Karagoz, Cagri Coltekin | 2024-03-22T15:54:03Z | cs.CL | The rapid spread of misinformation through social media platforms has raised
concerns regarding its impact on public opinion. While misinformation is
prevalent in other languages, the majority of research in this field has
concentrated on the English language. Hence, there is a scarcity of datasets
for other languages, including Turkish. To address this concern, we have
introduced the FCTR dataset, consisting of 3238 real-world claims. This dataset
spans multiple domains and incorporates evidence collected from three Turkish
fact-checking organizations. Additionally, we aim to assess the effectiveness
of cross-lingual transfer learning for low-resource languages, with a
particular focus on Turkish. We demonstrate in-context learning (zero-shot and
few-shot) performance of large language models in this context. The
experimental results indicate that the dataset has the potential to advance
research in the Turkish language. | cs.CL | Cross-Lingual Learning vs. Low-Resource Fine-Tuning: A Case Study with Fact-Checking in Turkish |
2024-03-01T09:55:18Z | http://arxiv.org/pdf/2403.00409v2 | http://arxiv.org/abs/2403.00409v2 | Sayak Ray Chowdhury, Anush Kini, Nagarajan Natarajan | 2024-04-12T01:09:37Z | cs.LG, cs.CL | Learning from preference-based feedback has recently gained traction as a
promising approach to align language models with human interests. While these
aligned generative models have demonstrated impressive capabilities across
various tasks, their dependence on high-quality human preference data poses a
bottleneck in practical applications. Specifically, noisy (incorrect and
ambiguous) preference pairs in the dataset might restrict the language models
from capturing human intent accurately. While practitioners have recently
proposed heuristics to mitigate the effect of noisy preferences, a complete
theoretical understanding of their workings remain elusive.
In this work, we aim to bridge this gap by by introducing a general framework
for policy optimization in the presence of random preference flips. We focus on
the direct preference optimization (DPO) algorithm in particular since it
assumes that preferences adhere to the Bradley-Terry-Luce (BTL) model, raising
concerns about the impact of noisy data on the learned policy. We design a
novel loss function, which de-bias the effect of noise on average, making a
policy trained by minimizing that loss robust to the noise. Under log-linear
parameterization of the policy class and assuming good feature coverage of the
SFT policy, we prove that the sub-optimality gap of the proposed robust DPO
(rDPO) policy compared to the optimal policy is of the order
$O(\frac{1}{1-2\epsilon}\sqrt{\frac{d}{n}})$, where $\epsilon < 1/2$ is flip
rate of labels, $d$ is policy parameter dimension and $n$ is size of dataset.
Our experiments on IMDb sentiment generation and Anthropic's helpful-harmless
dataset show that rDPO is robust to noise in preference labels compared to
vanilla DPO and other heuristics proposed by practitioners. | cs.LG | Provably Robust DPO: Aligning Language Models with Noisy Feedback |
2024-03-01T09:28:38Z | http://arxiv.org/pdf/2403.00393v2 | http://arxiv.org/abs/2403.00393v2 | Tanmay Rajore, Nishanth Chandran, Sunayana Sitaram, Divya Gupta, Rahul Sharma, Kashish Mittal, Manohar Swaminathan | 2024-06-24T08:28:18Z | cs.CR, cs.CL | Benchmarking is the de-facto standard for evaluating LLMs, due to its speed,
replicability and low cost. However, recent work has pointed out that the
majority of the open source benchmarks available today have been contaminated
or leaked into LLMs, meaning that LLMs have access to test data during
pretraining and/or fine-tuning. This raises serious concerns about the validity
of benchmarking studies conducted so far and the future of evaluation using
benchmarks. To solve this problem, we propose Private Benchmarking, a solution
where test datasets are kept private and models are evaluated without revealing
the test data to the model. We describe various scenarios (depending on the
trust placed on model owners or dataset owners), and present solutions to avoid
data contamination using private benchmarking. For scenarios where the model
weights need to be kept private, we describe solutions from confidential
computing and cryptography that can aid in private benchmarking. We build an
end-to-end system, TRUCE, that enables such private benchmarking showing that
the overheads introduced to protect models and benchmark are negligible (in the
case of confidential computing) and tractable (when cryptographic security is
required). Finally, we also discuss solutions to the problem of benchmark
dataset auditing, to ensure that private benchmarks are of sufficiently high
quality. | cs.CR | TRUCE: Private Benchmarking to Prevent Contamination and Improve Comparative Evaluation of LLMs |
2024-03-01T08:53:52Z | http://arxiv.org/pdf/2403.00370v1 | http://arxiv.org/abs/2403.00370v1 | Heyang Liu, Yu Wang, Yanfeng Wang | 2024-03-01T08:53:52Z | cs.CL, cs.SD, eess.AS | End-to-end (E2E) approach is gradually replacing hybrid models for automatic
speech recognition (ASR) tasks. However, the optimization of E2E models lacks
an intuitive method for handling decoding shifts, especially in scenarios with
a large number of domain-specific rare words that hold specific important
meanings. Furthermore, the absence of knowledge-intensive speech datasets in
academia has been a significant limiting factor, and the commonly used speech
corpora exhibit significant disparities with realistic conversation. To address
these challenges, we present Medical Interview (MED-IT), a multi-turn
consultation speech dataset that contains a substantial number of
knowledge-intensive named entities. We also explore methods to enhance the
recognition performance of rare words for E2E models. We propose a novel
approach, post-decoder biasing, which constructs a transform probability matrix
based on the distribution of training transcriptions. This guides the model to
prioritize recognizing words in the biasing list. In our experiments, for
subsets of rare words appearing in the training speech between 10 and 20 times,
and between 1 and 5 times, the proposed method achieves a relative improvement
of 9.3% and 5.1%, respectively. | cs.CL | Post-decoder Biasing for End-to-End Speech Recognition of Multi-turn Medical Interview |
2024-03-01T08:34:02Z | http://arxiv.org/pdf/2403.00354v2 | http://arxiv.org/abs/2403.00354v2 | Jieyong Kim, Ryang Heo, Yongsik Seo, SeongKu Kang, Jinyoung Yeo, Dongha Lee | 2024-06-08T08:41:53Z | cs.CL | In the task of aspect sentiment quad prediction (ASQP), generative methods
for predicting sentiment quads have shown promising results. However, they
still suffer from imprecise predictions and limited interpretability, caused by
data scarcity and inadequate modeling of the quadruplet composition process. In
this paper, we propose Self-Consistent Reasoning-based Aspect-sentiment
quadruple Prediction (SCRAP), optimizing its model to generate reasonings and
the corresponding sentiment quadruplets in sequence. SCRAP adopts the
Extract-Then-Assign reasoning strategy, which closely mimics human cognition.
In the end, SCRAP significantly improves the model's ability to handle complex
reasoning tasks and correctly predict quadruplets through consistency voting,
resulting in enhanced interpretability and accuracy in ASQP. | cs.CL | Self-Consistent Reasoning-based Aspect-Sentiment Quad Prediction with Extract-Then-Assign Strategy |
2024-03-01T08:13:48Z | http://arxiv.org/pdf/2403.00876v1 | http://arxiv.org/abs/2403.00876v1 | Qinghua Zhao, Vinit Ravishankar, Nicolas Garneau, Anders Søgaard | 2024-03-01T08:13:48Z | cs.CL, cs.AI | Word order is an important concept in natural language, and in this work, we
study how word order affects the induction of world knowledge from raw text
using language models. We use word analogies to probe for such knowledge.
Specifically, in addition to the natural word order, we first respectively
extract texts of six fixed word orders from five languages and then pretrain
the language models on these texts. Finally, we analyze the experimental
results of the fixed word orders on word analogies and show that i) certain
fixed word orders consistently outperform or underperform others, though the
specifics vary across languages, and ii) the Wov2Lex hypothesis is not hold in
pre-trained language models, and the natural word order typically yields
mediocre results. The source code will be made publicly available at
https://github.com/lshowway/probing_by_analogy. | cs.CL | Word Order and World Knowledge |
2024-03-01T08:05:44Z | http://arxiv.org/pdf/2403.00338v1 | http://arxiv.org/abs/2403.00338v1 | Xianzhen Luo, Qingfu Zhu, Zhiming Zhang, Xu Wang, Qing Yang, Dongliang Xu, Wanxiang Che | 2024-03-01T08:05:44Z | cs.CL | Instruction tuning plays a pivotal role in Code Large Language Models (Code
LLMs) for the task of program synthesis. Presently, two dominant paradigms for
collecting tuning data are natural-instruct (human-written) and self-instruct
(automatically generated). Natural-instruct includes diverse and correct codes
but lacks instruction-code pairs, and exists improper code formats like nested
single-line codes. In contrast, self-instruct automatically generates proper
paired data. However, it suffers from low diversity due to generating
duplicates and cannot ensure the correctness of codes. To bridge the both
paradigms, we propose \textbf{Semi-Instruct}. It first converts diverse but
improper codes from natural-instruct into proper instruction-code pairs through
a method similar to self-instruct. To verify the correctness of generated
codes, we design a novel way to construct test cases by generating cases'
inputs and executing correct codes from natural-instruct to get outputs.
Finally, diverse and correct instruction-code pairs are retained for
instruction tuning. Experiments show that semi-instruct is significantly better
than natural-instruct and self-instruct. Furthermore, the performance steadily
improves as data scale increases. | cs.CL | Semi-Instruct: Bridging Natural-Instruct and Self-Instruct for Code Large Language Models |
2024-03-01T06:15:07Z | http://arxiv.org/pdf/2403.00871v1 | http://arxiv.org/abs/2403.00871v1 | Ashwinee Panda, Christopher A. Choquette-Choo, Zhengming Zhang, Yaoqing Yang, Prateek Mittal | 2024-03-01T06:15:07Z | cs.CR, cs.AI, cs.CL, cs.LG | When large language models are trained on private data, it can be a
significant privacy risk for them to memorize and regurgitate sensitive
information. In this work, we propose a new practical data extraction attack
that we call "neural phishing". This attack enables an adversary to target and
extract sensitive or personally identifiable information (PII), e.g., credit
card numbers, from a model trained on user data with upwards of 10% attack
success rates, at times, as high as 50%. Our attack assumes only that an
adversary can insert as few as 10s of benign-appearing sentences into the
training dataset using only vague priors on the structure of the user data. | cs.CR | Teach LLMs to Phish: Stealing Private Information from Language Models |
2024-03-01T05:28:06Z | http://arxiv.org/pdf/2403.00292v1 | http://arxiv.org/abs/2403.00292v1 | Xu Zhang, Xiaojun Wan | 2024-03-01T05:28:06Z | cs.CL | Language models risk generating mindless and offensive content, which hinders
their safe deployment. Therefore, it is crucial to discover and modify
potential toxic outputs of pre-trained language models before deployment. In
this work, we elicit toxic content by automatically searching for a prompt that
directs pre-trained language models towards the generation of a specific target
output. The problem is challenging due to the discrete nature of textual data
and the considerable computational resources required for a single forward pass
of the language model. To combat these challenges, we introduce Auto-regressive
Selective Replacement Ascent (ASRA), a discrete optimization algorithm that
selects prompts based on both quality and similarity with determinantal point
process (DPP). Experimental results on six different pre-trained language
models demonstrate the efficacy of ASRA for eliciting toxic content.
Furthermore, our analysis reveals a strong correlation between the success rate
of ASRA attacks and the perplexity of target outputs, while indicating limited
association with the quantity of model parameters. | cs.CL | DPP-Based Adversarial Prompt Searching for Lanugage Models |
2024-03-01T04:47:16Z | http://arxiv.org/pdf/2403.00277v1 | http://arxiv.org/abs/2403.00277v1 | Jinman Zhao, Yitian Ding, Chen Jia, Yining Wang, Zifan Qian | 2024-03-01T04:47:16Z | cs.CL | With the growing deployment of large language models (LLMs) across various
applications, assessing the influence of gender biases embedded in LLMs becomes
crucial. The topic of gender bias within the realm of natural language
processing (NLP) has gained considerable focus, particularly in the context of
English. Nonetheless, the investigation of gender bias in languages other than
English is still relatively under-explored and insufficiently analyzed. In this
work, We examine gender bias in LLMs-generated outputs for different languages.
We use three measurements: 1) gender bias in selecting descriptive words given
the gender-related context. 2) gender bias in selecting gender-related pronouns
(she/he) given the descriptive words. 3) gender bias in the topics of
LLM-generated dialogues. We investigate the outputs of the GPT series of LLMs
in various languages using our three measurement methods. Our findings revealed
significant gender biases across all the languages we examined. | cs.CL | Gender Bias in Large Language Models across Multiple Languages |
2024-03-01T04:39:16Z | http://arxiv.org/pdf/2403.00868v2 | http://arxiv.org/abs/2403.00868v2 | Ye Chen, Igor Couto, Wei Cai, Cong Fu, Bruno Dorneles | 2024-03-27T03:03:00Z | cs.CL, cs.AI | We introduce SoftTiger, a clinical large language model (CLaM) designed as a
foundation model for healthcare workflows. The narrative and unstructured
nature of clinical notes is a major obstacle for healthcare intelligentization.
We address a critical problem of structuring clinical notes into clinical data,
according to international interoperability standards. We collect and annotate
data for three subtasks, namely, international patient summary, clinical
impression and medical encounter. We then supervised fine-tuned a
state-of-the-art LLM using public and credentialed clinical data. The training
is orchestrated in a way that the target model can first support basic clinical
tasks such as abbreviation expansion and temporal information extraction, and
then learn to perform more complex downstream clinical tasks. Moreover, we
address several modeling challenges in the healthcare context, e.g., extra long
context window. Our blind pairwise evaluation shows that SoftTiger outperforms
other popular open-source models and GPT-3.5, comparable to Gemini-pro, with a
mild gap from GPT-4. We believe that LLMs may become a step-stone towards
healthcare digitalization and democratization. Therefore, we publicly release
SoftTiger models at scales of 13 billion and 70 billion parameters, as well as
datasets and code for our innovative scalable evaluation, hopefully, making a
significant contribution to the healthcare industry. | cs.CL | SoftTiger: A Clinical Foundation Model for Healthcare Workflows |
2024-03-01T03:51:56Z | http://arxiv.org/pdf/2403.00260v1 | http://arxiv.org/abs/2403.00260v1 | Ghazal Khalighinejad, Defne Circi, L. C. Brinson, Bhuwan Dhingra | 2024-03-01T03:51:56Z | cs.CL | This paper investigates the use of large language models (LLMs) for
extracting sample lists of polymer nanocomposites (PNCs) from full-length
materials science research papers. The challenge lies in the complex nature of
PNC samples, which have numerous attributes scattered throughout the text. The
complexity of annotating detailed information on PNCs limits the availability
of data, making conventional document-level relation extraction techniques
impractical due to the challenge in creating comprehensive named entity span
annotations. To address this, we introduce a new benchmark and an evaluation
technique for this task and explore different prompting strategies in a
zero-shot manner. We also incorporate self-consistency to improve the
performance. Our findings show that even advanced LLMs struggle to extract all
of the samples from an article. Finally, we analyze the errors encountered in
this process, categorizing them into three main challenges, and discuss
potential strategies for future research to overcome them. | cs.CL | Extracting Polymer Nanocomposite Samples from Full-Length Documents |
2024-03-01T03:30:38Z | http://arxiv.org/pdf/2403.00252v2 | http://arxiv.org/abs/2403.00252v2 | Olivier Salaün, Frédéric Piedboeuf, Guillaume Le Berre, David Alfonso Hermelo, Philippe Langlais | 2024-06-14T13:51:01Z | cs.CL, cs.AI, cs.LG | Keyphrase generation has primarily been explored within the context of
academic research articles, with a particular focus on scientific domains and
the English language. In this work, we present EUROPA, a dataset for
multilingual keyphrase generation in the legal domain. It is derived from legal
judgments from the Court of Justice of the European Union (EU), and contains
instances in all 24 EU official languages. We run multilingual models on our
corpus and analyze the results, showing room for improvement on a
domain-specific multilingual corpus such as the one we present. | cs.CL | EUROPA: A Legal Multilingual Keyphrase Generation Dataset |
2024-03-01T03:29:54Z | http://arxiv.org/pdf/2403.00867v2 | http://arxiv.org/abs/2403.00867v2 | Xiaomeng Hu, Pin-Yu Chen, Tsung-Yi Ho | 2024-03-05T13:46:50Z | cs.CR, cs.AI, cs.CL, cs.LG | Large Language Models (LLMs) are becoming a prominent generative AI tool,
where the user enters a query and the LLM generates an answer. To reduce harm
and misuse, efforts have been made to align these LLMs to human values using
advanced training techniques such as Reinforcement Learning from Human Feedback
(RLHF). However, recent studies have highlighted the vulnerability of LLMs to
adversarial jailbreak attempts aiming at subverting the embedded safety
guardrails. To address this challenge, this paper defines and investigates the
Refusal Loss of LLMs and then proposes a method called Gradient Cuff to detect
jailbreak attempts. Gradient Cuff exploits the unique properties observed in
the refusal loss landscape, including functional values and its smoothness, to
design an effective two-step detection strategy. Experimental results on two
aligned LLMs (LLaMA-2-7B-Chat and Vicuna-7B-V1.5) and six types of jailbreak
attacks (GCG, AutoDAN, PAIR, TAP, Base64, and LRL) show that Gradient Cuff can
significantly improve the LLM's rejection capability for malicious jailbreak
queries, while maintaining the model's performance for benign user queries by
adjusting the detection threshold. | cs.CR | Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes |
2024-03-01T03:07:32Z | http://arxiv.org/pdf/2403.00241v2 | http://arxiv.org/abs/2403.00241v2 | Leane Jourdan, Florian Boudin, Nicolas Hernandez, Richard Dufour | 2024-03-19T08:34:05Z | cs.CL | Writing a scientific article is a challenging task as it is a highly codified
and specific genre, consequently proficiency in written communication is
essential for effectively conveying research findings and ideas. In this
article, we propose an original textual resource on the revision step of the
writing process of scientific articles. This new dataset, called CASIMIR,
contains the multiple revised versions of 15,646 scientific articles from
OpenReview, along with their peer reviews. Pairs of consecutive versions of an
article are aligned at sentence-level while keeping paragraph location
information as metadata for supporting future revision studies at the discourse
level. Each pair of revised sentences is enriched with automatically extracted
edits and associated revision intention. To assess the initial quality on the
dataset, we conducted a qualitative study of several state-of-the-art text
revision approaches and compared various evaluation metrics. Our experiments
led us to question the relevance of the current evaluation methods for the text
revision task. | cs.CL | CASIMIR: A Corpus of Scientific Articles enhanced with Multiple Author-Integrated Revisions |
2024-03-01T02:33:26Z | http://arxiv.org/pdf/2403.00236v1 | http://arxiv.org/abs/2403.00236v1 | Rachith Aiyappa, Shruthi Senthilmani, Jisun An, Haewoon Kwak, Yong-Yeol Ahn | 2024-03-01T02:33:26Z | cs.CL, cs.AI, cs.LG | We investigate the performance of LLM-based zero-shot stance detection on
tweets. Using FlanT5-XXL, an instruction-tuned open-source LLM, with the
SemEval 2016 Tasks 6A, 6B, and P-Stance datasets, we study the performance and
its variations under different prompts and decoding strategies, as well as the
potential biases of the model. We show that the zero-shot approach can match or
outperform state-of-the-art benchmarks, including fine-tuned models. We provide
various insights into its performance including the sensitivity to instructions
and prompts, the decoding strategies, the perplexity of the prompts, and to
negations and oppositions present in prompts. Finally, we ensure that the LLM
has not been trained on test datasets, and identify a positivity bias which may
partially explain the performance differences across decoding strategie | cs.CL | Benchmarking zero-shot stance detection with FlanT5-XXL: Insights from training data, prompting, and decoding strategies into its near-SoTA performance |
2024-03-01T02:21:30Z | http://arxiv.org/pdf/2403.00231v3 | http://arxiv.org/abs/2403.00231v3 | Lei Li, Yuqi Wang, Runxin Xu, Peiyi Wang, Xiachong Feng, Lingpeng Kong, Qi Liu | 2024-06-02T15:47:16Z | cs.CV, cs.CL | Large vision-language models (LVLMs) excel across diverse tasks involving
concrete images from natural scenes. However, their ability to interpret
abstract figures, such as geometry shapes and scientific plots, remains limited
due to a scarcity of training datasets in scientific domains. To fill this gap,
we introduce Multimodal ArXiv, consisting of ArXivCap and ArXivQA, for
enhancing LVLMs scientific comprehension. ArXivCap is a figure-caption dataset
comprising 6.4M images and 3.9M captions, sourced from 572K ArXiv papers
spanning various scientific domains. Drawing from ArXivCap, we introduce
ArXivQA, a question-answering dataset generated by prompting GPT-4V based on
scientific figures. ArXivQA greatly enhances open-sourced LVLMs' mathematical
reasoning capabilities, achieving a 10.4\% absolute accuracy gain on a
multimodal mathematical reasoning benchmark. Furthermore, employing ArXivCap,
we devise four vision-to-text tasks for benchmarking LVLMs. Evaluation results
with state-of-the-art LVLMs underscore their struggle with the nuanced
semantics of academic figures, while domain-specific training yields
substantial performance gains. Our error analysis uncovers misinterpretations
of visual context, recognition errors, and the production of overly simplified
captions by current LVLMs, shedding light on future improvements. | cs.CV | Multimodal ArXiv: A Dataset for Improving Scientific Comprehension of Large Vision-Language Models |
2024-03-01T02:09:25Z | http://arxiv.org/pdf/2403.00226v3 | http://arxiv.org/abs/2403.00226v3 | Taichi Aida, Danushka Bollegala | 2024-06-01T09:23:22Z | cs.CL | Detecting temporal semantic changes of words is an important task for various
NLP applications that must make time-sensitive predictions. Lexical Semantic
Change Detection (SCD) task involves predicting whether a given target word,
$w$, changes its meaning between two different text corpora, $C_1$ and $C_2$.
For this purpose, we propose a supervised two-staged SCD method that uses
existing Word-in-Context (WiC) datasets. In the first stage, for a target word
$w$, we learn two sense-aware encoders that represent the meaning of $w$ in a
given sentence selected from a corpus. Next, in the second stage, we learn a
sense-aware distance metric that compares the semantic representations of a
target word across all of its occurrences in $C_1$ and $C_2$. Experimental
results on multiple benchmark datasets for SCD show that our proposed method
achieves strong performance in multiple languages. Additionally, our method
achieves significant improvements on WiC benchmarks compared to a sense-aware
encoder with conventional distance functions. Source code is available at
https://github.com/LivNLP/svp-sdml . | cs.CL | A Semantic Distance Metric Learning approach for Lexical Semantic Change Detection |
2024-03-01T01:15:45Z | http://arxiv.org/pdf/2403.00212v1 | http://arxiv.org/abs/2403.00212v1 | Aniket Tathe, Anand Kamble, Suyash Kumbharkar, Atharva Bhandare, Anirban C. Mitra | 2024-03-01T01:15:45Z | cs.CL, cs.CV, cs.LG, cs.SD, eess.AS | This research addresses the challenge of training an ASR model for
personalized voices with minimal data. Utilizing just 14 minutes of custom
audio from a YouTube video, we employ Retrieval-Based Voice Conversion (RVC) to
create a custom Common Voice 16.0 corpus. Subsequently, a Cross-lingual
Self-supervised Representations (XLSR) Wav2Vec2 model is fine-tuned on this
dataset. The developed web-based GUI efficiently transcribes and translates
input Hindi videos. By integrating XLSR Wav2Vec2 and mBART, the system aligns
the translated text with the video timeline, delivering an accessible solution
for multilingual video content transcription and translation for personalized
voice. | cs.CL | Transcription and translation of videos using fine-tuned XLSR Wav2Vec2 on custom dataset and mBART |
2024-03-01T00:08:20Z | http://arxiv.org/pdf/2403.00199v3 | http://arxiv.org/abs/2403.00199v3 | Nischal Ashok Kumar, Andrew Lan | 2024-04-19T02:36:26Z | cs.CL, cs.CY, cs.LG | The Socratic method is a way of guiding students toward solving a problem
independently without directly revealing the solution to the problem. Although
this method has been shown to significantly improve student learning outcomes,
it remains a complex labor-intensive task for instructors. Large language
models (LLMs) can be used to augment human effort by automatically generating
Socratic questions for students. However, existing methods that involve
prompting these LLMs sometimes produce invalid outputs, e.g., those that
directly reveal the solution to the problem or provide irrelevant or premature
questions. To alleviate this problem, inspired by reinforcement learning with
AI feedback (RLAIF), we first propose a data augmentation method to enrich
existing Socratic questioning datasets with questions that are invalid in
specific ways. Next, we propose a method to optimize open-source LLMs such as
LLama 2 to prefer ground-truth questions over generated invalid ones, using
direct preference optimization (DPO). Our experiments on a Socratic questions
dataset for student code debugging show that a DPO-optimized 7B LLama 2 model
can effectively avoid generating invalid questions, and as a result,
outperforms existing state-of-the-art prompting methods. | cs.CL | Improving Socratic Question Generation using Data Augmentation and Preference Optimization |
2024-03-01T00:02:37Z | http://arxiv.org/pdf/2403.00198v1 | http://arxiv.org/abs/2403.00198v1 | Sana Ebrahimi, Kaiwen Chen, Abolfazl Asudeh, Gautam Das, Nick Koudas | 2024-03-01T00:02:37Z | cs.CL, cs.AI, cs.CY, cs.LG | Pre-trained Large Language Models (LLMs) have significantly advanced natural
language processing capabilities but are susceptible to biases present in their
training data, leading to unfair outcomes in various applications. While
numerous strategies have been proposed to mitigate bias, they often require
extensive computational resources and may compromise model performance. In this
work, we introduce AXOLOTL, a novel post-processing framework, which operates
agnostically across tasks and models, leveraging public APIs to interact with
LLMs without direct access to internal parameters. Through a three-step process
resembling zero-shot learning, AXOLOTL identifies biases, proposes resolutions,
and guides the model to self-debias its outputs. This approach minimizes
computational costs and preserves model performance, making AXOLOTL a promising
tool for debiasing LLM outputs with broad applicability and ease of use. | cs.CL | AXOLOTL: Fairness through Assisted Self-Debiasing of Large Language Model Outputs |
2024-02-29T23:11:55Z | http://arxiv.org/pdf/2403.00180v2 | http://arxiv.org/abs/2403.00180v2 | Karina Halevy, Anna Sotnikova, Badr AlKhamissi, Syrielle Montariol, Antoine Bosselut | 2024-06-16T18:08:11Z | cs.CL | Model editing has emerged as a cost-effective strategy to update knowledge
stored in language models. However, model editing can have unintended
consequences after edits are applied: information unrelated to the edits can
also be changed, and other general behaviors of the model can be wrongly
altered. In this work, we investigate how model editing methods unexpectedly
amplify model biases post-edit. We introduce a novel benchmark dataset,
Seesaw-CF, for measuring bias-related harms of model editing and conduct the
first in-depth investigation of how different weight-editing methods impact
model bias. Specifically, we focus on biases with respect to demographic
attributes such as race, geographic origin, and gender, as well as qualitative
flaws in long-form texts generated by edited language models. We find that
edited models exhibit, to various degrees, more biased behavior as they become
less confident in attributes for Asian, African, and South American subjects.
Furthermore, edited models amplify sexism and xenophobia in text generations
while remaining seemingly coherent and logical. Finally, editing facts about
place of birth, country of citizenship, or gender have particularly negative
effects on the model's knowledge about unrelated features like field of work. | cs.CL | "Flex Tape Can't Fix That": Bias and Misinformation in Edited Language Models |
2024-02-29T23:03:19Z | http://arxiv.org/pdf/2403.00863v2 | http://arxiv.org/abs/2403.00863v2 | Chenhao Fang, Xiaohan Li, Zezhong Fan, Jianpeng Xu, Kaushiki Nag, Evren Korpeoglu, Sushant Kumar, Kannan Achan | 2024-06-20T07:10:28Z | cs.IR, cs.AI, cs.CL | Product attribute value extraction is a pivotal component in Natural Language
Processing (NLP) and the contemporary e-commerce industry. The provision of
precise product attribute values is fundamental in ensuring high-quality
recommendations and enhancing customer satisfaction. The recently emerging
Large Language Models (LLMs) have demonstrated state-of-the-art performance in
numerous attribute extraction tasks, without the need for domain-specific
training data. Nevertheless, varying strengths and weaknesses are exhibited by
different LLMs due to the diversity in data, architectures, and
hyperparameters. This variation makes them complementary to each other, with no
single LLM dominating all others. Considering the diverse strengths and
weaknesses of LLMs, it becomes necessary to develop an ensemble method that
leverages their complementary potentials. In this paper, we propose a novel
algorithm called LLM-ensemble to ensemble different LLMs' outputs for attribute
value extraction. We iteratively learn the weights for different LLMs to
aggregate the labels with weights to predict the final attribute value. Not
only can our proposed method be proven theoretically optimal, but it also
ensures efficient computation, fast convergence, and safe deployment. We have
also conducted extensive experiments with various state-of-the-art LLMs,
including Llama2-13B, Llama2-70B, PaLM-2, GPT-3.5, and GPT-4, on Walmart's
internal data. Our offline metrics demonstrate that the LLM-ensemble method
outperforms all the state-of-the-art single LLMs on Walmart's internal dataset.
This method has been launched in several production models, leading to improved
Gross Merchandise Volume (GMV), Click-Through Rate (CTR), Conversion Rate
(CVR), and Add-to-Cart Rate (ATC). | cs.IR | LLM-Ensemble: Optimal Large Language Model Ensemble Method for E-commerce Product Attribute Value Extraction |
2024-02-29T22:26:07Z | http://arxiv.org/pdf/2403.00165v2 | http://arxiv.org/abs/2403.00165v2 | Yunyi Zhang, Ruozhen Yang, Xueqiang Xu, Rui Li, Jinfeng Xiao, Jiaming Shen, Jiawei Han | 2024-06-16T19:10:39Z | cs.CL, cs.LG | Hierarchical text classification aims to categorize each document into a set
of classes in a label taxonomy. Most earlier works focus on fully or
semi-supervised methods that require a large amount of human annotated data
which is costly and time-consuming to acquire. To alleviate human efforts, in
this paper, we work on hierarchical text classification with the minimal amount
of supervision: using the sole class name of each node as the only supervision.
Recently, large language models (LLM) show competitive performance on various
tasks through zero-shot prompting, but this method performs poorly in the
hierarchical setting, because it is ineffective to include the large and
structured label space in a prompt. On the other hand, previous
weakly-supervised hierarchical text classification methods only utilize the raw
taxonomy skeleton and ignore the rich information hidden in the text corpus
that can serve as additional class-indicative features. To tackle the above
challenges, we propose TELEClass, Taxonomy Enrichment and LLM-Enhanced
weakly-supervised hierarchical text Classification, which (1) automatically
enriches the label taxonomy with class-indicative terms to facilitate
classifier training and (2) utilizes LLMs for both data annotation and creation
tailored for the hierarchical label space. Experiments show that TELEClass can
outperform previous weakly-supervised methods and LLM-based zero-shot prompting
methods on two public datasets. | cs.CL | TELEClass: Taxonomy Enrichment and LLM-Enhanced Hierarchical Text Classification with Minimal Supervision |
2024-02-29T21:49:31Z | http://arxiv.org/pdf/2403.00143v1 | http://arxiv.org/abs/2403.00143v1 | Behzad Shayegh, Yuqiao Wen, Lili Mou | 2024-02-29T21:49:31Z | cs.CL, cs.AI, cs.LG | We address unsupervised discontinuous constituency parsing, where we observe
a high variance in the performance of the only previous model. We propose to
build an ensemble of different runs of the existing discontinuous parser by
averaging the predicted trees, to stabilize and boost performance. To begin
with, we provide comprehensive computational complexity analysis (in terms of P
and NP-complete) for tree averaging under different setups of binarity and
continuity. We then develop an efficient exact algorithm to tackle the task,
which runs in a reasonable time for all samples in our experiments. Results on
three datasets show our method outperforms all baselines in all metrics; we
also provide in-depth analyses of our approach. | cs.CL | Ensemble-Based Unsupervised Discontinuous Constituency Parsing by Tree Averaging |
2024-02-29T21:49:31Z | http://arxiv.org/pdf/2403.00144v1 | http://arxiv.org/abs/2403.00144v1 | Yuqiao Wen, Behzad Shayegh, Chenyang Huang, Yanshuai Cao, Lili Mou | 2024-02-29T21:49:31Z | cs.CL, cs.AI, cs.LG, I.2.7; I.2.6; I.2.m; I.5.1; I.7.m | The ability of zero-shot translation emerges when we train a multilingual
model with certain translation directions; the model can then directly
translate in unseen directions. Alternatively, zero-shot translation can be
accomplished by pivoting through a third language (e.g., English). In our work,
we observe that both direct and pivot translations are noisy and achieve less
satisfactory performance. We propose EBBS, an ensemble method with a novel
bi-level beam search algorithm, where each ensemble component explores its own
prediction step by step at the lower level but they are synchronized by a "soft
voting" mechanism at the upper level. Results on two popular multilingual
translation datasets show that EBBS consistently outperforms direct and pivot
translations as well as existing ensemble techniques. Further, we can distill
the ensemble's knowledge back to the multilingual model to improve inference
efficiency; profoundly, our EBBS-based distillation does not sacrifice, or even
improves, the translation quality. | cs.CL | EBBS: An Ensemble with Bi-Level Beam Search for Zero-Shot Machine Translation |
2024-02-29T21:44:50Z | http://arxiv.org/pdf/2403.00141v1 | http://arxiv.org/abs/2403.00141v1 | Joykirat Singh, Sehban Fazili, Rohan Jain, Md Shad Akhtar | 2024-02-29T21:44:50Z | cs.CL, cs.AI | Privacy policy documents have a crucial role in educating individuals about
the collection, usage, and protection of users' personal data by organizations.
However, they are notorious for their lengthy, complex, and convoluted language
especially involving privacy-related entities. Hence, they pose a significant
challenge to users who attempt to comprehend organization's data usage policy.
In this paper, we propose to enhance the interpretability and readability of
policy documents by using controlled abstractive summarization -- we enforce
the generated summaries to include critical privacy-related entities (e.g.,
data and medium) and organization's rationale (e.g.,target and reason) in
collecting those entities. To achieve this, we develop PD-Sum, a
policy-document summarization dataset with marked privacy-related entity
labels. Our proposed model, EROS, identifies critical entities through a
span-based entity extraction model and employs them to control the information
content of the summaries using proximal policy optimization (PPO). Comparison
shows encouraging improvement over various baselines. Furthermore, we furnish
qualitative and human evaluations to establish the efficacy of EROS. | cs.CL | EROS: Entity-Driven Controlled Policy Document Summarization |
2024-02-29T21:05:38Z | http://arxiv.org/pdf/2403.00127v2 | http://arxiv.org/abs/2403.00127v2 | Sui He | 2024-04-28T09:45:58Z | cs.CL, cs.CY, cs.HC | Prompt engineering has shown potential for improving translation quality in
LLMs. However, the possibility of using translation concepts in prompt design
remains largely underexplored. Against this backdrop, the current paper
discusses the effectiveness of incorporating the conceptual tool of translation
brief and the personas of translator and author into prompt design for
translation tasks in ChatGPT. Findings suggest that, although certain elements
are constructive in facilitating human-to-human communication for translation
tasks, their effectiveness is limited for improving translation quality in
ChatGPT. This accentuates the need for explorative research on how translation
theorists and practitioners can develop the current set of conceptual tools
rooted in the human-to-human communication paradigm for translation purposes in
this emerging workflow involving human-machine interaction, and how translation
concepts developed in translation studies can inform the training of GPT models
for translation tasks. | cs.CL | Prompting ChatGPT for Translation: A Comparative Analysis of Translation Brief and Persona Prompts |
2024-02-29T21:05:37Z | http://arxiv.org/pdf/2403.00126v1 | http://arxiv.org/abs/2403.00126v1 | Xiaoqiang Wang, Bang Liu, Lingfei Wu | 2024-02-29T21:05:37Z | cs.CL | Large language models (LLMs) are primarily evaluated by overall performance
on various text understanding and generation tasks. However, such a paradigm
fails to comprehensively differentiate the fine-grained language and cognitive
skills, rendering the lack of sufficient interpretation to LLMs' capabilities.
In this paper, we present FAC$^2$E, a framework for Fine-grAined and
Cognition-grounded LLMs' Capability Evaluation. Specifically, we formulate
LLMs' evaluation in a multi-dimensional and explainable manner by dissociating
the language-related capabilities and the cognition-related ones. Besides,
through extracting the intermediate reasoning from LLMs, we further break down
the process of applying a specific capability into three sub-steps: recalling
relevant knowledge, utilizing knowledge, and solving problems. Finally,
FAC$^2$E evaluates each sub-step of each fine-grained capability, providing a
two-faceted diagnosis for LLMs. Utilizing FAC$^2$E, we identify a common
shortfall in knowledge utilization among models and propose a straightforward,
knowledge-enhanced method to mitigate this issue. Our results not only showcase
promising performance enhancements but also highlight a direction for future
LLM advancements. | cs.CL | FAC$^2$E: Better Understanding Large Language Model Capabilities by Dissociating Language and Cognition |
2024-02-29T21:05:14Z | http://arxiv.org/pdf/2403.00862v4 | http://arxiv.org/abs/2403.00862v4 | Miao Li, Ming-Bin Chen, Bo Tang, Shengbin Hou, Pengyu Wang, Haiying Deng, Zhiyu Li, Feiyu Xiong, Keming Mao, Peng Cheng, Yi Luo | 2024-06-04T14:50:58Z | cs.CL, cs.AI | We present NewsBench, a novel evaluation framework to systematically assess
the capabilities of Large Language Models (LLMs) for editorial capabilities in
Chinese journalism. Our constructed benchmark dataset is focused on four facets
of writing proficiency and six facets of safety adherence, and it comprises
manually and carefully designed 1,267 test samples in the types of multiple
choice questions and short answer questions for five editorial tasks in 24 news
domains. To measure performances, we propose different GPT-4 based automatic
evaluation protocols to assess LLM generations for short answer questions in
terms of writing proficiency and safety adherence, and both are validated by
the high correlations with human evaluations. Based on the systematic
evaluation framework, we conduct a comprehensive analysis of ten popular LLMs
which can handle Chinese. The experimental results highlight GPT-4 and ERNIE
Bot as top performers, yet reveal a relative deficiency in journalistic safety
adherence in creative writing tasks. Our findings also underscore the need for
enhanced ethical guidance in machine-generated journalistic content, marking a
step forward in aligning LLMs with journalistic standards and safety
considerations. | cs.CL | NewsBench: A Systematic Evaluation Framework for Assessing Editorial Capabilities of Large Language Models in Chinese Journalism |
2024-02-29T20:25:16Z | http://arxiv.org/pdf/2403.00108v1 | http://arxiv.org/abs/2403.00108v1 | Hongyi Liu, Zirui Liu, Ruixiang Tang, Jiayi Yuan, Shaochen Zhong, Yu-Neng Chuang, Li Li, Rui Chen, Xia Hu | 2024-02-29T20:25:16Z | cs.CR, cs.AI, cs.CL | Fine-tuning LLMs is crucial to enhancing their task-specific performance and
ensuring model behaviors are aligned with human preferences. Among various
fine-tuning methods, LoRA is popular for its efficiency and ease to use,
allowing end-users to easily post and adopt lightweight LoRA modules on
open-source platforms to tailor their model for different customization.
However, such a handy share-and-play setting opens up new attack surfaces, that
the attacker can render LoRA as an attacker, such as backdoor injection, and
widely distribute the adversarial LoRA to the community easily. This can result
in detrimental outcomes. Despite the huge potential risks of sharing LoRA
modules, this aspect however has not been fully explored. To fill the gap, in
this study we thoroughly investigate the attack opportunities enabled in the
growing share-and-play scenario. Specifically, we study how to inject backdoor
into the LoRA module and dive deeper into LoRA's infection mechanisms. We found
that training-free mechanism is possible in LoRA backdoor injection. We also
discover the impact of backdoor attacks with the presence of multiple LoRA
adaptions concurrently as well as LoRA based backdoor transferability. Our aim
is to raise awareness of the potential risks under the emerging share-and-play
scenario, so as to proactively prevent potential consequences caused by
LoRA-as-an-Attack. Warning: the paper contains potential offensive content
generated by models. | cs.CR | LoRA-as-an-Attack! Piercing LLM Safety Under The Share-and-Play Scenario |
2024-02-29T19:55:06Z | http://arxiv.org/pdf/2403.00858v4 | http://arxiv.org/abs/2403.00858v4 | Raghavv Goel, Mukul Gagrani, Wonseok Jeon, Junyoung Park, Mingu Lee, Christopher Lott | 2024-05-13T18:34:30Z | cs.LG, cs.AI, cs.CL | Text generation with Large Language Models (LLMs) is known to be memory bound
due to the combination of their auto-regressive nature, huge parameter counts,
and limited memory bandwidths, often resulting in low token rates. Speculative
decoding has been proposed as a solution for LLM inference acceleration.
However, since draft models are often unavailable in the modern open-source LLM
families, e.g., for Llama 2 7B, training a high-quality draft model is required
to enable inference acceleration via speculative decoding. In this paper, we
propose a simple draft model training framework for direct alignment to
chat-capable target models. With the proposed framework, we train Llama 2 Chat
Drafter 115M, a draft model for Llama 2 Chat 7B or larger, with only 1.64\% of
the original size. Our training framework only consists of pretraining,
distillation dataset generation, and finetuning with knowledge distillation,
with no additional alignment procedure. For the finetuning step, we use
instruction-response pairs generated by target model for distillation in
plausible data distribution, and propose a new Total Variation Distance++
(TVD++) loss that incorporates variance reduction techniques inspired from the
policy gradient method in reinforcement learning. Our empirical results show
that Llama 2 Chat Drafter 115M with speculative decoding achieves up to 2.3
block efficiency and 2.4$\times$ speed-up relative to autoregressive decoding
on various tasks with no further task-specific fine-tuning. | cs.LG | Direct Alignment of Draft Model for Speculative Decoding with Chat-Fine-Tuned LLMs |
2024-02-29T19:40:25Z | http://arxiv.org/pdf/2403.00092v2 | http://arxiv.org/abs/2403.00092v2 | Tianyi Zhang, Li Zhang, Zhaoyi Hou, Ziyu Wang, Yuling Gu, Peter Clark, Chris Callison-Burch, Niket Tandon | 2024-07-02T04:50:36Z | cs.CL | Planning in a text-based environment continues to be a major challenge for AI
systems. Recent approaches have used language models to predict a planning
domain definition (e.g., PDDL) but have only been evaluated in closed-domain
simulated environments. To address this, we present Proc2PDDL , the first
dataset containing open-domain procedural texts paired with expert-annotated
PDDL representations. Using this dataset, we evaluate state-of-the-art models
on defining the preconditions and effects of actions. We show that Proc2PDDL is
highly challenging, with GPT-3.5's success rate close to 0% and GPT-4's around
35%. Our analysis shows both syntactic and semantic errors, indicating LMs'
deficiency in both generating domain-specific prgorams and reasoning about
events. We hope this analysis and dataset helps future progress towards
integrating the best of LMs and formal planning. | cs.CL | PROC2PDDL: Open-Domain Planning Representations from Texts |
2024-02-29T19:02:03Z | http://arxiv.org/pdf/2403.00071v2 | http://arxiv.org/abs/2403.00071v2 | Suyuchen Wang, Ivan Kobyzev, Peng Lu, Mehdi Rezagholizadeh, Bang Liu | 2024-06-10T13:30:34Z | cs.CL, cs.AI | This paper addresses the challenge of train-short-test-long (TSTL) scenarios
in Large Language Models (LLMs) equipped with Rotary Position Embedding (RoPE),
where models pre-trained on shorter sequences face difficulty with
out-of-distribution (OOD) token positions in longer sequences. We introduce
Resonance RoPE, a novel approach designed to narrow the generalization gap in
TSTL scenarios by refining the interpolation of RoPE features for OOD
positions, significantly improving the model performance without additional
online computational costs. Furthermore, we present PosGen, a new synthetic
benchmark specifically designed for fine-grained behavior analysis in TSTL
scenarios, aiming to isolate the constantly increasing difficulty of token
generation on long contexts from the challenges of recognizing new token
positions. Our experiments on synthetic tasks show that after applying
Resonance RoPE, Transformers recognize OOD position better and more robustly.
Our extensive LLM experiments also show superior performance after applying
Resonance RoPE to the current state-of-the-art RoPE scaling method, YaRN, on
both upstream language modeling tasks and a variety of downstream long-text
applications. | cs.CL | Resonance RoPE: Improving Context Length Generalization of Large Language Models |
2024-02-29T19:00:47Z | http://arxiv.org/pdf/2403.00067v1 | http://arxiv.org/abs/2403.00067v1 | Md Tahmid Rahman Laskar, Elena Khasanova, Xue-Yong Fu, Cheng Chen, Shashi Bhushan TN | 2024-02-29T19:00:47Z | cs.CL | This work focuses on the task of query-based meeting summarization in which
the summary of a context (meeting transcript) is generated in response to a
specific query. When using Large Language Models (LLMs) for this task, a new
call to the LLM inference endpoint/API is required for each new query even if
the context stays the same. However, repeated calls to the LLM inference
endpoints would significantly increase the costs of using them in production,
making LLMs impractical for many real-world use cases. To address this problem,
in this paper, we investigate whether combining the queries for the same input
context in a single prompt to minimize repeated calls can be successfully used
in meeting summarization. In this regard, we conduct extensive experiments by
comparing the performance of various popular LLMs: GPT-4, PaLM-2, LLaMA-2,
Mistral, and FLAN-T5 in single-query and multi-query settings. We observe that
while most LLMs tend to respond to the multi-query instructions, almost all of
them (except GPT-4), even after fine-tuning, could not properly generate the
response in the required output format. We conclude that while multi-query
prompting could be useful to optimize the inference costs by reducing calls to
the inference endpoints/APIs for the task of meeting summarization, this
capability to reliably generate the response in the expected format is only
limited to certain LLMs. | cs.CL | Query-OPT: Optimizing Inference of Large Language Models via Multi-Query Instructions in Meeting Summarization |
2024-02-29T18:58:15Z | http://arxiv.org/pdf/2402.19471v2 | http://arxiv.org/abs/2402.19471v2 | Gabriel Grand, Valerio Pepe, Jacob Andreas, Joshua B. Tenenbaum | 2024-05-01T19:00:06Z | cs.CL, cs.AI | Questions combine our mastery of language with our remarkable facility for
reasoning about uncertainty. How do people navigate vast hypothesis spaces to
pose informative questions given limited cognitive resources? We study these
tradeoffs in a classic grounded question-asking task based on the board game
Battleship. Our language-informed program sampling (LIPS) model uses large
language models (LLMs) to generate natural language questions, translate them
into symbolic programs, and evaluate their expected information gain. We find
that with a surprisingly modest resource budget, this simple Monte Carlo
optimization strategy yields informative questions that mirror human
performance across varied Battleship board scenarios. In contrast, LLM-only
baselines struggle to ground questions in the board state; notably, GPT-4V
provides no improvement over non-visual baselines. Our results illustrate how
Bayesian models of question-asking can leverage the statistics of language to
capture human priors, while highlighting some shortcomings of pure LLMs as
grounded reasoners. | cs.CL | Loose LIPS Sink Ships: Asking Questions in Battleship with Language-Informed Program Sampling |
2024-02-29T18:57:01Z | http://arxiv.org/pdf/2402.19467v3 | http://arxiv.org/abs/2402.19467v3 | Kate Sanders, Nathaniel Weir, Benjamin Van Durme | 2024-03-11T01:31:58Z | cs.CL, cs.AI, cs.CV, I.2.7; I.2.10 | It is challenging to perform question-answering over complex, multimodal
content such as television clips. This is in part because current
video-language models rely on single-modality reasoning, have lowered
performance on long inputs, and lack interpetability. We propose TV-TREES, the
first multimodal entailment tree generator. TV-TREES serves as an approach to
video understanding that promotes interpretable joint-modality reasoning by
producing trees of entailment relationships between simple premises directly
entailed by the videos and higher-level conclusions. We then introduce the task
of multimodal entailment tree generation to evaluate the reasoning quality of
such methods. Our method's experimental results on the challenging TVQA dataset
demonstrate intepretable, state-of-the-art zero-shot performance on full video
clips, illustrating a best-of-both-worlds contrast to black-box methods. | cs.CL | TV-TREES: Multimodal Entailment Trees for Neuro-Symbolic Video Reasoning |
2024-02-29T18:55:06Z | http://arxiv.org/pdf/2402.19465v1 | http://arxiv.org/abs/2402.19465v1 | Chen Qian, Jie Zhang, Wei Yao, Dongrui Liu, Zhenfei Yin, Yu Qiao, Yong Liu, Jing Shao | 2024-02-29T18:55:06Z | cs.CL, cs.AI | Ensuring the trustworthiness of large language models (LLMs) is crucial. Most
studies concentrate on fully pre-trained LLMs to better understand and improve
LLMs' trustworthiness. In this paper, to reveal the untapped potential of
pre-training, we pioneer the exploration of LLMs' trustworthiness during this
period, focusing on five key dimensions: reliability, privacy, toxicity,
fairness, and robustness. To begin with, we apply linear probing to LLMs. The
high probing accuracy suggests that \textit{LLMs in early pre-training can
already distinguish concepts in each trustworthiness dimension}. Therefore, to
further uncover the hidden possibilities of pre-training, we extract steering
vectors from a LLM's pre-training checkpoints to enhance the LLM's
trustworthiness. Finally, inspired by~\citet{choi2023understanding} that mutual
information estimation is bounded by linear probing accuracy, we also probe
LLMs with mutual information to investigate the dynamics of trustworthiness
during pre-training. We are the first to observe a similar two-phase
phenomenon: fitting and compression~\citep{shwartz2017opening}. This research
provides an initial exploration of trustworthiness modeling during LLM
pre-training, seeking to unveil new insights and spur further developments in
the field. We will make our code publicly accessible at
\url{https://github.com/ChnQ/TracingLLM}. | cs.CL | Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models |
2024-02-29T18:55:03Z | http://arxiv.org/pdf/2402.19464v1 | http://arxiv.org/abs/2402.19464v1 | Zhang-Wei Hong, Idan Shenfeld, Tsun-Hsuan Wang, Yung-Sung Chuang, Aldo Pareja, James Glass, Akash Srivastava, Pulkit Agrawal | 2024-02-29T18:55:03Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) hold great potential for many natural language
applications but risk generating incorrect or toxic content. To probe when an
LLM generates unwanted content, the current paradigm is to recruit a
\textit{red team} of human testers to design input prompts (i.e., test cases)
that elicit undesirable responses from LLMs. However, relying solely on human
testers is expensive and time-consuming. Recent works automate red teaming by
training a separate red team LLM with reinforcement learning (RL) to generate
test cases that maximize the chance of eliciting undesirable responses from the
target LLM. However, current RL methods are only able to generate a small
number of effective test cases resulting in a low coverage of the span of
prompts that elicit undesirable responses from the target LLM. To overcome this
limitation, we draw a connection between the problem of increasing the coverage
of generated test cases and the well-studied approach of curiosity-driven
exploration that optimizes for novelty. Our method of curiosity-driven red
teaming (CRT) achieves greater coverage of test cases while mantaining or
increasing their effectiveness compared to existing methods. Our method, CRT
successfully provokes toxic responses from LLaMA2 model that has been heavily
fine-tuned using human preferences to avoid toxic outputs. Code is available at
\url{https://github.com/Improbable-AI/curiosity_redteam} | cs.LG | Curiosity-driven Red-teaming for Large Language Models |
2024-02-29T18:54:46Z | http://arxiv.org/pdf/2402.19462v2 | http://arxiv.org/abs/2402.19462v2 | Pranav Shetty, Aishat Adeboye, Sonakshi Gupta, Chao Zhang, Rampi Ramprasad | 2024-06-22T03:56:31Z | cond-mat.mtrl-sci, cs.CL, physics.app-ph | We present a simulation of various active learning strategies for the
discovery of polymer solar cell donor/acceptor pairs using data extracted from
the literature spanning $\sim$20 years by a natural language processing
pipeline. While data-driven methods have been well established to discover
novel materials faster than Edisonian trial-and-error approaches, their
benefits have not been quantified for material discovery problems that can take
decades. Our approach demonstrates a potential reduction in discovery time by
approximately 75 %, equivalent to a 15 year acceleration in material
innovation. Our pipeline enables us to extract data from greater than 3300
papers which is $\sim$5 times larger and therefore more diverse than similar
data sets reported by others. We also trained machine learning models to
predict the power conversion efficiency and used our model to identify
promising donor-acceptor combinations that are as yet unreported. We thus
demonstrate a pipeline that goes from published literature to extracted
material property data which in turn is used to obtain data-driven insights.
Our insights include active learning strategies that can be used to train
strong predictive models of material properties or be robust to the initial
material system used. This work provides a valuable framework for data-driven
research in materials science. | cond-mat.mtrl-sci | Accelerating materials discovery for polymer solar cells: Data-driven insights enabled by natural language processing |
2024-02-29T18:51:23Z | http://arxiv.org/pdf/2402.19457v2 | http://arxiv.org/abs/2402.19457v2 | Maxime Darrin, Philippe Formont, Jackie Chi Kit Cheung, Pablo Piantanida | 2024-03-01T15:29:52Z | cs.CL, cs.AI | Assessing the quality of summarizers poses significant challenges. In
response, we propose a novel task-oriented evaluation approach that assesses
summarizers based on their capacity to produce summaries that are useful for
downstream tasks, while preserving task outcomes. We theoretically establish a
direct relationship between the resulting error probability of these tasks and
the mutual information between source texts and generated summaries. We
introduce $\texttt{COSMIC}$ as a practical implementation of this metric,
demonstrating its strong correlation with human judgment-based metrics and its
effectiveness in predicting downstream task performance. Comparative analyses
against established metrics like $\texttt{BERTScore}$ and $\texttt{ROUGE}$
highlight the competitive performance of $\texttt{COSMIC}$. | cs.CL | $\texttt{COSMIC}$: Mutual Information for Task-Agnostic Summarization Evaluation |
2024-02-29T18:48:18Z | http://arxiv.org/pdf/2402.19450v1 | http://arxiv.org/abs/2402.19450v1 | Saurabh Srivastava, Annarose M B, Anto P V, Shashank Menon, Ajay Sukumar, Adwaith Samod T, Alan Philipose, Stevin Prince, Sooraj Thomas | 2024-02-29T18:48:18Z | cs.AI, cs.CL | We propose a framework for robust evaluation of reasoning capabilities of
language models, using functional variants of benchmarks. Models that solve a
reasoning test should exhibit no difference in performance over the static
version of a problem compared to a snapshot of the functional variant. We have
rewritten the relevant fragment of the MATH benchmark into its functional
variant MATH(), with functionalization of other benchmarks to follow. When
evaluating current state-of-the-art models over snapshots of MATH(), we find a
reasoning gap -- the percentage difference between the static and functional
accuracies. We find reasoning gaps from 58.35% to 80.31% among the
state-of-the-art closed and open weights models that perform well on static
benchmarks, with the caveat that the gaps are likely to be smaller with more
sophisticated prompting strategies. Here we show that models which anecdotally
have good reasoning performance over real-world tasks, have quantifiable lower
gaps, motivating the open problem of building "gap 0" models. Code for
evaluation and new evaluation datasets, three MATH() snapshots, are publicly
available at https://github.com/consequentai/fneval/. | cs.AI | Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gap |
2024-02-29T18:47:52Z | http://arxiv.org/pdf/2402.19449v1 | http://arxiv.org/abs/2402.19449v1 | Frederik Kunstner, Robin Yadav, Alan Milligan, Mark Schmidt, Alberto Bietti | 2024-02-29T18:47:52Z | cs.LG, cs.CL, math.OC, stat.ML | Adam has been shown to outperform gradient descent in optimizing large
language transformers empirically, and by a larger margin than on other tasks,
but it is unclear why this happens. We show that the heavy-tailed class
imbalance found in language modeling tasks leads to difficulties in the
optimization dynamics. When training with gradient descent, the loss associated
with infrequent words decreases slower than the loss associated with frequent
ones. As most samples come from relatively infrequent words, the average loss
decreases slowly with gradient descent. On the other hand, Adam and sign-based
methods do not suffer from this problem and improve predictions on all classes.
To establish that this behavior is indeed caused by class imbalance, we show
empirically that it persist through different architectures and data types, on
language transformers, vision CNNs, and linear models. We further study this
phenomenon on a linear classification with cross-entropy loss, showing that
heavy-tailed class imbalance leads to ill-conditioning, and that the
normalization used by Adam can counteract it. | cs.LG | Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Models |
2024-02-29T18:45:56Z | http://arxiv.org/pdf/2402.19446v1 | http://arxiv.org/abs/2402.19446v1 | Yifei Zhou, Andrea Zanette, Jiayi Pan, Sergey Levine, Aviral Kumar | 2024-02-29T18:45:56Z | cs.LG, cs.AI, cs.CL | A broad use case of large language models (LLMs) is in goal-directed
decision-making tasks (or "agent" tasks), where an LLM needs to not just
generate completions for a given prompt, but rather make intelligent decisions
over a multi-turn interaction to accomplish a task (e.g., when interacting with
the web, using tools, or providing customer support). Reinforcement learning
(RL) provides a general paradigm to address such agent tasks, but current RL
methods for LLMs largely focus on optimizing single-turn rewards. By
construction, most single-turn RL methods cannot endow LLMs with the ability to
intelligently seek information over multiple turns, perform credit assignment,
or reason about their past actions -- all of which are critical in agent tasks.
This raises the question: how can we design effective and efficient multi-turn
RL algorithms for LLMs? In this paper, we develop a framework for building
multi-turn RL algorithms for fine-tuning LLMs, that preserves the flexibility
of existing single-turn RL methods for LLMs (e.g., proximal policy
optimization), while accommodating multiple turns, long horizons, and delayed
rewards effectively. To do this, our framework adopts a hierarchical RL
approach and runs two RL algorithms in parallel: a high-level off-policy
value-based RL algorithm to aggregate reward over utterances, and a low-level
RL algorithm that utilizes this high-level value function to train a token
policy within each utterance or turn. Our hierarchical framework, Actor-Critic
Framework with a Hierarchical Structure (ArCHer), can also give rise to other
RL methods. Empirically, we find that ArCHer significantly improves efficiency
and performance on agent tasks, attaining a sample efficiency of about 100x
over existing methods, while also improving with larger model capacity (upto
the 7 billion scale that we tested on). | cs.LG | ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL |
2024-02-29T18:30:52Z | http://arxiv.org/pdf/2403.00854v1 | http://arxiv.org/abs/2403.00854v1 | Lauren Stumpf, Balasundaram Kadirvelu, Sigourney Waibel, A. Aldo Faisal | 2024-02-29T18:30:52Z | q-bio.NC, cs.AI, cs.CL, cs.LG, cs.SD, eess.AS, I.2.7; I.2.1; J.3 | Dysarthria, a condition resulting from impaired control of the speech muscles
due to neurological disorders, significantly impacts the communication and
quality of life of patients. The condition's complexity, human scoring and
varied presentations make its assessment and management challenging. This study
presents a transformer-based framework for automatically assessing dysarthria
severity from raw speech data. It can offer an objective, repeatable,
accessible, standardised and cost-effective and compared to traditional methods
requiring human expert assessors. We develop a transformer framework, called
Speaker-Agnostic Latent Regularisation (SALR), incorporating a multi-task
learning objective and contrastive learning for speaker-independent multi-class
dysarthria severity classification. The multi-task framework is designed to
reduce reliance on speaker-specific characteristics and address the intrinsic
intra-class variability of dysarthric speech. We evaluated on the Universal
Access Speech dataset using leave-one-speaker-out cross-validation, our model
demonstrated superior performance over traditional machine learning approaches,
with an accuracy of $70.48\%$ and an F1 score of $59.23\%$. Our SALR model also
exceeded the previous benchmark for AI-based classification, which used support
vector machines, by $16.58\%$. We open the black box of our model by
visualising the latent space where we can observe how the model substantially
reduces speaker-specific cues and amplifies task-specific ones, thereby showing
its robustness. In conclusion, SALR establishes a new benchmark in
speaker-independent multi-class dysarthria severity classification using
generative AI. The potential implications of our findings for broader clinical
applications in automated dysarthria severity assessments. | q-bio.NC | Speaker-Independent Dysarthria Severity Classification using Self-Supervised Transformers and Multi-Task Learning |
2024-02-29T18:27:27Z | http://arxiv.org/pdf/2402.19431v1 | http://arxiv.org/abs/2402.19431v1 | Zexiong Ma, Shengnan An, Bing Xie, Zeqi Lin | 2024-02-29T18:27:27Z | cs.SE, cs.AI, cs.CL | Large language models (LLMs) have achieved exceptional performance in code
generation. However, the performance remains unsatisfactory in generating
library-oriented code, especially for the libraries not present in the training
data of LLMs. Previous work utilizes API recommendation technology to help LLMs
use libraries: it retrieves APIs related to the user requirements, then
leverages them as context to prompt LLMs. However, developmental requirements
can be coarse-grained, requiring a combination of multiple fine-grained APIs.
This granularity inconsistency makes API recommendation a challenging task. To
address this, we propose CAPIR (Compositional API Recommendation), which adopts
a "divide-and-conquer" strategy to recommend APIs for coarse-grained
requirements. Specifically, CAPIR employs an LLM-based Decomposer to break down
a coarse-grained task description into several detailed subtasks. Then, CAPIR
applies an embedding-based Retriever to identify relevant APIs corresponding to
each subtask. Moreover, CAPIR leverages an LLM-based Reranker to filter out
redundant APIs and provides the final recommendation. To facilitate the
evaluation of API recommendation methods on coarse-grained requirements, we
present two challenging benchmarks, RAPID (Recommend APIs based on
Documentation) and LOCG (Library-Oriented Code Generation). Experimental
results on these benchmarks, demonstrate the effectiveness of CAPIR in
comparison to existing baselines. Specifically, on RAPID's Torchdata-AR
dataset, compared to the state-of-the-art API recommendation approach, CAPIR
improves recall@5 from 18.7% to 43.2% and precision@5 from 15.5% to 37.1%. On
LOCG's Torchdata-Code dataset, compared to code generation without API
recommendation, CAPIR improves pass@100 from 16.0% to 28.0%. | cs.SE | Compositional API Recommendation for Library-Oriented Code Generation |
2024-02-29T18:24:46Z | http://arxiv.org/pdf/2402.19427v1 | http://arxiv.org/abs/2402.19427v1 | Soham De, Samuel L. Smith, Anushan Fernando, Aleksandar Botev, George Cristian-Muraru, Albert Gu, Ruba Haroun, Leonard Berrada, Yutian Chen, Srivatsan Srinivasan, Guillaume Desjardins, Arnaud Doucet, David Budden, Yee Whye Teh, Razvan Pascanu, Nando De Freitas, Caglar Gulcehre | 2024-02-29T18:24:46Z | cs.LG, cs.CL | Recurrent neural networks (RNNs) have fast inference and scale efficiently on
long sequences, but they are difficult to train and hard to scale. We propose
Hawk, an RNN with gated linear recurrences, and Griffin, a hybrid model that
mixes gated linear recurrences with local attention. Hawk exceeds the reported
performance of Mamba on downstream tasks, while Griffin matches the performance
of Llama-2 despite being trained on over 6 times fewer tokens. We also show
that Griffin can extrapolate on sequences significantly longer than those seen
during training. Our models match the hardware efficiency of Transformers
during training, and during inference they have lower latency and significantly
higher throughput. We scale Griffin up to 14B parameters, and explain how to
shard our models for efficient distributed training. | cs.LG | Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models |
2024-02-29T18:09:03Z | http://arxiv.org/pdf/2402.19411v1 | http://arxiv.org/abs/2402.19411v1 | Mainak Ghosh, Sebastian Erhardt, Michael E. Rose, Erik Buunk, Dietmar Harhoff | 2024-02-29T18:09:03Z | cs.IR, cs.CL, cs.LG | PaECTER is a publicly available, open-source document-level encoder specific
for patents. We fine-tune BERT for Patents with examiner-added citation
information to generate numerical representations for patent documents. PaECTER
performs better in similarity tasks than current state-of-the-art models used
in the patent domain. More specifically, our model outperforms the next-best
patent specific pre-trained language model (BERT for Patents) on our patent
citation prediction test dataset on two different rank evaluation metrics.
PaECTER predicts at least one most similar patent at a rank of 1.32 on average
when compared against 25 irrelevant patents. Numerical representations
generated by PaECTER from patent text can be used for downstream tasks such as
classification, tracing knowledge flows, or semantic similarity search.
Semantic similarity search is especially relevant in the context of prior art
search for both inventors and patent examiners. PaECTER is available on Hugging
Face. | cs.IR | PaECTER: Patent-level Representation Learning using Citation-informed Transformers |
2024-02-29T18:04:11Z | http://arxiv.org/pdf/2402.19406v2 | http://arxiv.org/abs/2402.19406v2 | Nathan Godey, Éric de la Clergerie, Benoît Sagot | 2024-03-04T11:35:02Z | cs.CL, cs.AI | Language models have long been shown to embed geographical information in
their hidden representations. This line of work has recently been revisited by
extending this result to Large Language Models (LLMs). In this paper, we
propose to fill the gap between well-established and recent literature by
observing how geographical knowledge evolves when scaling language models. We
show that geographical knowledge is observable even for tiny models, and that
it scales consistently as we increase the model size. Notably, we observe that
larger language models cannot mitigate the geographical bias that is inherent
to the training data. | cs.CL | On the Scaling Laws of Geographical Representation in Language Models |
2024-02-29T18:03:00Z | http://arxiv.org/pdf/2402.19404v4 | http://arxiv.org/abs/2402.19404v4 | Junzhe Zhang, Huixuan Zhang, Xunjian Yin, Xiaojun Wan | 2024-05-06T14:41:56Z | cs.CV, cs.CL | News image captioning requires model to generate an informative caption rich
in entities, with the news image and the associated news article. Though
Multimodal Large Language Models (MLLMs) have demonstrated remarkable
capabilities in addressing various vision-language tasks, our research finds
that current MLLMs still bear limitations in handling entity information on
news image captioning task. Besides, while MLLMs have the ability to process
long inputs, generating high-quality news image captions still requires a
trade-off between sufficiency and conciseness of textual input information. To
explore the potential of MLLMs and address problems we discovered, we propose :
an Entity-Aware Multimodal Alignment based approach for news image captioning.
Our approach first aligns the MLLM through Balance Training Strategy with two
extra alignment tasks: Entity-Aware Sentence Selection task and Entity
Selection task, together with News Image Captioning task, to enhance its
capability in handling multimodal entity information. The aligned MLLM will
utilizes the additional entity-related information it explicitly extracts to
supplement its textual input while generating news image captions. Our approach
achieves better results than all previous models in CIDEr score on GoodNews
dataset (72.33 -> 88.39) and NYTimes800k dataset (70.83 -> 85.61). | cs.CV | EAMA : Entity-Aware Multimodal Alignment Based Approach for News Image Captioning |
2024-02-29T17:27:59Z | http://arxiv.org/pdf/2402.19379v5 | http://arxiv.org/abs/2402.19379v5 | Philipp Schoenegger, Indre Tuminauskaite, Peter S. Park, Rafael Valdece Sousa Bastos, Philip E. Tetlock | 2024-06-17T11:38:00Z | cs.CY, cs.AI, cs.CL, cs.LG | Human forecasting accuracy in practice relies on the 'wisdom of the crowd'
effect, in which predictions about future events are significantly improved by
aggregating across a crowd of individual forecasters. Past work on the
forecasting ability of large language models (LLMs) suggests that frontier
LLMs, as individual forecasters, underperform compared to the gold standard of
a human-crowd forecasting-tournament aggregate. In Study 1, we expand this
research by using an LLM ensemble approach consisting of a crowd of 12 LLMs. We
compare the aggregated LLM predictions on 31 binary questions to those of a
crowd of 925 human forecasters from a three-month forecasting tournament. Our
preregistered main analysis shows that the LLM crowd outperforms a simple
no-information benchmark, and is not statistically different from the human
crowd. We also observe a set of human-like biases in machine responses, such as
an acquiescence effect and a tendency to favour round numbers. In Study 2, we
test whether LLM predictions (of GPT-4 and Claude 2) can be improved by drawing
on human cognitive output. We find that both models' forecasting accuracy
benefits from exposure to the median human prediction as information, improving
accuracy by between 17% and 28%, though this leads to less accurate predictions
than simply averaging human and machine forecasts. Our results suggest that
LLMs can achieve forecasting accuracy rivaling that of the human crowd: via the
simple, practically applicable method of forecast aggregation. | cs.CY | Wisdom of the Silicon Crowd: LLM Ensemble Prediction Capabilities Rival Human Crowd Accuracy |
2024-02-29T17:19:39Z | http://arxiv.org/pdf/2402.19371v1 | http://arxiv.org/abs/2402.19371v1 | Jenish Maharjan, Anurag Garikipati, Navan Preet Singh, Leo Cyrus, Mayank Sharma, Madalina Ciobanu, Gina Barnes, Rahul Thapa, Qingqing Mao, Ritankar Das | 2024-02-29T17:19:39Z | cs.CL, cs.AI, cs.IR | LLMs have become increasingly capable at accomplishing a range of
specialized-tasks and can be utilized to expand equitable access to medical
knowledge. Most medical LLMs have involved extensive fine-tuning, leveraging
specialized medical data and significant, thus costly, amounts of computational
power. Many of the top performing LLMs are proprietary and their access is
limited to very few research groups. However, open-source (OS) models represent
a key area of growth for medical LLMs due to significant improvements in
performance and an inherent ability to provide the transparency and compliance
required in healthcare. We present OpenMedLM, a prompting platform which
delivers state-of-the-art (SOTA) performance for OS LLMs on medical benchmarks.
We evaluated a range of OS foundation LLMs (7B-70B) on four medical benchmarks
(MedQA, MedMCQA, PubMedQA, MMLU medical-subset). We employed a series of
prompting strategies, including zero-shot, few-shot, chain-of-thought (random
selection and kNN selection), and ensemble/self-consistency voting. We found
that OpenMedLM delivers OS SOTA results on three common medical LLM benchmarks,
surpassing the previous best performing OS models that leveraged
computationally costly extensive fine-tuning. The model delivers a 72.6%
accuracy on the MedQA benchmark, outperforming the previous SOTA by 2.4%, and
achieves 81.7% accuracy on the MMLU medical-subset, establishing itself as the
first OS LLM to surpass 80% accuracy on this benchmark. Our results highlight
medical-specific emergent properties in OS LLMs which have not yet been
documented to date elsewhere, and showcase the benefits of further leveraging
prompt engineering to improve the performance of accessible LLMs for medical
applications. | cs.CL | OpenMedLM: Prompt engineering can out-perform fine-tuning in medical question-answering with open-source large language models |
2024-02-29T16:56:36Z | http://arxiv.org/pdf/2402.19350v5 | http://arxiv.org/abs/2402.19350v5 | Guangming Huang, Yunfei Long, Cunjin Luo, Jiaxing Shen, Xia Sun | 2024-06-28T00:26:41Z | cs.CL | Pre-trained language models (PLMs) leverage chains-of-thought (CoT) to
simulate human reasoning and inference processes, achieving proficient
performance in multi-hop QA. However, a gap persists between PLMs' reasoning
abilities and those of humans when tackling complex problems. Psychological
studies suggest a vital connection between explicit information in passages and
human prior knowledge during reading. Nevertheless, current research has given
insufficient attention to linking input passages and PLMs' pre-training-based
knowledge from the perspective of human cognition studies. In this study, we
introduce a Prompting Explicit and Implicit knowledge (PEI) framework, which
uses prompts to connect explicit and implicit knowledge, aligning with human
reading process for multi-hop QA. We consider the input passages as explicit
knowledge, employing them to elicit implicit knowledge through unified prompt
reasoning. Furthermore, our model incorporates type-specific reasoning via
prompts, a form of implicit knowledge. Experimental results show that PEI
performs comparably to the state-of-the-art on HotpotQA. Ablation studies
confirm the efficacy of our model in bridging and integrating explicit and
implicit knowledge. | cs.CL | Prompting Explicit and Implicit Knowledge for Multi-hop Question Answering Based on Human Reading Process |
2024-02-29T16:37:08Z | http://arxiv.org/pdf/2402.19334v2 | http://arxiv.org/abs/2402.19334v2 | Ansh Arora, Xuanli He, Maximilian Mozes, Srinibas Swain, Mark Dras, Qiongkai Xu | 2024-06-03T16:19:03Z | cs.CL | The democratization of pre-trained language models through open-source
initiatives has rapidly advanced innovation and expanded access to cutting-edge
technologies. However, this openness also brings significant security risks,
including backdoor attacks, where hidden malicious behaviors are triggered by
specific inputs, compromising natural language processing (NLP) system
integrity and reliability. This paper suggests that merging a backdoored model
with other homogeneous models can significantly remediate backdoor
vulnerabilities even if such models are not entirely secure. In our
experiments, we verify our hypothesis on various models (BERT-Base,
RoBERTa-Large, Llama2-7B, and Mistral-7B) and datasets (SST-2, OLID, AG News,
and QNLI). Compared to multiple advanced defensive approaches, our method
offers an effective and efficient inference-stage defense against backdoor
attacks on classification and instruction-tuned tasks without additional
resources or specific knowledge. Our approach consistently outperforms recent
advanced baselines, leading to an average of about 75% reduction in the attack
success rate. Since model merging has been an established approach for
improving model performance, the extra advantage it provides regarding defense
can be seen as a cost-free bonus. | cs.CL | Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge |
2024-02-29T16:36:51Z | http://arxiv.org/pdf/2402.19333v2 | http://arxiv.org/abs/2402.19333v2 | Tsz Kin Lam, Alexandra Birch, Barry Haddow | 2024-06-26T09:50:28Z | cs.CL, cs.SD, eess.AS | We propose a pretraining method to use Self-Supervised Speech (SSS) model to
creating more compact Speech-to-text Translation. In contrast to using the SSS
model for initialization, our method is more suitable to memory constrained
scenario such as on-device deployment. Our method is based on Discrete Speech
Units (DSU) extracted from the SSS model. In the first step, our method
pretrains two smaller encoder-decoder models on 1) Filterbank-to-DSU
(Fbk-to-DSU) and 2) DSU-to-Translation (DSU-to-Trl) data respectively. The DSU
thus become the distillation inputs of the smaller models. Subsequently, the
encoder from the Fbk-to-DSU model and the decoder from the DSU-to-Trl model are
taken to initialise the compact model. Finally, the compact model is finetuned
on the paired Fbk-Trl data. In addition to being compact, our method requires
no transcripts, making it applicable to low-resource settings. It also avoids
speech discretization in inference and is more robust to the DSU tokenization.
Evaluation on CoVoST-2 (X-En) shows that our method has consistent improvement
over the baseline in three metrics while being compact i.e., only half the SSS
model size. | cs.CL | Compact Speech Translation Models via Discrete Speech Units Pretraining |
2024-02-29T16:09:02Z | http://arxiv.org/pdf/2403.00046v2 | http://arxiv.org/abs/2403.00046v2 | Xue Jiang, Yihong Dong, Zhi Jin, Ge Li | 2024-03-23T16:51:11Z | cs.SE, cs.AI, cs.CL | Although Large Language Models (LLMs) have made significant progress in code
generation, they still struggle with code generation tasks in specific
scenarios. These scenarios usually necessitate the adaptation of LLMs to
fulfill specific needs, but the limited training samples available in practice
lead to poor code generation performance. Therefore, how to effectively adapt
LLMs to new scenarios with few training samples is a major challenge for
current code generation. In this paper, we propose a novel adaptation approach
named SEED, which stands for Sample-Efficient adaptation with Error-Driven
learning for code generation. SEED leverages the errors made by LLMs as
learning opportunities, using error revision to overcome its own shortcomings,
thus achieving efficient learning. Specifically, SEED involves identifying
error code generated by LLMs, employing Self-revise for code revision,
optimizing the model with revised code, and iteratively adapting the process
for continuous improvement. Experimental results show that, compared to other
mainstream fine-tuning approaches, SEED achieves superior performance with few
training samples, showing an average relative improvement of 54.7% in Pass@1 on
multiple code generation benchmarks. We also validate the effectiveness of
Self-revise, which generates revised code that optimizes the model more
efficiently compared to the code samples from datasets. Moreover, SEED
consistently demonstrates strong performance across various LLMs, underscoring
its generalizability. | cs.SE | SEED: Customize Large Language Models with Sample-Efficient Adaptation for Code Generation |
2024-02-29T15:49:15Z | http://arxiv.org/pdf/2402.19282v6 | http://arxiv.org/abs/2402.19282v6 | Jiantao Qiu, Haijun Lv, Zhenjiang Jin, Rui Wang, Wenchang Ning, Jia Yu, ChaoBin Zhang, Zhenxiang Li, Pei Chu, Yuan Qu, Jin Shi, Lindong Lu, Runyu Peng, Zhiyuan Zeng, Huanze Tang, Zhikai Lei, Jiawei Hong, Keyu Chen, Zhaoye Fei, Ruiliang Xu, Wei Li, Zhongying Tu, Lin Dahua, Yu Qiao, Hang Yan, Conghui He | 2024-03-18T03:18:58Z | cs.CL | This paper presents WanJuan-CC, a safe and high-quality open-sourced English
webtext dataset derived from Common Crawl data. The study addresses the
challenges of constructing large-scale pre-training datasets for language
models, which require vast amounts of high-quality data. A comprehensive
process was designed to handle Common Crawl data, including extraction,
heuristic rule filtering, fuzzy deduplication, content safety filtering, and
data quality filtering. From approximately 68 billion original English
documents, we obtained 2.22T Tokens of safe data and selected 1.0T Tokens of
high-quality data as part of WanJuan-CC. We have open-sourced 100B Tokens from
this dataset. The paper also provides statistical information related to data
quality, enabling users to select appropriate data according to their needs. To
evaluate the quality and utility of the dataset, we trained 1B-parameter and
3B-parameter models using WanJuan-CC and another dataset, RefinedWeb. Results
show that WanJuan-CC performs better on validation datasets and downstream
tasks. | cs.CL | WanJuan-CC: A Safe and High-Quality Open-sourced English Webtext Dataset |
2024-02-29T15:41:20Z | http://arxiv.org/pdf/2402.19273v1 | http://arxiv.org/abs/2402.19273v1 | He Zhu, Wenjia Zhang, Nuoxian Huang, Boyang Li, Luyao Niu, Zipei Fan, Tianle Lun, Yicheng Tao, Junyou Su, Zhaoya Gong, Chenyu Fang, Xing Liu | 2024-02-29T15:41:20Z | cs.CL | In the field of urban planning, general-purpose large language models often
struggle to meet the specific needs of planners. Tasks like generating urban
planning texts, retrieving related information, and evaluating planning
documents pose unique challenges. To enhance the efficiency of urban
professionals and overcome these obstacles, we introduce PlanGPT, the first
specialized Large Language Model tailored for urban and spatial planning.
Developed through collaborative efforts with institutions like the Chinese
Academy of Urban Planning, PlanGPT leverages a customized local database
retrieval framework, domain-specific fine-tuning of base models, and advanced
tooling capabilities. Empirical tests demonstrate that PlanGPT has achieved
advanced performance, delivering responses of superior quality precisely
tailored to the intricacies of urban planning. | cs.CL | PlanGPT: Enhancing Urban Planning with Tailored Language Model and Efficient Retrieval |
2024-02-29T15:38:28Z | http://arxiv.org/pdf/2402.19267v2 | http://arxiv.org/abs/2402.19267v2 | Seunghyun Ji, Hagai Raja Sinulingga, Darongsae Kwon | 2024-05-21T17:19:37Z | cs.CL, cs.AI | Low-resourced data presents a significant challenge for neural machine
translation. In most cases, the low-resourced environment is caused by high
costs due to the need for domain experts or the lack of language experts.
Therefore, identifying the most training-efficient data within an unsupervised
setting emerges as a practical strategy. Recent research suggests that such
effective data can be identified by selecting 'appropriately complex data'
based on its volume, providing strong intuition for unsupervised data
selection. However, we have discovered that establishing criteria for
unsupervised data selection remains a challenge, as the 'appropriate level of
difficulty' may vary depending on the data domain. We introduce a novel
unsupervised data selection method named 'Capturing Perplexing Named Entities,'
which leverages the maximum inference entropy in translated named entities as a
metric for selection. When tested with the 'Korean-English Parallel Corpus of
Specialized Domains,' our method served as robust guidance for identifying
training-efficient data across different domains, in contrast to existing
methods. | cs.CL | Robust Guidance for Unsupervised Data Selection: Capturing Perplexing Named Entities for Domain-Specific Machine Translation |
2024-02-29T15:26:14Z | http://arxiv.org/pdf/2402.19255v2 | http://arxiv.org/abs/2402.19255v2 | Qintong Li, Leyang Cui, Xueliang Zhao, Lingpeng Kong, Wei Bi | 2024-07-02T03:46:03Z | cs.CL | Large language models (LLMs) have achieved impressive performance across
various mathematical reasoning benchmarks. However, there are increasing
debates regarding whether these models truly understand and apply mathematical
knowledge or merely rely on shortcuts for mathematical reasoning. One essential
and frequently occurring evidence is that when the math questions are slightly
changed, LLMs can behave incorrectly. This motivates us to evaluate the
robustness of LLMs' math reasoning capability by testing a wide range of
question variations. We introduce the adversarial grade school math (GSM-Plus)
dataset, an extension of GSM8K augmented with various mathematical
perturbations. Our experiments on 25 LLMs and 4 prompting techniques show that
while LLMs exhibit different levels of math reasoning abilities, their
performances are far from robust. In particular, even for problems that have
been solved in GSM8K, LLMs can make mistakes when new statements are added or
the question targets are altered. We also explore whether more robust
performance can be achieved by composing existing prompting methods, in which
we try an iterative method that generates and verifies each intermediate
thought based on its reasoning goal and calculation result. | cs.CL | GSM-Plus: A Comprehensive Benchmark for Evaluating the Robustness of LLMs as Mathematical Problem Solvers |
2024-02-29T15:22:13Z | http://arxiv.org/pdf/2402.19248v2 | http://arxiv.org/abs/2402.19248v2 | Zhikun Xu, Yinghui Li, Ruixue Ding, Xinyu Wang, Boli Chen, Yong Jiang, Hai-Tao Zheng, Wenlian Lu, Pengjun Xie, Fei Huang | 2024-03-02T04:37:37Z | cs.CL | How to better evaluate the capabilities of Large Language Models (LLMs) is
the focal point and hot topic in current LLMs research. Previous work has noted
that due to the extremely high cost of iterative updates of LLMs, they are
often unable to answer the latest dynamic questions well. To promote the
improvement of Chinese LLMs' ability to answer dynamic questions, in this
paper, we introduce CDQA, a Chinese Dynamic QA benchmark containing
question-answer pairs related to the latest news on the Chinese Internet. We
obtain high-quality data through a pipeline that combines humans and models,
and carefully classify the samples according to the frequency of answer changes
to facilitate a more fine-grained observation of LLMs' capabilities. We have
also evaluated and analyzed mainstream and advanced Chinese LLMs on CDQA.
Extensive experiments and valuable insights suggest that our proposed CDQA is
challenging and worthy of more further study. We believe that the benchmark we
provide will become one of the key data resources for improving LLMs' Chinese
question-answering ability in the future. | cs.CL | Let LLMs Take on the Latest Challenges! A Chinese Dynamic Question Answering Benchmark |
2024-02-29T14:47:24Z | http://arxiv.org/pdf/2402.19218v1 | http://arxiv.org/abs/2402.19218v1 | Stephan Raaijmakers, Roos Bakker, Anita Cremers, Roy de Kleijn, Tom Kouwenhoven, Tessa Verhoef | 2024-02-29T14:47:24Z | cs.CL | Conversational AI systems that rely on Large Language Models, like
Transformers, have difficulty interweaving external data (like facts) with the
language they generate. Vanilla Transformer architectures are not designed for
answering factual questions with high accuracy. This paper investigates a
possible route for addressing this problem. We propose to extend the standard
Transformer architecture with an additional memory bank holding extra
information (such as facts drawn from a knowledge base), and an extra attention
layer for addressing this memory. We add this augmented memory to a Generative
Adversarial Network-inspired Transformer architecture. This setup allows for
implementing arbitrary felicity conditions on the generated language of the
Transformer. We first demonstrate how this machinery can be deployed for
handling factual questions in goal-oriented dialogues. Secondly, we demonstrate
that our approach can be useful for applications like {\it style adaptation} as
well: the adaptation of utterances according to certain stylistic (external)
constraints, like social properties of human interlocutors in dialogues. | cs.CL | Memory-Augmented Generative Adversarial Transformers |
2024-02-29T14:34:03Z | http://arxiv.org/pdf/2402.19204v1 | http://arxiv.org/abs/2402.19204v1 | Guilherme Lamartine de Mello, Marcelo Finger, and Felipe Serras, Miguel de Mello Carpi, Marcos Menon Jose, Pedro Henrique Domingues, Paulo Cavalim | 2024-02-29T14:34:03Z | cs.CL, I.2.7 | In this paper we present PeLLE, a family of large language models based on
the RoBERTa architecture, for Brazilian Portuguese, trained on curated, open
data from the Carolina corpus. Aiming at reproducible results, we describe
details of the pretraining of the models. We also evaluate PeLLE models against
a set of existing multilingual and PT-BR refined pretrained Transformer-based
LLM encoders, contrasting performance of large versus smaller-but-curated
pretrained models in several downstream tasks. We conclude that several tasks
perform better with larger models, but some tasks benefit from
smaller-but-curated data in its pretraining. | cs.CL | PeLLE: Encoder-based language models for Brazilian Portuguese based on open data |
2024-02-29T14:30:28Z | http://arxiv.org/pdf/2402.19200v2 | http://arxiv.org/abs/2402.19200v2 | Yong Yang, Changjiang Li, Yi Jiang, Xi Chen, Haoyu Wang, Xuhong Zhang, Zonghui Wang, Shouling Ji | 2024-06-08T03:43:12Z | cs.CR, cs.CL | In recent years, "prompt as a service" has greatly enhanced the utility of
large language models (LLMs) by enabling them to perform various downstream
tasks efficiently without fine-tuning. This has also increased the commercial
value of prompts. However, the potential risk of leakage in these
commercialized prompts remains largely underexplored. In this paper, we
introduce a novel attack framework, PRSA, designed for prompt stealing attacks
against LLMs. The main idea of PRSA is to infer the intent behind a prompt by
analyzing its input-output content, enabling the generation of a surrogate
prompt that replicates the original's functionality. Specifically, PRSA mainly
consists of two key phases: prompt mutation and prompt pruning. In the mutation
phase, we propose a prompt attention algorithm based on output difference. The
algorithm facilitates the generation of effective surrogate prompts by learning
key factors that influence the accurate inference of prompt intent. During the
pruning phase, we employ a two-step related word identification strategy to
detect and mask words that are highly related to the input, thus improving the
generalizability of the surrogate prompts. We verify the actual threat of PRSA
through evaluation in both real-world settings, non-interactive and interactive
prompt services. The results strongly confirm the PRSA's effectiveness and
generalizability. We have reported these findings to prompt service providers
and actively collaborate with them to implement defensive measures. | cs.CR | PRSA: PRompt Stealing Attacks against Large Language Models |
2024-02-29T14:06:34Z | http://arxiv.org/pdf/2403.14661v1 | http://arxiv.org/abs/2403.14661v1 | Seyed Parsa Neshaei, Richard Lee Davis, Adam Hazimeh, Bojan Lazarevski, Pierre Dillenbourg, Tanja Käser | 2024-02-29T14:06:34Z | cs.CY, cs.CL, cs.LG | Recent work exploring the capabilities of pre-trained large language models
(LLMs) has demonstrated their ability to act as general pattern machines by
completing complex token sequences representing a wide array of tasks,
including time-series prediction and robot control. This paper investigates
whether the pattern recognition and sequence modeling capabilities of LLMs can
be extended to the domain of knowledge tracing, a critical component in the
development of intelligent tutoring systems (ITSs) that tailor educational
experiences by predicting learner performance over time. In an empirical
evaluation across multiple real-world datasets, we compare two approaches to
using LLMs for this task, zero-shot prompting and model fine-tuning, with
existing, non-LLM approaches to knowledge tracing. While LLM-based approaches
do not achieve state-of-the-art performance, fine-tuned LLMs surpass the
performance of naive baseline models and perform on par with standard Bayesian
Knowledge Tracing approaches across multiple metrics. These findings suggest
that the pattern recognition capabilities of LLMs can be used to model complex
learning trajectories, opening a novel avenue for applying LLMs to educational
contexts. The paper concludes with a discussion of the implications of these
findings for future research, suggesting that further refinements and a deeper
understanding of LLMs' predictive mechanisms could lead to enhanced performance
in knowledge tracing tasks. | cs.CY | Towards Modeling Learner Performance with Large Language Models |
2024-02-29T13:52:33Z | http://arxiv.org/pdf/2402.19170v2 | http://arxiv.org/abs/2402.19170v2 | Mihai Masala, Traian Rebedea, Horia Velicu | 2024-03-04T20:54:34Z | cs.CL, cs.AI | In recent years,the entire field of Natural Language Processing (NLP) has
enjoyed amazing novel results achieving almost human-like performance on a
variety of tasks. Legal NLP domain has also been part of this process, as it
has seen an impressive growth. However, general-purpose models are not readily
applicable for legal domain. Due to the nature of the domain (e.g. specialized
vocabulary, long documents) specific models and methods are often needed for
Legal NLP. In this work we investigate both specialized and general models for
predicting the final ruling of a legal case, task known as Legal Judgment
Prediction (LJP). We particularly focus on methods to extend to sequence length
of Transformer-based models to better understand the long documents present in
legal corpora. Extensive experiments on 4 LJP datasets in Romanian, originating
from 2 sources with significantly different sizes and document lengths, show
that specialized models and handling long texts are critical for a good
performance. | cs.CL | Improving Legal Judgement Prediction in Romanian with Long Text Encoders |
2024-02-29T13:50:47Z | http://arxiv.org/pdf/2402.19167v2 | http://arxiv.org/abs/2402.19167v2 | Chen Zhang, Xiao Liu, Jiuheng Lin, Yansong Feng | 2024-06-13T04:58:21Z | cs.CL | Existing large language models struggle to support numerous low-resource
languages, particularly the extremely low-resource ones, for which there is
minimal training data available for effective parameter updating. We thus
investigate whether LLMs can learn a new language on the fly solely through
prompting. To study this question, we collect a research suite for Zhuang, a
language supported by no LLMs currently. We introduce DiPMT++, a framework for
adapting LLMs to unseen languages by in-context learning. Using a dictionary
and 5K parallel sentences only, DiPMT++ significantly enhances the performance
of GPT-4 from 0 to 16 BLEU for Chinese-to-Zhuang translation and achieves 32
BLEU for Zhuang-to-Chinese translation. We also validate the effectiveness of
our framework on Kalamang, another unseen language. Furthermore, we demonstrate
the practical utility of DiPMT++ in aiding humans in translating completely
unseen languages, which could contribute to the preservation of linguistic
diversity. | cs.CL | Teaching Large Language Models an Unseen Language on the Fly |
2024-02-29T13:49:56Z | http://arxiv.org/pdf/2403.00843v2 | http://arxiv.org/abs/2403.00843v2 | Wentao Shi, Xiangnan He, Yang Zhang, Chongming Gao, Xinyue Li, Jizhi Zhang, Qifan Wang, Fuli Feng | 2024-04-26T07:41:07Z | cs.IR, cs.AI, cs.CL, cs.LG | Planning for both immediate and long-term benefits becomes increasingly
important in recommendation. Existing methods apply Reinforcement Learning (RL)
to learn planning capacity by maximizing cumulative reward for long-term
recommendation. However, the scarcity of recommendation data presents
challenges such as instability and susceptibility to overfitting when training
RL models from scratch, resulting in sub-optimal performance. In this light, we
propose to leverage the remarkable planning capabilities over sparse data of
Large Language Models (LLMs) for long-term recommendation. The key to achieving
the target lies in formulating a guidance plan following principles of
enhancing long-term engagement and grounding the plan to effective and
executable actions in a personalized manner. To this end, we propose a Bi-level
Learnable LLM Planner framework, which consists of a set of LLM instances and
breaks down the learning process into macro-learning and micro-learning to
learn macro-level guidance and micro-level personalized recommendation
policies, respectively. Extensive experiments validate that the framework
facilitates the planning ability of LLMs for long-term recommendation. Our code
and data can be found at https://github.com/jizhi-zhang/BiLLP. | cs.IR | Large Language Models are Learnable Planners for Long-Term Recommendation |
2024-02-29T13:09:26Z | http://arxiv.org/pdf/2402.19133v1 | http://arxiv.org/abs/2402.19133v1 | Stephanie Brandl, Oliver Eberle, Tiago Ribeiro, Anders Søgaard, Nora Hollenstein | 2024-02-29T13:09:26Z | cs.CL | Rationales in the form of manually annotated input spans usually serve as
ground truth when evaluating explainability methods in NLP. They are, however,
time-consuming and often biased by the annotation process. In this paper, we
debate whether human gaze, in the form of webcam-based eye-tracking recordings,
poses a valid alternative when evaluating importance scores. We evaluate the
additional information provided by gaze data, such as total reading times, gaze
entropy, and decoding accuracy with respect to human rationale annotations. We
compare WebQAmGaze, a multilingual dataset for information-seeking QA, with
attention and explainability-based importance scores for 4 different
multilingual Transformer-based language models (mBERT, distil-mBERT, XLMR, and
XLMR-L) and 3 languages (English, Spanish, and German). Our pipeline can easily
be applied to other tasks and languages. Our findings suggest that gaze data
offers valuable linguistic insights that could be leveraged to infer task
difficulty and further show a comparable ranking of explainability methods to
that of human rationales. | cs.CL | Evaluating Webcam-based Gaze Data as an Alternative for Human Rationale Annotations |
2024-02-29T12:56:18Z | http://arxiv.org/pdf/2402.19119v2 | http://arxiv.org/abs/2402.19119v2 | Alexander Black, Jing Shi, Yifei Fan, Tu Bui, John Collomosse | 2024-03-14T19:18:43Z | cs.CV, cs.CL | We present VIXEN - a technique that succinctly summarizes in text the visual
differences between a pair of images in order to highlight any content
manipulation present. Our proposed network linearly maps image features in a
pairwise manner, constructing a soft prompt for a pretrained large language
model. We address the challenge of low volume of training data and lack of
manipulation variety in existing image difference captioning (IDC) datasets by
training on synthetically manipulated images from the recent InstructPix2Pix
dataset generated via prompt-to-prompt editing framework. We augment this
dataset with change summaries produced via GPT-3. We show that VIXEN produces
state-of-the-art, comprehensible difference captions for diverse image contents
and edit types, offering a potential mitigation against misinformation
disseminated via manipulated image content. Code and data are available at
http://github.com/alexblck/vixen | cs.CV | VIXEN: Visual Text Comparison Network for Image Difference Captioning |
2024-02-29T12:49:48Z | http://arxiv.org/pdf/2402.19116v2 | http://arxiv.org/abs/2402.19116v2 | Jiamin Luo, Jianing Zhao, Jingjing Wang, Guodong Zhou | 2024-03-04T08:42:56Z | cs.CL, cs.AI | Weakly-supervised Phrase Grounding (WPG) is an emerging task of inferring the
fine-grained phrase-region matching, while merely leveraging the coarse-grained
sentence-image pairs for training. However, existing studies on WPG largely
ignore the implicit phrase-region matching relations, which are crucial for
evaluating the capability of models in understanding the deep multimodal
semantics. To this end, this paper proposes an Implicit-Enhanced Causal
Inference (IECI) approach to address the challenges of modeling the implicit
relations and highlighting them beyond the explicit. Specifically, this
approach leverages both the intervention and counterfactual techniques to
tackle the above two challenges respectively. Furthermore, a high-quality
implicit-enhanced dataset is annotated to evaluate IECI and detailed
evaluations show the great advantages of IECI over the state-of-the-art
baselines. Particularly, we observe an interesting finding that IECI
outperforms the advanced multimodal LLMs by a large margin on this
implicit-enhanced dataset, which may facilitate more research to evaluate the
multimodal LLMs in this direction. | cs.CL | How to Understand "Support"? An Implicit-enhanced Causal Inference Approach for Weakly-supervised Phrase Grounding |
2024-02-29T12:35:45Z | http://arxiv.org/pdf/2402.19103v1 | http://arxiv.org/abs/2402.19103v1 | Hongbang Yuan, Pengfei Cao, Zhuoran Jin, Yubo Chen, Daojian Zeng, Kang Liu, Jun Zhao | 2024-02-29T12:35:45Z | cs.CL, cs.AI | Large Language Models (LLMs) have shown impressive capabilities but still
suffer from the issue of hallucinations. A significant type of this issue is
the false premise hallucination, which we define as the phenomenon when LLMs
generate hallucinated text when confronted with false premise questions. In
this paper, we perform a comprehensive analysis of the false premise
hallucination and elucidate its internal working mechanism: a small subset of
attention heads (which we designate as false premise heads) disturb the
knowledge extraction process, leading to the occurrence of false premise
hallucination. Based on our analysis, we propose \textbf{FAITH} (\textbf{F}alse
premise \textbf{A}ttention head constra\textbf{I}ining for mi\textbf{T}igating
\textbf{H}allucinations), a novel and effective method to mitigate false
premise hallucinations. It constrains the false premise attention heads during
the model inference process. Impressively, extensive experiments demonstrate
that constraining only approximately $1\%$ of the attention heads in the model
yields a notable increase of nearly $20\%$ of model performance. | cs.CL | Whispers that Shake Foundations: Analyzing and Mitigating False Premise Hallucinations in Large Language Models |
2024-02-29T12:25:45Z | http://arxiv.org/pdf/2402.19097v1 | http://arxiv.org/abs/2402.19097v1 | Alexander Shabalin, Viacheslav Meshchaninov, Tingir Badmaev, Dmitry Molchanov, Grigory Bartosh, Sergey Markov, Dmitry Vetrov | 2024-02-29T12:25:45Z | cs.CL, I.2; I.7 | Drawing inspiration from the success of diffusion models in various domains,
numerous research papers proposed methods for adapting them to text data.
Despite these efforts, none of them has managed to achieve the quality of the
large language models. In this paper, we conduct a comprehensive analysis of
key components of the text diffusion models and introduce a novel approach
named Text Encoding Diffusion Model (TEncDM). Instead of the commonly used
token embedding space, we train our model in the space of the language model
encodings. Additionally, we propose to use a Transformer-based decoder that
utilizes contextual information for text reconstruction. We also analyse
self-conditioning and find that it increases the magnitude of the model
outputs, allowing the reduction of the number of denoising steps at the
inference stage. Evaluation of TEncDM on two downstream text generation tasks,
QQP and XSum, demonstrates its superiority over existing non-autoregressive
models. | cs.CL | TEncDM: Understanding the Properties of Diffusion Model in the Space of Language Model Encodings |
2024-02-29T12:13:50Z | http://arxiv.org/pdf/2402.19088v2 | http://arxiv.org/abs/2402.19088v2 | Jader Martins Camboim de Sá, Marcos Da Silveira, Cédric Pruski | 2024-03-11T15:21:57Z | cs.CL, cs.AI | Live languages continuously evolve to integrate the cultural change of human
societies. This evolution manifests through neologisms (new words) or
\textbf{semantic changes} of words (new meaning to existing words).
Understanding the meaning of words is vital for interpreting texts coming from
different cultures (regionalism or slang), domains (e.g., technical terms), or
periods. In computer science, these words are relevant to computational
linguistics algorithms such as translation, information retrieval, question
answering, etc. Semantic changes can potentially impact the quality of the
outcomes of these algorithms. Therefore, it is important to understand and
characterize these changes formally. The study of this impact is a recent
problem that has attracted the attention of the computational linguistics
community. Several approaches propose methods to detect semantic changes with
good precision, but more effort is needed to characterize how the meaning of
words changes and to reason about how to reduce the impact of semantic change.
This survey provides an understandable overview of existing approaches to the
\textit{characterization of semantic changes} and also formally defines three
classes of characterizations: if the meaning of a word becomes more general or
narrow (change in dimension) if the word is used in a more pejorative or
positive/ameliorated sense (change in orientation), and if there is a trend to
use the word in a, for instance, metaphoric or metonymic context (change in
relation). We summarized the main aspects of the selected publications in a
table and discussed the needs and trends in the research activities on semantic
change characterization. | cs.CL | Survey in Characterization of Semantic Change |
2024-02-29T12:12:30Z | http://arxiv.org/pdf/2402.19085v1 | http://arxiv.org/abs/2402.19085v1 | Yiju Guo, Ganqu Cui, Lifan Yuan, Ning Ding, Jiexin Wang, Huimin Chen, Bowen Sun, Ruobing Xie, Jie Zhou, Yankai Lin, Zhiyuan Liu, Maosong Sun | 2024-02-29T12:12:30Z | cs.CL, cs.AI, cs.SY, eess.SY | Alignment in artificial intelligence pursues the consistency between model
responses and human preferences as well as values. In practice, the
multifaceted nature of human preferences inadvertently introduces what is known
as the "alignment tax" -a compromise where enhancements in alignment within one
objective (e.g.,harmlessness) can diminish performance in others
(e.g.,helpfulness). However, existing alignment techniques are mostly
unidirectional, leading to suboptimal trade-offs and poor flexibility over
various objectives. To navigate this challenge, we argue the prominence of
grounding LLMs with evident preferences. We introduce controllable preference
optimization (CPO), which explicitly specifies preference scores for different
objectives, thereby guiding the model to generate responses that meet the
requirements. Our experimental analysis reveals that the aligned models can
provide responses that match various preferences among the "3H" (helpfulness,
honesty, harmlessness) desiderata. Furthermore, by introducing diverse data and
alignment goals, we surpass baseline methods in aligning with single
objectives, hence mitigating the impact of the alignment tax and achieving
Pareto improvements in multi-objective alignment. | cs.CL | Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment |
2024-02-29T12:01:46Z | http://arxiv.org/pdf/2402.19076v1 | http://arxiv.org/abs/2402.19076v1 | Gennaro Nolano, Moritz Blum, Basil Ell, Philipp Cimiano | 2024-02-29T12:01:46Z | cs.CL | In recent years, large language models have achieved state-of-the-art
performance across various NLP tasks. However, investigations have shown that
these models tend to rely on shortcut features, leading to inaccurate
predictions and causing the models to be unreliable at generalization to
out-of-distribution (OOD) samples. For instance, in the context of relation
extraction (RE), we would expect a model to identify the same relation
independently of the entities involved in it. For example, consider the
sentence "Leonardo da Vinci painted the Mona Lisa" expressing the
created(Leonardo_da_Vinci, Mona_Lisa) relation. If we substiute "Leonardo da
Vinci" with "Barack Obama", then the sentence still expresses the created
relation. A robust model is supposed to detect the same relation in both cases.
In this work, we describe several semantically-motivated strategies to generate
adversarial examples by replacing entity mentions and investigate how
state-of-the-art RE models perform under pressure. Our analyses show that the
performance of these models significantly deteriorates on the modified datasets
(avg. of -48.5% in F1), which indicates that these models rely to a great
extent on shortcuts, such as surface forms (or patterns therein) of entities,
without making full use of the information present in the sentences. | cs.CL | Pointing out the Shortcomings of Relation Extraction Models with Semantically Motivated Adversarials |
2024-02-29T11:29:47Z | http://arxiv.org/pdf/2402.19052v1 | http://arxiv.org/abs/2402.19052v1 | Prottay Kumar Adhikary, Aseem Srivastava, Shivani Kumar, Salam Michael Singh, Puneet Manuja, Jini K Gopinath, Vijay Krishnan, Swati Kedia, Koushik Sinha Deb, Tanmoy Chakraborty | 2024-02-29T11:29:47Z | cs.CL, cs.HC | Comprehensive summaries of sessions enable an effective continuity in mental
health counseling, facilitating informed therapy planning. Yet, manual
summarization presents a significant challenge, diverting experts' attention
from the core counseling process. This study evaluates the effectiveness of
state-of-the-art Large Language Models (LLMs) in selectively summarizing
various components of therapy sessions through aspect-based summarization,
aiming to benchmark their performance. We introduce MentalCLOUDS, a
counseling-component guided summarization dataset consisting of 191 counseling
sessions with summaries focused on three distinct counseling components (aka
counseling aspects). Additionally, we assess the capabilities of 11
state-of-the-art LLMs in addressing the task of component-guided summarization
in counseling. The generated summaries are evaluated quantitatively using
standard summarization metrics and verified qualitatively by mental health
professionals. Our findings demonstrate the superior performance of
task-specific LLMs such as MentalLlama, Mistral, and MentalBART in terms of
standard quantitative metrics such as Rouge-1, Rouge-2, Rouge-L, and BERTScore
across all aspects of counseling components. Further, expert evaluation reveals
that Mistral supersedes both MentalLlama and MentalBART based on six parameters
-- affective attitude, burden, ethicality, coherence, opportunity costs, and
perceived effectiveness. However, these models share the same weakness by
demonstrating a potential for improvement in the opportunity costs and
perceived effectiveness metrics. | cs.CL | Exploring the Efficacy of Large Language Models in Summarizing Mental Health Counseling Sessions: A Benchmark Study |
2024-02-29T09:35:41Z | http://arxiv.org/pdf/2403.00840v1 | http://arxiv.org/abs/2403.00840v1 | Xiaolan Chen, Ziwei Zhao, Weiyi Zhang, Pusheng Xu, Le Gao, Mingpu Xu, Yue Wu, Yinwen Li, Danli Shi, Mingguang He | 2024-02-29T09:35:41Z | cs.CL, cs.AI | Artificial intelligence (AI) has gained significant attention in healthcare
consultation due to its potential to improve clinical workflow and enhance
medical communication. However, owing to the complex nature of medical
information, large language models (LLM) trained with general world knowledge
might not possess the capability to tackle medical-related tasks at an expert
level. Here, we introduce EyeGPT, a specialized LLM designed specifically for
ophthalmology, using three optimization strategies including role-playing,
finetuning, and retrieval-augmented generation. In particular, we proposed a
comprehensive evaluation framework that encompasses a diverse dataset, covering
various subspecialties of ophthalmology, different users, and diverse inquiry
intents. Moreover, we considered multiple evaluation metrics, including
accuracy, understandability, trustworthiness, empathy, and the proportion of
hallucinations. By assessing the performance of different EyeGPT variants, we
identify the most effective one, which exhibits comparable levels of
understandability, trustworthiness, and empathy to human ophthalmologists (all
Ps>0.05). Overall, ur study provides valuable insights for future research,
facilitating comprehensive comparisons and evaluations of different strategies
for developing specialized LLMs in ophthalmology. The potential benefits
include enhancing the patient experience in eye care and optimizing
ophthalmologists' services. | cs.CL | EyeGPT: Ophthalmic Assistant with Large Language Models |
2024-02-29T08:28:04Z | http://arxiv.org/pdf/2402.18950v1 | http://arxiv.org/abs/2402.18950v1 | Erxin Yu, Jing Li, Chunpu Xu | 2024-02-29T08:28:04Z | cs.CL | Social media platforms are daily exhibiting millions of events. To
preliminarily predict the mainstream public reaction to these events, we study
trendy response prediction to automatically generate top-liked user replies to
social media events. While previous works focus on generating responses without
factoring in popularity, we propose Popularity-Aligned Language Models (PopALM)
to distinguish responses liked by a larger audience through reinforcement
learning. Recognizing the noisy labels from user "likes", we tailor-make
curriculum learning in proximal policy optimization (PPO) to help models
capture the essential samples for easy-to-hard training. In experiments, we
build a large-scale Weibo dataset for trendy response prediction, and its
results show that PopALM can help boost the performance of advanced language
models. | cs.CL | PopALM: Popularity-Aligned Language Models for Social Media Trendy Response Prediction |
2024-02-29T08:20:49Z | http://arxiv.org/pdf/2402.18945v2 | http://arxiv.org/abs/2402.18945v2 | Pengzhou Cheng, Wei Du, Zongru Wu, Fengwei Zhang, Libo Chen, Gongshen Liu | 2024-05-24T15:21:55Z | cs.CR, cs.AI, cs.CL | Pre-training has been a necessary phase for deploying pre-trained language
models (PLMs) to achieve remarkable performance in downstream tasks. However,
we empirically show that backdoor attacks exploit such a phase as a vulnerable
entry point for task-agnostic. In this paper, we first propose
$\mathtt{maxEntropy}$, an entropy-based poisoning filtering defense, to prove
that existing task-agnostic backdoors are easily exposed, due to explicit
triggers used. Then, we present $\mathtt{SynGhost}$, an imperceptible and
universal task-agnostic backdoor attack in PLMs. Specifically,
$\mathtt{SynGhost}$ hostilely manipulates clean samples through different
syntactic and then maps the backdoor to representation space without disturbing
the primitive representation. $\mathtt{SynGhost}$ further leverages contrastive
learning to achieve universal, which performs a uniform distribution of
backdoors in the representation space. In light of the syntactic properties, we
also introduce an awareness module to alleviate the interference between
different syntactic. Experiments show that $\mathtt{SynGhost}$ holds more
serious threats. Not only do severe harmfulness to various downstream tasks on
two tuning paradigms but also to any PLMs. Meanwhile, $\mathtt{SynGhost}$ is
imperceptible against three countermeasures based on perplexity, fine-pruning,
and the proposed $\mathtt{maxEntropy}$. | cs.CR | SynGhost: Imperceptible and Universal Task-agnostic Backdoor Attack in Pre-trained Language Models |
2024-02-29T08:20:06Z | http://arxiv.org/pdf/2402.18944v1 | http://arxiv.org/abs/2402.18944v1 | Shivani Kumar, Md Shad Akhtar, Erik Cambria, Tanmoy Chakraborty | 2024-02-29T08:20:06Z | cs.CL, cs.AI | We present SemEval-2024 Task 10, a shared task centred on identifying
emotions and finding the rationale behind their flips within monolingual
English and Hindi-English code-mixed dialogues. This task comprises three
distinct subtasks - emotion recognition in conversation for code-mixed
dialogues, emotion flip reasoning for code-mixed dialogues, and emotion flip
reasoning for English dialogues. Participating systems were tasked to
automatically execute one or more of these subtasks. The datasets for these
tasks comprise manually annotated conversations focusing on emotions and
triggers for emotion shifts (The task data is available at
https://github.com/LCS2-IIITD/EDiReF-SemEval2024.git). A total of 84
participants engaged in this task, with the most adept systems attaining
F1-scores of 0.70, 0.79, and 0.76 for the respective subtasks. This paper
summarises the results and findings from 24 teams alongside their system
descriptions. | cs.CL | SemEval 2024 -- Task 10: Emotion Discovery and Reasoning its Flip in Conversation (EDiReF) |
2024-02-29T07:29:42Z | http://arxiv.org/pdf/2402.18923v1 | http://arxiv.org/abs/2402.18923v1 | Jeehyun Lee, Yerin Choi, Tae-Jin Song, Myoung-Wan Koo | 2024-02-29T07:29:42Z | cs.CL, cs.SD, eess.AS | Dysarthria, a common issue among stroke patients, severely impacts speech
intelligibility. Inappropriate pauses are crucial indicators in severity
assessment and speech-language therapy. We propose to extend a large-scale
speech recognition model for inappropriate pause detection in dysarthric
speech. To this end, we propose task design, labeling strategy, and a speech
recognition model with an inappropriate pause prediction layer. First, we treat
pause detection as speech recognition, using an automatic speech recognition
(ASR) model to convert speech into text with pause tags. According to the newly
designed task, we label pause locations at the text level and their
appropriateness. We collaborate with speech-language pathologists to establish
labeling criteria, ensuring high-quality annotated data. Finally, we extend the
ASR model with an inappropriate pause prediction layer for end-to-end
inappropriate pause detection. Moreover, we propose a task-tailored metric for
evaluating inappropriate pause detection independent of ASR performance. Our
experiments show that the proposed method better detects inappropriate pauses
in dysarthric speech than baselines. (Inappropriate Pause Error Rate: 14.47%) | cs.CL | Inappropriate Pause Detection In Dysarthric Speech Using Large-Scale Speech Recognition |
2024-02-29T07:11:24Z | http://arxiv.org/pdf/2402.18913v1 | http://arxiv.org/abs/2402.18913v1 | Yiran Zhao, Wenxuan Zhang, Huiming Wang, Kenji Kawaguchi, Lidong Bing | 2024-02-29T07:11:24Z | cs.CL, cs.AI | As an effective alternative to the direct fine-tuning on target tasks in
specific languages, cross-lingual transfer addresses the challenges of limited
training data by decoupling ''task ability'' and ''language ability'' by
fine-tuning on the target task in the source language and another selected task
in the target language, respectively. However, they fail to fully separate the
task ability from the source language or the language ability from the chosen
task. In this paper, we acknowledge the mutual reliance between task ability
and language ability and direct our attention toward the gap between the target
language and the source language on tasks. As the gap removes the impact of
tasks, we assume that it remains consistent across tasks. Based on this
assumption, we propose a new cross-lingual transfer method called
$\texttt{AdaMergeX}$ that utilizes adaptive adapter merging. By introducing a
reference task, we can determine that the divergence of adapters fine-tuned on
the reference task in both languages follows the same distribution as the
divergence of adapters fine-tuned on the target task in both languages. Hence,
we can obtain target adapters by combining the other three adapters.
Furthermore, we propose a structure-adaptive adapter merging method. Our
empirical results demonstrate that our approach yields new and effective
cross-lingual transfer, outperforming existing methods across all settings. | cs.CL | AdaMergeX: Cross-Lingual Transfer with Large Language Models via Adaptive Adapter Merging |
2024-02-29T07:08:34Z | http://arxiv.org/pdf/2402.18909v1 | http://arxiv.org/abs/2402.18909v1 | Xiaobao Wu, Liangming Pan, William Yang Wang, Anh Tuan Luu | 2024-02-29T07:08:34Z | cs.CL, cs.AI | Knowledge editing aims to inject knowledge updates into language models to
keep them correct and up-to-date. However, its current evaluation strategies
are notably impractical: they solely update with well-curated structured facts
(triplets with subjects, relations, and objects), whereas real-world knowledge
updates commonly emerge in unstructured texts like news articles. In this
paper, we propose a new benchmark, Unstructured Knowledge Editing (UKE). It
evaluates editing performance directly using unstructured texts as knowledge
updates, termed unstructured facts. Hence UKE avoids the laborious construction
of structured facts and enables efficient and responsive knowledge editing,
becoming a more practical benchmark. We conduct extensive experiments on newly
built datasets and demonstrate that UKE poses a significant challenge to
state-of-the-art knowledge editing methods, resulting in their critical
performance declines. We further show that this challenge persists even if we
extract triplets as structured facts. Our analysis discloses key insights to
motivate future research in UKE for more practical knowledge editing. | cs.CL | Updating Language Models with Unstructured Facts: Towards Practical Knowledge Editing |
2024-02-29T06:40:53Z | http://arxiv.org/pdf/2403.00037v1 | http://arxiv.org/abs/2403.00037v1 | Jiajun Zhang, Zhixun Li, Qiang Liu, Shu Wu, Liang Wang | 2024-02-29T06:40:53Z | cs.SI, cs.AI, cs.CL | With the rapid development of social media, the wide dissemination of fake
news on social media is increasingly threatening both individuals and society.
In the dynamic landscape of social media, fake news detection aims to develop a
model trained on news reporting past events. The objective is to predict and
identify fake news about future events, which often relate to subjects entirely
different from those in the past. However, existing fake detection methods
exhibit a lack of robustness and cannot generalize to unseen events. To address
this, we introduce Future ADaptive Event-based Fake news Detection (FADE)
framework. Specifically, we train a target predictor through an adaptive
augmentation strategy and graph contrastive learning to make more robust
overall predictions. Simultaneously, we independently train an event-only
predictor to obtain biased predictions. Then we further mitigate event bias by
obtaining the final prediction by subtracting the output of the event-only
predictor from the output of the target predictor. Encouraging results from
experiments designed to emulate real-world social media conditions validate the
effectiveness of our method in comparison to existing state-of-the-art
approaches. | cs.SI | Evolving to the Future: Unseen Event Adaptive Fake News Detection on Social Media |
2024-02-29T05:47:34Z | http://arxiv.org/pdf/2402.18877v1 | http://arxiv.org/abs/2402.18877v1 | Yugo Murawaki | 2024-02-29T05:47:34Z | cs.CL | Bayesian approaches to reconstructing the evolutionary history of languages
rely on the tree model, which assumes that these languages descended from a
common ancestor and underwent modifications over time. However, this assumption
can be violated to different extents due to contact and other factors.
Understanding the degree to which this assumption is violated is crucial for
validating the accuracy of phylolinguistic inference. In this paper, we propose
a simple sanity check: projecting a reconstructed tree onto a space generated
by principal component analysis. By using both synthetic and real data, we
demonstrate that our method effectively visualizes anomalies, particularly in
the form of jogging. | cs.CL | Principal Component Analysis as a Sanity Check for Bayesian Phylolinguistic Reconstruction |
2024-02-29T05:43:43Z | http://arxiv.org/pdf/2402.18873v1 | http://arxiv.org/abs/2402.18873v1 | Fangwei Zhu, Peiyi Wang, Zhifang Sui | 2024-02-29T05:43:43Z | cs.CL | Entity abstract summarization aims to generate a coherent description of a
given entity based on a set of relevant Internet documents. Pretrained language
models (PLMs) have achieved significant success in this task, but they may
suffer from hallucinations, i.e. generating non-factual information about the
entity. To address this issue, we decompose the summary into two components:
Facts that represent the factual information about the given entity, which PLMs
are prone to fabricate; and Template that comprises generic content with
designated slots for facts, which PLMs can generate competently. Based on the
facts-template decomposition, we propose SlotSum, an explainable framework for
entity abstract summarization. SlotSum first creates the template and then
predicts the fact for each template slot based on the input documents.
Benefiting from our facts-template decomposition, SlotSum can easily locate
errors and further rectify hallucinated predictions with external knowledge. We
construct a new dataset WikiFactSum to evaluate the performance of SlotSum.
Experimental results demonstrate that SlotSum could generate summaries that are
significantly more factual with credible external knowledge. | cs.CL | Reducing Hallucinations in Entity Abstract Summarization with Facts-Template Decomposition |
2024-02-29T05:27:45Z | http://arxiv.org/pdf/2402.18865v1 | http://arxiv.org/abs/2402.18865v1 | Weijieying Ren, Xinlong Li, Lei Wang, Tianxiang Zhao, Wei Qin | 2024-02-29T05:27:45Z | cs.LG, cs.AI, cs.CL | Existing research has shown that large language models (LLMs) exhibit
remarkable performance in language understanding and generation. However, when
LLMs are continuously fine-tuned on complex and diverse domain-specific
downstream tasks, the inference performance on historical tasks decreases
dramatically, which is known as a catastrophic forgetting problem. A trade-off
needs to be kept between learning plasticity and memory stability. Plenty of
existing works have explored strategies like memory replay, regularization and
parameter isolation, but little is known about the geometric connection of
various adjacent minima in the continual LLMs fine-tuning scenarios. In this
work, we investigate the geometric connections of different minima through the
lens of mode connectivity, which means different minima can be connected by a
low-loss valley. Through extensive experiments, we uncover the mode
connectivity phenomenon in the LLMs continual learning scenario and find that
it can strike a balance between plasticity and stability. Building upon these
findings, we propose a simple yet effective method called Interpolation-based
LoRA (I-LoRA), which constructs a dual-memory experience replay framework based
on LoRA parameter interpolations. Extensive experiments and analysis on eight
domain-specific CL benchmarks demonstrate that I-LoRA consistently show
significant improvement over the previous state-of-the-art approaches with up
to $11\%$ performance gains, providing a strong baseline and insights for
future research on the large language model continual learning problem. Our
code is available at \url{https://github.com/which47/LLMCL}. | cs.LG | Analyzing and Reducing Catastrophic Forgetting in Parameter Efficient Tuning |
2024-02-29T05:16:19Z | http://arxiv.org/pdf/2403.15407v1 | http://arxiv.org/abs/2403.15407v1 | Shafiuddin Rehan Ahmed, Jon Z. Cai, Martha Palmer, James H. Martin | 2024-02-29T05:16:19Z | cs.CL, cs.AI | This paper presents a novel Cross-document Abstract Meaning Representation
(X-AMR) annotation tool designed for annotating key corpus-level event
semantics. Leveraging machine assistance through the Prodigy Annotation Tool,
we enhance the user experience, ensuring ease and efficiency in the annotation
process. Through empirical analyses, we demonstrate the effectiveness of our
tool in augmenting an existing event corpus, highlighting its advantages when
integrated with GPT-4. Code and annotations:
https://github.com/ahmeshaf/gpt_coref | cs.CL | X-AMR Annotation Tool |
2024-02-29T04:53:06Z | http://arxiv.org/pdf/2402.18849v1 | http://arxiv.org/abs/2402.18849v1 | Mingyang Li, Maoqin Yuan, Luyao Li, Han Pengsihua | 2024-02-29T04:53:06Z | cs.CV, cs.AI, cs.CL | This study discusses a new method combining image steganography technology
with Natural Language Processing (NLP) large models, aimed at improving the
accuracy and robustness of extracting steganographic text. Traditional Least
Significant Bit (LSB) steganography techniques face challenges in accuracy and
robustness of information extraction when dealing with complex character
encoding, such as Chinese characters. To address this issue, this study
proposes an innovative LSB-NLP hybrid framework. This framework integrates the
advanced capabilities of NLP large models, such as error detection, correction,
and semantic consistency analysis, as well as information reconstruction
techniques, thereby significantly enhancing the robustness of steganographic
text extraction. Experimental results show that the LSB-NLP hybrid framework
excels in improving the extraction accuracy of steganographic text, especially
in handling Chinese characters. The findings of this study not only confirm the
effectiveness of combining image steganography technology and NLP large models
but also propose new ideas for research and application in the field of
information hiding. The successful implementation of this interdisciplinary
approach demonstrates the great potential of integrating image steganography
technology with natural language processing technology in solving complex
information processing problems. | cs.CV | Enhancing Steganographic Text Extraction: Evaluating the Impact of NLP Models on Accuracy and Semantic Coherence |
2024-02-29T04:11:10Z | http://arxiv.org/pdf/2402.18838v2 | http://arxiv.org/abs/2402.18838v2 | Xuanda Chen, Timothy O'Donnell, Siva Reddy | 2024-03-01T17:40:04Z | cs.CL | Language models (LMs) may appear insensitive to word order changes in natural
language understanding (NLU) tasks. In this paper, we propose that linguistic
redundancy can explain this phenomenon, whereby word order and other linguistic
cues such as case markers provide overlapping and thus redundant information.
Our hypothesis is that models exhibit insensitivity to word order when the
order provides redundant information, and the degree of insensitivity varies
across tasks. We quantify how informative word order is using mutual
information (MI) between unscrambled and scrambled sentences. Our results show
the effect that the less informative word order is, the more consistent the
model's predictions are between unscrambled and scrambled sentences. We also
find that the effect varies across tasks: for some tasks, like SST-2, LMs'
prediction is almost always consistent with the original one even if the
Pointwise-MI (PMI) changes, while for others, like RTE, the consistency is near
random when the PMI gets lower, i.e., word order is really important. | cs.CL | When does word order matter and when doesn't it? |
2024-02-29T03:20:45Z | http://arxiv.org/pdf/2402.18825v2 | http://arxiv.org/abs/2402.18825v2 | Zihan Wang, Peiyi Wang, Houfeng Wang | 2024-03-29T08:08:41Z | cs.CL | Hierarchical text classification (HTC) is a challenging subtask of
multi-label classification due to its complex taxonomic structure. Nearly all
recent HTC works focus on how the labels are structured but ignore the
sub-structure of ground-truth labels according to each input text which
contains fruitful label co-occurrence information. In this work, we introduce
this local hierarchy with an adversarial framework. We propose a HiAdv
framework that can fit in nearly all HTC models and optimize them with the
local hierarchy as auxiliary information. We test on two typical HTC models and
find that HiAdv is effective in all scenarios and is adept at dealing with
complex taxonomic hierarchies. Further experiments demonstrate that the
promotion of our framework indeed comes from the local hierarchy and the local
hierarchy is beneficial for rare classes which have insufficient training data. | cs.CL | Utilizing Local Hierarchy with Adversarial Training for Hierarchical Text Classification |
2024-02-29T02:55:26Z | http://arxiv.org/pdf/2402.18815v2 | http://arxiv.org/abs/2402.18815v2 | Yiran Zhao, Wenxuan Zhang, Guizhen Chen, Kenji Kawaguchi, Lidong Bing | 2024-05-24T07:59:10Z | cs.CL, cs.AI | Large language models (LLMs) have demonstrated impressive capabilities across
diverse languages. This study explores how LLMs handle multilingualism. Based
on observed language ratio shifts among layers and the relationships between
network structures and certain capabilities, we hypothesize the LLM's
multilingual workflow ($\texttt{MWork}$): LLMs initially understand the query,
converting multilingual inputs into English for task-solving. In the
intermediate layers, they employ English for thinking and incorporate
multilingual knowledge with self-attention and feed-forward structures,
respectively. In the final layers, LLMs generate responses aligned with the
original language of the query. To verify $\texttt{MWork}$, we introduce
Parallel Language-specific Neuron Detection ($\texttt{PLND}$) to identify
activated neurons for inputs in different languages without any labeled data.
Using $\texttt{PLND}$, we validate $\texttt{MWork}$ through extensive
experiments involving the deactivation of language-specific neurons across
various layers and structures. Moreover, $\texttt{MWork}$ allows fine-tuning of
language-specific neurons with a small dataset, enhancing multilingual
abilities in a specific language without compromising others. This approach
results in an average improvement of $3.6\%$ for high-resource languages and
$2.3\%$ for low-resource languages across all tasks with just $400$ documents. | cs.CL | How do Large Language Models Handle Multilingualism? |
2024-02-29T02:22:23Z | http://arxiv.org/pdf/2402.18807v1 | http://arxiv.org/abs/2402.18807v1 | Chenglei Shen, Guofu Xie, Xiao Zhang, Jun Xu | 2024-02-29T02:22:23Z | cs.CL, cs.AI | Large language models (LLMs) are now increasingly utilized for role-playing
tasks, especially in impersonating domain-specific experts, primarily through
role-playing prompts. When interacting in real-world scenarios, the
decision-making abilities of a role significantly shape its behavioral
patterns. In this paper, we concentrate on evaluating the decision-making
abilities of LLMs post role-playing thereby validating the efficacy of
role-playing. Our goal is to provide metrics and guidance for enhancing the
decision-making abilities of LLMs in role-playing tasks. Specifically, we first
use LLMs to generate virtual role descriptions corresponding to the 16
personality types of Myers-Briggs Type Indicator (abbreviated as MBTI)
representing a segmentation of the population. Then we design specific
quantitative operations to evaluate the decision-making abilities of LLMs post
role-playing from four aspects: adaptability, exploration$\&$exploitation
trade-off ability, reasoning ability, and safety. Finally, we analyze the
association between the performance of decision-making and the corresponding
MBTI types through GPT-4. Extensive experiments demonstrate stable differences
in the four aspects of decision-making abilities across distinct roles,
signifying a robust correlation between decision-making abilities and the roles
emulated by LLMs. These results underscore that LLMs can effectively
impersonate varied roles while embodying their genuine sociological
characteristics. | cs.CL | On the Decision-Making Abilities in Role-Playing using Large Language Models |
2024-02-29T02:04:00Z | http://arxiv.org/pdf/2403.00839v1 | http://arxiv.org/abs/2403.00839v1 | Xukun Liu, Zhiyuan Peng, Xiaoyuan Yi, Xing Xie, Lirong Xiang, Yuchen Liu, Dongkuan Xu | 2024-02-29T02:04:00Z | cs.AI, cs.CL | While achieving remarkable progress in a broad range of tasks, large language
models (LLMs) remain significantly limited in properly using massive external
tools. Existing in-context learning approaches simply format tools into a list
of plain text descriptions and input them to LLMs, from which, LLMs generate a
sequence of tool calls to solve problems step by step. Such a paradigm ignores
the intrinsic dependency between tools and offloads all reasoning loads to
LLMs, making them restricted to a limited number of specifically designed
tools. It thus remains challenging for LLMs to operate on a library of massive
tools, casting a great limitation when confronted with real-world scenarios.
This paper proposes ToolNet, a plug-and-play framework that scales up the
number of tools to thousands with a moderate increase in token consumption.
ToolNet organizes tools into a directed graph. Each node represents a tool, and
weighted edges denote tool transition. Starting from an initial tool node, an
LLM navigates in the graph by iteratively choosing the next one from its
successors until the task is resolved. Extensive experiments show that ToolNet
can achieve impressive results in challenging multi-hop tool learning datasets
and is resilient to tool failures. | cs.AI | ToolNet: Connecting Large Language Models with Massive Tools via Tool Graph |
2024-02-29T01:58:49Z | http://arxiv.org/pdf/2402.18797v1 | http://arxiv.org/abs/2402.18797v1 | Guande Wu, Jing Qian, Sonia Castelo, Shaoyu Chen, Joao Rulff, Claudio Silva | 2024-02-29T01:58:49Z | cs.HC, cs.CL, H.1.2; I.2.7 | Text presented in augmented reality provides in-situ, real-time information
for users. However, this content can be challenging to apprehend quickly when
engaging in cognitively demanding AR tasks, especially when it is presented on
a head-mounted display. We propose ARTiST, an automatic text simplification
system that uses a few-shot prompt and GPT-3 models to specifically optimize
the text length and semantic content for augmented reality. Developed out of a
formative study that included seven users and three experts, our system
combines a customized error calibration model with a few-shot prompt to
integrate the syntactic, lexical, elaborative, and content simplification
techniques, and generate simplified AR text for head-worn displays. Results
from a 16-user empirical study showed that ARTiST lightens the cognitive load
and improves performance significantly over both unmodified text and text
modified via traditional methods. Our work constitutes a step towards
automating the optimization of batch text data for readability and performance
in augmented reality. | cs.HC | ARTiST: Automated Text Simplification for Task Guidance in Augmented Reality |
2024-02-29T01:49:18Z | http://arxiv.org/pdf/2402.18792v1 | http://arxiv.org/abs/2402.18792v1 | Fangyuan Zhang, Huichi Zhou, Shuangjiao Li, Hongtao Wang | 2024-02-29T01:49:18Z | cs.LG, cs.CL, cs.CR | Deep neural networks have been proven to be vulnerable to adversarial
examples and various methods have been proposed to defend against adversarial
attacks for natural language processing tasks. However, previous defense
methods have limitations in maintaining effective defense while ensuring the
performance of the original task. In this paper, we propose a malicious
perturbation based adversarial training method (MPAT) for building robust deep
neural networks against textual adversarial attacks. Specifically, we construct
a multi-level malicious example generation strategy to generate adversarial
examples with malicious perturbations, which are used instead of original
inputs for model training. Additionally, we employ a novel training objective
function to ensure achieving the defense goal without compromising the
performance on the original task. We conduct comprehensive experiments to
evaluate our defense method by attacking five victim models on three benchmark
datasets. The result demonstrates that our method is more effective against
malicious adversarial attacks compared with previous defense methods while
maintaining or further improving the performance on the original task. | cs.LG | MPAT: Building Robust Deep Neural Networks against Textual Adversarial Attacks |
2024-02-29T01:33:08Z | http://arxiv.org/pdf/2402.18789v1 | http://arxiv.org/abs/2402.18789v1 | Xupeng Miao, Gabriele Oliaro, Xinhao Cheng, Mengdi Wu, Colin Unger, Zhihao Jia | 2024-02-29T01:33:08Z | cs.DC, cs.CL, cs.LG | Parameter-efficient finetuning (PEFT) is a widely used technique to adapt
large language models for different tasks. Service providers typically create
separate systems for users to perform PEFT model finetuning and inference
tasks. This is because existing systems cannot handle workloads that include a
mix of inference and PEFT finetuning requests. As a result, shared GPU
resources are underutilized, leading to inefficiencies. To address this
problem, we present FlexLLM, the first system that can serve inference and
parameter-efficient finetuning requests in the same iteration. Our system
leverages the complementary nature of these two tasks and utilizes shared GPU
resources to run them jointly, using a method called co-serving. To achieve
this, FlexLLM introduces a novel token-level finetuning mechanism, which breaks
down the finetuning computation of a sequence into smaller token-level
computations and uses dependent parallelization and graph pruning, two static
compilation optimizations, to minimize the memory overhead and latency for
co-serving. Compared to existing systems, FlexLLM's co-serving approach reduces
the activation GPU memory overhead by up to 8x, and the end-to-end GPU memory
requirement of finetuning by up to 36% while maintaining a low inference
latency and improving finetuning throughput. For example, under a heavy
inference workload, FlexLLM can still preserve more than 80% of the peak
finetuning throughput, whereas existing systems cannot make any progress with
finetuning. The source code of FlexLLM is publicly available at
https://github.com/flexflow/FlexFlow. | cs.DC | FlexLLM: A System for Co-Serving Large Language Model Inference and Parameter-Efficient Finetuning |
2024-02-29T00:19:13Z | http://arxiv.org/pdf/2402.18766v2 | http://arxiv.org/abs/2402.18766v2 | Rodrigo Santos, João Silva, Luís Gomes, João Rodrigues, António Branco | 2024-03-05T10:44:03Z | cs.CL | To advance the neural decoding of Portuguese, in this paper we present a
fully open Transformer-based, instruction-tuned decoder model that sets a new
state of the art in this respect. To develop this decoder, which we named
Gerv\'asio PT*, a strong LLaMA~2 7B model was used as a starting point, and its
further improvement through additional training was done over language
resources that include new instruction data sets of Portuguese prepared for
this purpose, which are also contributed in this paper. All versions of
Gerv\'asio are open source and distributed for free under an open license,
including for either research or commercial usage, and can be run on
consumer-grade hardware, thus seeking to contribute to the advancement of
research and innovation in language technology for Portuguese. | cs.CL | Advancing Generative AI for Portuguese with Open Decoder Gervásio PT* |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.