publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2024-06-12T13:35:10Z | http://arxiv.org/pdf/2406.08202v1 | http://arxiv.org/abs/2406.08202v1 | Isidora Jeknić, David Schlangen, Alexander Koller | 2024-06-12T13:35:10Z | cs.CL | Collaboration is an integral part of human dialogue. Typical task-oriented
dialogue games assign asymmetric roles to the participants, which limits their
ability to elicit naturalistic role-taking in collaboration and its
negotiation. We present a novel and simple online setup that favors balanced
collaboration: a two-player 2D object placement game in which the players must
negotiate the goal state themselves. We show empirically that human players
exhibit a variety of role distributions, and that balanced collaboration
improves task performance. We also present an LLM-based baseline agent which
demonstrates that automatic playing of our game is an interesting challenge for
artificial systems. | cs.CL | A Dialogue Game for Eliciting Balanced Collaboration |
2024-06-12T13:30:48Z | http://arxiv.org/pdf/2406.12907v1 | http://arxiv.org/abs/2406.12907v1 | Tim Pearce, Jinyeop Song | 2024-06-12T13:30:48Z | cs.LG, cs.CL | Kaplan et al. [2020] (`Kaplan') and Hoffmann et al. [2022] (`Chinchilla')
studied the scaling behavior of transformers trained on next-token language
prediction. These studies produced different estimates for how the number of
parameters ($N$) and training tokens ($D$) should be set to achieve the lowest
possible loss for a given compute budget ($C$). Kaplan: $N_\text{optimal}
\propto C^{0.73}$, Chinchilla: $N_\text{optimal} \propto C^{0.50}$. This note
finds that much of this discrepancy can be attributed to Kaplan counting
non-embedding rather than total parameters, combined with their analysis being
performed at small scale. Simulating the Chinchilla study under these
conditions produces biased scaling coefficients close to Kaplan's. Hence, this
note reaffirms Chinchilla's scaling coefficients, by explaining the cause of
Kaplan's original overestimation. | cs.LG | Reconciling Kaplan and Chinchilla Scaling Laws |
2024-06-12T13:14:19Z | http://arxiv.org/pdf/2406.08183v2 | http://arxiv.org/abs/2406.08183v2 | Micol Spitale, Jiaee Cheong, Hatice Gunes | 2024-06-14T09:34:35Z | cs.CL | Recent studies show bias in many machine learning models for depression
detection, but bias in LLMs for this task remains unexplored. This work
presents the first attempt to investigate the degree of gender bias present in
existing LLMs (ChatGPT, LLaMA 2, and Bard) using both quantitative and
qualitative approaches. From our quantitative evaluation, we found that ChatGPT
performs the best across various performance metrics and LLaMA 2 outperforms
other LLMs in terms of group fairness metrics. As qualitative fairness
evaluation remains an open research question we propose several strategies
(e.g., word count, thematic analysis) to investigate whether and how a
qualitative evaluation can provide valuable insights for bias analysis beyond
what is possible with quantitative evaluation. We found that ChatGPT
consistently provides a more comprehensive, well-reasoned explanation for its
prediction compared to LLaMA 2. We have also identified several themes adopted
by LLMs to qualitatively evaluate gender fairness. We hope our results can be
used as a stepping stone towards future attempts at improving qualitative
evaluation of fairness for LLMs especially for high-stakes tasks such as
depression detection. | cs.CL | Underneath the Numbers: Quantitative and Qualitative Gender Fairness in LLMs for Depression Prediction |
2024-06-12T13:05:27Z | http://arxiv.org/pdf/2406.08173v1 | http://arxiv.org/abs/2406.08173v1 | Huijie Yao, Wengang Zhou, Hao Zhou, Houqiang Li | 2024-06-12T13:05:27Z | cs.CL | Spoken language glossification (SLG) aims to translate the spoken language
text into the sign language gloss, i.e., a written record of sign language. In
this work, we present a framework named $S$emi-$S$upervised $S$poken $L$anguage
$G$lossification ($S^3$LG) for SLG. To tackle the bottleneck of limited
parallel data in SLG, our $S^3$LG incorporates large-scale monolingual spoken
language text into SLG training. The proposed framework follows the
self-training structure that iteratively annotates and learns from pseudo
labels. Considering the lexical similarity and syntactic difference between
sign language and spoken language, our $S^3$LG adopts both the rule-based
heuristic and model-based approach for auto-annotation. During training, we
randomly mix these complementary synthetic datasets and mark their differences
with a special token. As the synthetic data may be less quality, the $S^3$LG
further leverages consistency regularization to reduce the negative impact of
noise in the synthetic data. Extensive experiments are conducted on public
benchmarks to demonstrate the effectiveness of the $S^3$LG. Our code is
available at \url{https://github.com/yaohj11/S3LG}. | cs.CL | Semi-Supervised Spoken Language Glossification |
2024-06-12T12:44:48Z | http://arxiv.org/pdf/2406.08155v1 | http://arxiv.org/abs/2406.08155v1 | Pingzhi Li, Xiaolong Jin, Yu Cheng, Tianlong Chen | 2024-06-12T12:44:48Z | cs.LG, cs.AI, cs.CL | Large Language Models~(LLMs) have become foundational in the realm of natural
language processing, demonstrating performance improvements as model sizes
increase. The Mixture-of-Experts~(MoE) approach offers a promising way to scale
LLMs more efficiently by using fewer computational FLOPs through sparse
activation. However, it suffers from significant memory overheads,
necessitating model compression techniques. Post-training quantization, a
popular method for model compression, proves less effective when directly
applied to MoE models due to MoE's overlooked inherent sparsity. This paper
explores several MoE structure-aware quantization heuristics, ranging from
coarse to fine granularity, from MoE block to individual linear weight. Our
investigations reveal critical principles: different MoE structures (i.e.,
blocks, experts, linear layers) require varying numbers of weight bits for
effective and efficient quantization. Conclusions are supported by extensive
benchmarking across two representative MoE models and six tasks. We further
introduce novel enhancements to more accurately identify the most critical
weights in MoE quantization that necessitate higher bit allocations, including
the linear weight outlier scorer and MoE block scorer. Additionally, subsequent
experiments validate our findings in the context of both weight and activation
quantization. | cs.LG | Examining Post-Training Quantization for Mixture-of-Experts: A Benchmark |
2024-06-12T12:06:32Z | http://arxiv.org/pdf/2406.08124v1 | http://arxiv.org/abs/2406.08124v1 | Duanyu Feng, Bowen Qin, Chen Huang, Youcheng Huang, Zheng Zhang, Wenqiang Lei | 2024-06-12T12:06:32Z | cs.CL, cs.AI | The success of the reward model in distinguishing between responses with
subtle safety differences depends critically on the high-quality preference
dataset, which should capture the fine-grained nuances of harmful and harmless
responses. This motivates the need to develop a dataset involving preference
margins, which accurately quantify how harmless one response is compared to
another. In this paper, we take the first step to propose an effective and
cost-efficient framework to promote the margin-enhanced preference dataset
development. Our framework, Legend, Leverages representation engineering to
annotate preference datasets. It constructs the specific direction within the
LLM's embedding space that represents safety. By leveraging this safety
direction, Legend can then leverage the semantic distances of paired responses
along this direction to annotate margins automatically. We experimentally
demonstrate our effectiveness in both reward modeling and harmless alignment
for LLMs. Legend also stands out for its efficiency, requiring only the
inference time rather than additional training. This efficiency allows for
easier implementation and scalability, making Legend particularly valuable for
practical applications in aligning LLMs with safe conversations. | cs.CL | Legend: Leveraging Representation Engineering to Annotate Safety Margin for Preference Datasets |
2024-06-12T11:52:35Z | http://arxiv.org/pdf/2406.08116v1 | http://arxiv.org/abs/2406.08116v1 | Zile Qiao, Wei Ye, Yong Jiang, Tong Mo, Pengjun Xie, Weiping Li, Fei Huang, Shikun Zhang | 2024-06-12T11:52:35Z | cs.CL, cs.AI | Retrieval-augmented language models (RALMs) have recently shown great
potential in mitigating the limitations of implicit knowledge in LLMs, such as
untimely updating of the latest expertise and unreliable retention of long-tail
knowledge. However, since the external knowledge base, as well as the
retriever, can not guarantee reliability, potentially leading to the knowledge
retrieved not being helpful or even misleading for LLM generation. In this
paper, we introduce Supportiveness-based Knowledge Rewriting (SKR), a robust
and pluggable knowledge rewriter inherently optimized for LLM generation.
Specifically, we introduce the novel concept of "supportiveness"--which
represents how effectively a knowledge piece facilitates downstream tasks--by
considering the perplexity impact of augmented knowledge on the response text
of a white-box LLM. Based on knowledge supportiveness, we first design a
training data curation strategy for our rewriter model, effectively identifying
and filtering out poor or irrelevant rewrites (e.g., with low supportiveness
scores) to improve data efficacy. We then introduce the direct preference
optimization (DPO) algorithm to align the generated rewrites to optimal
supportiveness, guiding the rewriter model to summarize augmented content that
better improves the final response. Comprehensive evaluations across six
popular knowledge-intensive tasks and four LLMs have demonstrated the
effectiveness and superiority of SKR. With only 7B parameters, SKR has shown
better knowledge rewriting capability over GPT-4, the current state-of-the-art
general-purpose LLM. | cs.CL | Supportiveness-based Knowledge Rewriting for Retrieval-augmented Language Modeling |
2024-06-12T11:27:10Z | http://arxiv.org/pdf/2406.08101v2 | http://arxiv.org/abs/2406.08101v2 | Qianli Wang, Tatiana Anikina, Nils Feldhus, Simon Ostermann, Sebastian Möller | 2024-06-13T03:16:47Z | cs.CL | Conversational explainable artificial intelligence (ConvXAI) systems based on
large language models (LLMs) have garnered significant interest from the
research community in natural language processing (NLP) and human-computer
interaction (HCI). Such systems can provide answers to user questions about
explanations in dialogues, have the potential to enhance users' comprehension
and offer more information about the decision-making and generation processes
of LLMs. Currently available ConvXAI systems are based on intent recognition
rather than free chat, as this has been found to be more precise and reliable
in identifying users' intentions. However, the recognition of intents still
presents a challenge in the case of ConvXAI, since little training data exist
and the domain is highly specific, as there is a broad range of XAI methods to
map requests onto. In order to bridge this gap, we present CoXQL, the first
dataset for user intent recognition in ConvXAI, covering 31 intents, seven of
which require filling multiple slots. Subsequently, we enhance an existing
parsing approach by incorporating template validations, and conduct an
evaluation of several LLMs on CoXQL using different parsing strategies. We
conclude that the improved parsing approach (MP+) surpasses the performance of
previous approaches. We also discover that intents with multiple slots remain
highly challenging for LLMs. | cs.CL | CoXQL: A Dataset for Parsing Explanation Requests in Conversational XAI Systems |
2024-06-12T11:27:03Z | http://arxiv.org/pdf/2406.08100v1 | http://arxiv.org/abs/2406.08100v1 | Mingyu Zheng, Xinwei Feng, Qingyi Si, Qiaoqiao She, Zheng Lin, Wenbin Jiang, Weiping Wang | 2024-06-12T11:27:03Z | cs.CL, cs.AI | Although great progress has been made by previous table understanding methods
including recent approaches based on large language models (LLMs), they rely
heavily on the premise that given tables must be converted into a certain text
sequence (such as Markdown or HTML) to serve as model input. However, it is
difficult to access such high-quality textual table representations in some
real-world scenarios, and table images are much more accessible. Therefore, how
to directly understand tables using intuitive visual information is a crucial
and urgent challenge for developing more practical applications. In this paper,
we propose a new problem, multimodal table understanding, where the model needs
to generate correct responses to various table-related requests based on the
given table image. To facilitate both the model training and evaluation, we
construct a large-scale dataset named MMTab, which covers a wide spectrum of
table images, instructions and tasks. On this basis, we develop Table-LLaVA, a
generalist tabular multimodal large language model (MLLM), which significantly
outperforms recent open-source MLLM baselines on 23 benchmarks under held-in
and held-out settings. The code and data is available at this
https://github.com/SpursGoZmy/Table-LLaVA | cs.CL | Multimodal Table Understanding |
2024-06-12T11:16:30Z | http://arxiv.org/pdf/2406.08092v1 | http://arxiv.org/abs/2406.08092v1 | Zhi Qu, Chenchen Ding, Taro Watanabe | 2024-06-12T11:16:30Z | cs.CL | Understanding representation transfer in multilingual neural machine
translation can reveal the representational issue causing the zero-shot
translation deficiency. In this work, we introduce the identity pair, a
sentence translated into itself, to address the lack of the base measure in
multilingual investigations, as the identity pair represents the optimal state
of representation among any language transfers. In our analysis, we demonstrate
that the encoder transfers the source language to the representational subspace
of the target language instead of the language-agnostic state. Thus, the
zero-shot translation deficiency arises because representations are entangled
with other languages and are not transferred effectively to the target
language. Based on our findings, we propose two methods: 1) low-rank
language-specific embedding at the encoder, and 2) language-specific
contrastive learning of the representation at the decoder. The experimental
results on Europarl-15, TED-19, and OPUS-100 datasets show that our methods
substantially enhance the performance of zero-shot translations by improving
language transfer capacity, thereby providing practical evidence to support our
conclusions. | cs.CL | Languages Transferred Within the Encoder: On Representation Transfer in Zero-Shot Multilingual Translation |
2024-06-12T11:04:11Z | http://arxiv.org/pdf/2406.08080v1 | http://arxiv.org/abs/2406.08080v1 | Pia Pachinger, Janis Goldzycher, Anna Maria Planitzer, Wojciech Kusa, Allan Hanbury, Julia Neidhardt | 2024-06-12T11:04:11Z | cs.CL, cs.AI, cs.CY, I.2.7 | Model interpretability in toxicity detection greatly profits from token-level
annotations. However, currently such annotations are only available in English.
We introduce a dataset annotated for offensive language detection sourced from
a news forum, notable for its incorporation of the Austrian German dialect,
comprising 4,562 user comments. In addition to binary offensiveness
classification, we identify spans within each comment constituting vulgar
language or representing targets of offensive statements. We evaluate
fine-tuned language models as well as large language models in a zero- and
few-shot fashion. The results indicate that while fine-tuned models excel in
detecting linguistic peculiarities such as vulgar dialect, large language
models demonstrate superior performance in detecting offensiveness in
AustroTox. We publish the data and code. | cs.CL | AustroTox: A Dataset for Target-Based Austrian German Offensive Language Detection |
2024-06-12T10:48:53Z | http://arxiv.org/pdf/2406.08074v1 | http://arxiv.org/abs/2406.08074v1 | Jayneel Parekh, Pegah Khayatan, Mustafa Shukor, Alasdair Newson, Matthieu Cord | 2024-06-12T10:48:53Z | cs.LG, cs.AI, cs.CL, cs.CV | Large multimodal models (LMMs) combine unimodal encoders and large language
models (LLMs) to perform multimodal tasks. Despite recent advancements towards
the interpretability of these models, understanding internal representations of
LMMs remains largely a mystery. In this paper, we present a novel framework for
the interpretation of LMMs. We propose a dictionary learning based approach,
applied to the representation of tokens. The elements of the learned dictionary
correspond to our proposed concepts. We show that these concepts are well
semantically grounded in both vision and text. Thus we refer to these as
"multi-modal concepts". We qualitatively and quantitatively evaluate the
results of the learnt concepts. We show that the extracted multimodal concepts
are useful to interpret representations of test samples. Finally, we evaluate
the disentanglement between different concepts and the quality of grounding
concepts visually and textually. We will publicly release our code. | cs.LG | A Concept-Based Explainability Framework for Large Multimodal Models |
2024-06-12T10:36:27Z | http://arxiv.org/pdf/2406.08068v1 | http://arxiv.org/abs/2406.08068v1 | Hao Yang, Yanyan Zhao, Yang Wu, Shilong Wang, Tian Zheng, Hongbo Zhang, Wanxiang Che, Bing Qin | 2024-06-12T10:36:27Z | cs.CL | Compared to traditional sentiment analysis, which only considers text,
multimodal sentiment analysis needs to consider emotional signals from
multimodal sources simultaneously and is therefore more consistent with the way
how humans process sentiment in real-world scenarios. It involves processing
emotional information from various sources such as natural language, images,
videos, audio, physiological signals, etc. However, although other modalities
also contain diverse emotional cues, natural language usually contains richer
contextual information and therefore always occupies a crucial position in
multimodal sentiment analysis. The emergence of ChatGPT has opened up immense
potential for applying large language models (LLMs) to text-centric multimodal
tasks. However, it is still unclear how existing LLMs can adapt better to
text-centric multimodal sentiment analysis tasks. This survey aims to (1)
present a comprehensive review of recent research in text-centric multimodal
sentiment analysis tasks, (2) examine the potential of LLMs for text-centric
multimodal sentiment analysis, outlining their approaches, advantages, and
limitations, (3) summarize the application scenarios of LLM-based multimodal
sentiment analysis technology, and (4) explore the challenges and potential
research directions for multimodal sentiment analysis in the future. | cs.CL | Large Language Models Meet Text-Centric Multimodal Sentiment Analysis: A Survey |
2024-06-12T10:12:52Z | http://arxiv.org/pdf/2406.08055v1 | http://arxiv.org/abs/2406.08055v1 | Napat Laosaengpha, Thanit Tativannarat, Chawan Piansaddhayanon, Attapol Rutherford, Ekapol Chuangsuwanich | 2024-06-12T10:12:52Z | cs.CL | Learning job title representation is a vital process for developing automatic
human resource tools. To do so, existing methods primarily rely on learning the
title representation through skills extracted from the job description,
neglecting the rich and diverse content within. Thus, we propose an alternative
framework for learning job titles through their respective job description (JD)
and utilize a Job Description Aggregator component to handle the lengthy
description and bidirectional contrastive loss to account for the bidirectional
relationship between the job title and its description. We evaluated the
performance of our method on both in-domain and out-of-domain settings,
achieving a superior performance over the skill-based approach. | cs.CL | Learning Job Title Representation from Job Description Aggregation Network |
2024-06-12T10:02:27Z | http://arxiv.org/pdf/2406.08050v1 | http://arxiv.org/abs/2406.08050v1 | João Vitorino, Eva Maia, Isabel Praça | 2024-06-12T10:02:27Z | cs.CL, cs.LG | Large Language Models (LLMs) are valuable for text classification, but their
vulnerabilities must not be disregarded. They lack robustness against
adversarial examples, so it is pertinent to understand the impacts of different
types of perturbations, and assess if those attacks could be replicated by
common users with a small amount of perturbations and a small number of queries
to a deployed LLM. This work presents an analysis of the effectiveness,
efficiency, and practicality of three different types of adversarial attacks
against five different LLMs in a sentiment classification task. The obtained
results demonstrated the very distinct impacts of the word-level and
character-level attacks. The word attacks were more effective, but the
character and more constrained attacks were more practical and required a
reduced number of perturbations and queries. These differences need to be
considered during the development of adversarial defense strategies to train
more robust LLMs for intelligent text classification applications. | cs.CL | Adversarial Evasion Attack Efficiency against Large Language Models |
2024-06-12T08:56:46Z | http://arxiv.org/pdf/2406.10284v1 | http://arxiv.org/abs/2406.10284v1 | Yuanyuan Zhang, Zhengjun Yue, Tanvina Patel, Odette Scharenborg | 2024-06-12T08:56:46Z | cs.CL, cs.SD, eess.AS | State-of-the-art ASRs show suboptimal performance for child speech. The
scarcity of child speech limits the development of child speech recognition
(CSR). Therefore, we studied child-to-child voice conversion (VC) from existing
child speakers in the dataset and additional (new) child speakers via
monolingual and cross-lingual (Dutch-to-German) VC, respectively. The results
showed that cross-lingual child-to-child VC significantly improved child ASR
performance. Experiments on the impact of the quantity of child-to-child
cross-lingual VC-generated data on fine-tuning (FT) ASR models gave the best
results with two-fold augmentation for our FT-Conformer model and FT-Whisper
model which reduced WERs with ~3% absolute compared to the baseline, and with
six-fold augmentation for the model trained from scratch, which improved by an
absolute 3.6% WER. Moreover, using a small amount of "high-quality"
VC-generated data achieved similar results to those of our best-FT models. | cs.CL | Improving child speech recognition with augmented child-like speech |
2024-06-12T08:27:44Z | http://arxiv.org/pdf/2406.10283v1 | http://arxiv.org/abs/2406.10283v1 | Zihan Pan, Tianchi Liu, Hardik B. Sailor, Qiongqiong Wang | 2024-06-12T08:27:44Z | cs.CL, cs.SD, eess.AS | Self-supervised learning (SSL) speech representation models, trained on large
speech corpora, have demonstrated effectiveness in extracting hierarchical
speech embeddings through multiple transformer layers. However, the behavior of
these embeddings in specific tasks remains uncertain. This paper investigates
the multi-layer behavior of the WavLM model in anti-spoofing and proposes an
attentive merging method to leverage the hierarchical hidden embeddings.
Results demonstrate the feasibility of fine-tuning WavLM to achieve the best
equal error rate (EER) of 0.65%, 3.50%, and 3.19% on the ASVspoof 2019LA,
2021LA, and 2021DF evaluation sets, respectively. Notably, We find that the
early hidden transformer layers of the WavLM large model contribute
significantly to anti-spoofing task, enabling computational efficiency by
utilizing a partial pre-trained model. | cs.CL | Attentive Merging of Hidden Embeddings from Pre-trained Speech Model for Anti-spoofing Detection |
2024-06-12T08:26:30Z | http://arxiv.org/pdf/2406.07990v1 | http://arxiv.org/abs/2406.07990v1 | Thomas Roland Barillot, Alex De Castro | 2024-06-12T08:26:30Z | cs.LG, cs.AI, cs.CL | This works reports evidence for the topological signatures of ambiguity in
sentence embeddings that could be leveraged for ranking and/or explanation
purposes in the context of vector search and Retrieval Augmented Generation
(RAG) systems. We proposed a working definition of ambiguity and designed an
experiment where we have broken down a proprietary dataset into collections of
chunks of varying size - 3, 5, and 10 lines and used the different collections
successively as queries and answers sets. It allowed us to test the signatures
of ambiguity with removal of confounding factors. Our results show that proxy
ambiguous queries (size 10 queries against size 3 documents) display different
distributions of homologies 0 and 1 based features than proxy clear queries
(size 5 queries against size 10 documents). We then discuss those results in
terms increased manifold complexity and/or approximately discontinuous
embedding submanifolds. Finally we propose a strategy to leverage those
findings as a new scoring strategy of semantic similarities. | cs.LG | Blowfish: Topological and statistical signatures for quantifying ambiguity in semantic search |
2024-06-12T07:52:17Z | http://arxiv.org/pdf/2406.07971v2 | http://arxiv.org/abs/2406.07971v2 | Taiming Lu, Lingfeng Shen, Xinyu Yang, Weiting Tan, Beidi Chen, Huaxiu Yao | 2024-06-13T05:13:50Z | cs.CL, cs.AI, cs.LG | Reinforcement Learning from Human Feedback (RLHF) involves training policy
models (PMs) and reward models (RMs) to align language models with human
preferences. Instead of focusing solely on PMs and RMs independently, we
propose to examine their interactions during fine-tuning, introducing the
concept of seamlessness. Our study starts with observing the saturation
phenomenon, where continual improvements in RM and PM do not translate into
RLHF progress. Our analysis shows that RMs fail to assign proper scores to PM
responses, resulting in a 35% mismatch rate with human preferences,
highlighting a significant discrepancy between PM and RM. To measure
seamlessness between PM and RM without human effort, we propose an automatic
metric, SEAM. SEAM quantifies the discrepancies between PM and RM judgments
induced by data samples. We validate the effectiveness of SEAM in data
selection and model augmentation. Our experiments demonstrate that (1) using
SEAM-filtered data for RL training improves RLHF performance by 4.5%, and (2)
SEAM-guided model augmentation results in a 4% performance improvement over
standard augmentation methods. | cs.CL | It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF |
2024-06-12T07:49:36Z | http://arxiv.org/pdf/2406.07970v1 | http://arxiv.org/abs/2406.07970v1 | Javad Pourmostafa Roshan Sharami, Dimitar Shterionov, Pieter Spronck | 2024-06-12T07:49:36Z | cs.CL | The quality of output from large language models (LLMs), particularly in
machine translation (MT), is closely tied to the quality of in-context examples
(ICEs) provided along with the query, i.e., the text to translate. The
effectiveness of these ICEs is influenced by various factors, such as the
domain of the source text, the order in which the ICEs are presented, the
number of these examples, and the prompt templates used. Naturally, selecting
the most impactful ICEs depends on understanding how these affect the resulting
translation quality, which ultimately relies on translation references or human
judgment. This paper presents a novel methodology for in-context learning (ICL)
that relies on a search algorithm guided by domain-specific quality estimation
(QE). Leveraging the XGLM model, our methodology estimates the resulting
translation quality without the need for translation references, selecting
effective ICEs for MT to maximize translation quality. Our results demonstrate
significant improvements over existing ICL methods and higher translation
performance compared to fine-tuning a pre-trained language model (PLM),
specifically mBART-50. | cs.CL | Guiding In-Context Learning of LLMs through Quality Estimation for Machine Translation |
2024-06-12T07:49:21Z | http://arxiv.org/pdf/2406.07969v1 | http://arxiv.org/abs/2406.07969v1 | Masaya Kawamura, Ryuichi Yamamoto, Yuma Shirahata, Takuya Hasumi, Kentaro Tachibana | 2024-06-12T07:49:21Z | eess.AS, cs.CL, cs.LG, cs.SD | We introduce LibriTTS-P, a new corpus based on LibriTTS-R that includes
utterance-level descriptions (i.e., prompts) of speaking style and
speaker-level prompts of speaker characteristics. We employ a hybrid approach
to construct prompt annotations: (1) manual annotations that capture human
perceptions of speaker characteristics and (2) synthetic annotations on
speaking style. Compared to existing English prompt datasets, our corpus
provides more diverse prompt annotations for all speakers of LibriTTS-R.
Experimental results for prompt-based controllable TTS demonstrate that the TTS
model trained with LibriTTS-P achieves higher naturalness than the model using
the conventional dataset. Furthermore, the results for style captioning tasks
show that the model utilizing LibriTTS-P generates 2.5 times more accurate
words than the model using a conventional dataset. Our corpus, LibriTTS-P, is
available at https://github.com/line/LibriTTS-P. | eess.AS | LibriTTS-P: A Corpus with Speaking Style and Speaker Identity Prompts for Text-to-Speech and Style Captioning |
2024-06-12T07:44:36Z | http://arxiv.org/pdf/2406.07967v1 | http://arxiv.org/abs/2406.07967v1 | Jie Ruan, Xiao Pu, Mingqi Gao, Xiaojun Wan, Yuesheng Zhu | 2024-06-12T07:44:36Z | cs.CL, cs.LG | Human evaluation is viewed as a reliable evaluation method for NLG which is
expensive and time-consuming. To save labor and costs, researchers usually
perform human evaluation on a small subset of data sampled from the whole
dataset in practice. However, different selection subsets will lead to
different rankings of the systems. To give a more correct inter-system ranking
and make the gold standard human evaluation more reliable, we propose a
Constrained Active Sampling Framework (CASF) for reliable human judgment. CASF
operates through a Learner, a Systematic Sampler and a Constrained Controller
to select representative samples for getting a more correct inter-system
ranking.Experiment results on 137 real NLG evaluation setups with 44 human
evaluation metrics across 16 datasets and 5 NLG tasks demonstrate CASF receives
93.18% top-ranked system recognition accuracy and ranks first or ranks second
on 90.91% of the human metrics with 0.83 overall inter-system ranking Kendall
correlation.Code and data are publicly available online. | cs.CL | Better than Random: Reliable NLG Human Evaluation with Constrained Active Sampling |
2024-06-12T07:42:12Z | http://arxiv.org/pdf/2406.07964v1 | http://arxiv.org/abs/2406.07964v1 | Joseba Fernandez de Landa, Rodrigo Agerri | 2024-06-12T07:42:12Z | cs.SI, cs.CL, cs.CY | Social media users express their political preferences via interaction with
other users, by spontaneous declarations or by participation in communities
within the network. This makes a social network such as Twitter a valuable data
source to study computational science approaches to political learning
inference. In this work we focus on three diverse regions in Spain (Basque
Country, Catalonia and Galicia) to explore various methods for multi-party
categorization, required to analyze evolving and complex political landscapes,
and compare it with binary left-right approaches. We use a two-step method
involving unsupervised user representations obtained from the retweets and
their subsequent use for political leaning detection. Comprehensive
experimentation on a newly collected and curated dataset comprising labeled
users and their interactions demonstrate the effectiveness of using Relational
Embeddings as representation method for political ideology detection in both
binary and multi-party frameworks, even with limited training data. Finally,
data visualization illustrates the ability of the Relational Embeddings to
capture intricate intra-group and inter-group political affinities. | cs.SI | Political Leaning Inference through Plurinational Scenarios |
2024-06-12T07:41:44Z | http://arxiv.org/pdf/2406.07962v1 | http://arxiv.org/abs/2406.07962v1 | Luis Miguel Vieira da Silva, Aljosha Köcher, Felix Gehlhoff, Alexander Fay | 2024-06-12T07:41:44Z | cs.AI, cs.CL | To achieve a flexible and adaptable system, capability ontologies are
increasingly leveraged to describe functions in a machine-interpretable way.
However, modeling such complex ontological descriptions is still a manual and
error-prone task that requires a significant amount of effort and ontology
expertise. This contribution presents an innovative method to automate
capability ontology modeling using Large Language Models (LLMs), which have
proven to be well suited for such tasks. Our approach requires only a natural
language description of a capability, which is then automatically inserted into
a predefined prompt using a few-shot prompting technique. After prompting an
LLM, the resulting capability ontology is automatically verified through
various steps in a loop with the LLM to check the overall correctness of the
capability ontology. First, a syntax check is performed, then a check for
contradictions, and finally a check for hallucinations and missing ontology
elements. Our method greatly reduces manual effort, as only the initial natural
language description and a final human review and possible correction are
necessary, thereby streamlining the capability ontology generation process. | cs.AI | Toward a Method to Generate Capability Ontologies from Natural Language Descriptions |
2024-06-12T06:59:31Z | http://arxiv.org/pdf/2406.07935v1 | http://arxiv.org/abs/2406.07935v1 | Jie Ruan, Wenqing Wang, Xiaojun Wan | 2024-06-12T06:59:31Z | cs.CL, cs.LG | Human evaluation serves as the gold standard for assessing the quality of
Natural Language Generation (NLG) systems. Nevertheless, the evaluation
guideline, as a pivotal element ensuring reliable and reproducible human
assessment, has received limited attention.Our investigation revealed that only
29.84% of recent papers involving human evaluation at top conferences release
their evaluation guidelines, with vulnerabilities identified in 77.09% of these
guidelines. Unreliable evaluation guidelines can yield inaccurate assessment
outcomes, potentially impeding the advancement of NLG in the right direction.
To address these challenges, we take an initial step towards reliable
evaluation guidelines and propose the first human evaluation guideline dataset
by collecting annotations of guidelines extracted from existing papers as well
as generated via Large Language Models (LLMs). We then introduce a taxonomy of
eight vulnerabilities and formulate a principle for composing evaluation
guidelines. Furthermore, a method for detecting guideline vulnerabilities has
been explored using LLMs, and we offer a set of recommendations to enhance
reliability in human evaluation. The annotated human evaluation guideline
dataset and code for the vulnerability detection method are publicly available
online. | cs.CL | Defining and Detecting Vulnerability in Human Evaluation Guidelines: A Preliminary Study Towards Reliable NLG Evaluation |
2024-06-12T06:56:20Z | http://arxiv.org/pdf/2406.07933v1 | http://arxiv.org/abs/2406.07933v1 | Chris Yuhao Liu, Yaxuan Wang, Jeffrey Flanigan, Yang Liu | 2024-06-12T06:56:20Z | cs.CL, cs.AI, cs.LG | Large language models (LLMs) have advanced to encompass extensive knowledge
across diverse domains. Yet controlling what a large language model should not
know is important for ensuring alignment and thus safe use. However, accurately
and efficiently unlearning knowledge from an LLM remains challenging due to the
potential collateral damage caused by the fuzzy boundary between retention and
forgetting, and the large computational requirements for optimization across
state-of-the-art models with hundreds of billions of parameters. In this work,
we present Embedding-COrrupted (ECO) Prompts, a lightweight unlearning
framework for large language models to address both the challenges of knowledge
entanglement and unlearning efficiency. Instead of relying on the LLM itself to
unlearn, we enforce an unlearned state during inference by employing a prompt
classifier to identify and safeguard prompts to forget. We learn corruptions
added to prompt embeddings via zeroth order optimization toward the unlearning
objective offline and corrupt prompts flagged by the classifier during
inference. We find that these embedding-corrupted prompts not only lead to
desirable outputs that satisfy the unlearning objective but also closely
approximate the output from a model that has never been trained on the data
intended for forgetting. Through extensive experiments on unlearning, we
demonstrate the superiority of our method in achieving promising unlearning at
nearly zero side effects in general domains and domains closely related to the
unlearned ones. Additionally, we highlight the scalability of our method to 100
LLMs, ranging from 0.5B to 236B parameters, incurring no additional cost as the
number of parameters increases. | cs.CL | Large Language Model Unlearning via Embedding-Corrupted Prompts |
2024-06-12T06:44:05Z | http://arxiv.org/pdf/2406.07922v1 | http://arxiv.org/abs/2406.07922v1 | Dongsuk Jang, Hyeryun Park, Jiye Son, Hyeonuk Hwang, Sujin Kim, Jinwook Choi | 2024-06-12T06:44:05Z | cs.CL | In the rapidly evolving field of healthcare, the integration of artificial
intelligence (AI) has become a pivotal component in the automation of clinical
workflows, ushering in a new era of efficiency and accuracy. This study focuses
on the transformative capabilities of the fine-tuned KoELECTRA model in
comparison to the GPT-4 model, aiming to facilitate automated information
extraction from thyroid operation narratives. The current research landscape is
dominated by traditional methods heavily reliant on regular expressions, which
often face challenges in processing free-style text formats containing critical
details of operation records, including frozen biopsy reports. Addressing this,
the study leverages advanced natural language processing (NLP) techniques to
foster a paradigm shift towards more sophisticated data processing systems.
Through this comparative study, we aspire to unveil a more streamlined,
precise, and efficient approach to document processing in the healthcare
domain, potentially revolutionizing the way medical data is handled and
analyzed. | cs.CL | Automated Information Extraction from Thyroid Operation Narrative: A Comparative Study of GPT-4 and Fine-tuned KoELECTRA |
2024-06-12T06:43:59Z | http://arxiv.org/pdf/2406.16925v1 | http://arxiv.org/abs/2406.16925v1 | Jingwei Wang | 2024-06-12T06:43:59Z | cs.CL, cs.AI | This project investigates the behavior of multi-head attention in Transformer
models, specifically focusing on the differences between benign and trojan
models in the context of sentiment analysis. Trojan attacks cause models to
perform normally on clean inputs but exhibit misclassifications when presented
with inputs containing predefined triggers. We characterize attention head
functions in trojan and benign models, identifying specific 'trojan' heads and
analyzing their behavior. | cs.CL | Analyzing Multi-Head Attention on Trojan BERT Models |
2024-06-12T06:33:54Z | http://arxiv.org/pdf/2406.07913v1 | http://arxiv.org/abs/2406.07913v1 | Yuxi Feng, Raymond Li, Zhenan Fan, Giuseppe Carenini, Mohammadreza Pourreza, Weiwei Zhang, Yong Zhang | 2024-06-12T06:33:54Z | cs.CL, cs.IR | While in-context Learning (ICL) has proven to be an effective technique to
improve the performance of Large Language Models (LLMs) in a variety of complex
tasks, notably in translating natural language questions into Structured Query
Language (NL2SQL), the question of how to select the most beneficial
demonstration examples remains an open research problem. While prior works
often adapted off-the-shelf encoders to retrieve examples dynamically, an
inherent discrepancy exists in the representational capacities between the
external retrievers and the LLMs. Further, optimizing the selection of examples
is a non-trivial task, since there are no straightforward methods to assess the
relative benefits of examples without performing pairwise inference. To address
these shortcomings, we propose DeTriever, a novel demonstration retrieval
framework that learns a weighted combination of LLM hidden states, where rich
semantic information is encoded. To train the model, we propose a proxy score
that estimates the relative benefits of examples based on the similarities
between output queries. Experiments on two popular NL2SQL benchmarks
demonstrate that our method significantly outperforms the state-of-the-art
baselines on one-shot NL2SQL tasks. | cs.CL | DeTriever: Decoder-representation-based Retriever for Improving NL2SQL In-Context Learning |
2024-06-12T06:22:52Z | http://arxiv.org/pdf/2406.07909v1 | http://arxiv.org/abs/2406.07909v1 | Eungbeom Kim, Hantae Kim, Kyogu Lee | 2024-06-12T06:22:52Z | eess.AS, cs.CL, cs.SD, stat.ML | Transformer encoder with connectionist temporal classification (CTC)
framework is widely used for automatic speech recognition (ASR). However,
knowledge distillation (KD) for ASR displays a problem of disagreement between
teacher-student models in frame-level alignment which ultimately hinders it
from improving the student model's performance. In order to resolve this
problem, this paper introduces a self-knowledge distillation (SKD) method that
guides the frame-level alignment during the training time. In contrast to the
conventional method using separate teacher and student models, this study
introduces a simple and effective method sharing encoder layers and applying
the sub-model as the student model. Overall, our approach is effective in
improving both the resource efficiency as well as performance. We also
conducted an experimental analysis of the spike timings to illustrate that the
proposed method improves performance by reducing the alignment disagreement. | eess.AS | Guiding Frame-Level CTC Alignments Using Self-knowledge Distillation |
2024-06-12T06:06:55Z | http://arxiv.org/pdf/2406.07900v1 | http://arxiv.org/abs/2406.07900v1 | Bulat Khaertdinov, Pedro Jeuris, Annanda Sousa, Enrique Hortal | 2024-06-12T06:06:55Z | cs.CL, cs.AI, cs.SD, eess.AS | Recent advancements in Deep and Self-Supervised Learning (SSL) have led to
substantial improvements in Speech Emotion Recognition (SER) performance,
reaching unprecedented levels. However, obtaining sufficient amounts of
accurately labeled data for training or fine-tuning the models remains a costly
and challenging task. In this paper, we propose a multi-view SSL pre-training
technique that can be applied to various representations of speech, including
the ones generated by large speech models, to improve SER performance in
scenarios where annotations are limited. Our experiments, based on wav2vec 2.0,
spectral and paralinguistic features, demonstrate that the proposed framework
boosts the SER performance, by up to 10% in Unweighted Average Recall, in
settings with extremely sparse data annotations. | cs.CL | Exploring Self-Supervised Multi-view Contrastive Learning for Speech Emotion Recognition with Limited Annotations |
2024-06-12T05:41:01Z | http://arxiv.org/pdf/2406.07890v1 | http://arxiv.org/abs/2406.07890v1 | Anfeng Xu, Kevin Huang, Tiantian Feng, Lue Shen, Helen Tager-Flusberg, Shrikanth Narayanan | 2024-06-12T05:41:01Z | eess.AS, cs.CL, cs.LG | Speech foundation models, trained on vast datasets, have opened unique
opportunities in addressing challenging low-resource speech understanding, such
as child speech. In this work, we explore the capabilities of speech foundation
models on child-adult speaker diarization. We show that exemplary foundation
models can achieve 39.5% and 62.3% relative reductions in Diarization Error
Rate and Speaker Confusion Rate, respectively, compared to previous speaker
diarization methods. In addition, we benchmark and evaluate the speaker
diarization results of the speech foundation models with varying the input
audio window size, speaker demographics, and training data ratio. Our results
highlight promising pathways for understanding and adopting speech foundation
models to facilitate child speech understanding. | eess.AS | Exploring Speech Foundation Models for Speaker Diarization in Child-Adult Dyadic Interactions |
2024-06-12T05:25:15Z | http://arxiv.org/pdf/2406.07887v1 | http://arxiv.org/abs/2406.07887v1 | Roger Waleffe, Wonmin Byeon, Duncan Riach, Brandon Norick, Vijay Korthikanti, Tri Dao, Albert Gu, Ali Hatamizadeh, Sudhakar Singh, Deepak Narayanan, Garvit Kulshreshtha, Vartika Singh, Jared Casper, Jan Kautz, Mohammad Shoeybi, Bryan Catanzaro | 2024-06-12T05:25:15Z | cs.LG, cs.CL | Selective state-space models (SSMs) like Mamba overcome some of the
shortcomings of Transformers, such as quadratic computational complexity with
sequence length and large inference-time memory requirements from the key-value
cache. Moreover, recent studies have shown that SSMs can match or exceed the
language modeling capabilities of Transformers, making them an attractive
alternative. In a controlled setting (e.g., same data), however, studies so far
have only presented small scale experiments comparing SSMs to Transformers. To
understand the strengths and weaknesses of these architectures at larger
scales, we present a direct comparison between 8B-parameter Mamba, Mamba-2, and
Transformer models trained on the same datasets of up to 3.5T tokens. We also
compare these models to a hybrid architecture consisting of 43% Mamba-2, 7%
attention, and 50% MLP layers (Mamba-2-Hybrid). Using a diverse set of tasks,
we answer the question of whether Mamba models can match Transformers at larger
training budgets. Our results show that while pure SSMs match or exceed
Transformers on many tasks, they lag behind Transformers on tasks which require
strong copying or in-context learning abilities (e.g., 5-shot MMLU, Phonebook)
or long-context reasoning. In contrast, we find that the 8B Mamba-2-Hybrid
exceeds the 8B Transformer on all 12 standard tasks we evaluated (+2.65 points
on average) and is predicted to be up to 8x faster when generating tokens at
inference time. To validate long-context capabilities, we provide additional
experiments evaluating variants of the Mamba-2-Hybrid and Transformer extended
to support 16K, 32K, and 128K sequences. On an additional 23 long-context
tasks, the hybrid model continues to closely match or exceed the Transformer on
average. To enable further study, we release the checkpoints as well as the
code used to train our models as part of NVIDIA's Megatron-LM project. | cs.LG | An Empirical Study of Mamba-based Language Models |
2024-06-12T05:24:58Z | http://arxiv.org/pdf/2406.07886v1 | http://arxiv.org/abs/2406.07886v1 | Jaehoon Kim, Seungwan Jin, Sohyun Park, Someen Park, Kyungsik Han | 2024-06-12T05:24:58Z | cs.CL | Detecting implicit hate speech that is not directly hateful remains a
challenge. Recent research has attempted to detect implicit hate speech by
applying contrastive learning to pre-trained language models such as BERT and
RoBERTa, but the proposed models still do not have a significant advantage over
cross-entropy loss-based learning. We found that contrastive learning based on
randomly sampled batch data does not encourage the model to learn hard negative
samples. In this work, we propose Label-aware Hard Negative sampling strategies
(LAHN) that encourage the model to learn detailed features from hard negative
samples, instead of naive negative samples in random batch, using
momentum-integrated contrastive learning. LAHN outperforms the existing models
for implicit hate speech detection both in- and cross-datasets. The code is
available at https://github.com/Hanyang-HCC-Lab/LAHN | cs.CL | Label-aware Hard Negative Sampling Strategies with Momentum Contrastive Learning for Implicit Hate Speech Detection |
2024-06-12T05:20:16Z | http://arxiv.org/pdf/2406.07882v2 | http://arxiv.org/abs/2406.07882v2 | Yida Chen, Aoyu Wu, Trevor DePodesta, Catherine Yeh, Kenneth Li, Nicholas Castillo Marin, Oam Patel, Jan Riecke, Shivam Raval, Olivia Seow, Martin Wattenberg, Fernanda Viégas | 2024-06-15T05:06:29Z | cs.CL, cs.AI, cs.HC | Conversational LLMs function as black box systems, leaving users guessing
about why they see the output they do. This lack of transparency is potentially
problematic, especially given concerns around bias and truthfulness. To address
this issue, we present an end-to-end prototype-connecting interpretability
techniques with user experience design-that seeks to make chatbots more
transparent. We begin by showing evidence that a prominent open-source LLM has
a "user model": examining the internal state of the system, we can extract data
related to a user's age, gender, educational level, and socioeconomic status.
Next, we describe the design of a dashboard that accompanies the chatbot
interface, displaying this user model in real time. The dashboard can also be
used to control the user model and the system's behavior. Finally, we discuss a
study in which users conversed with the instrumented system. Our results
suggest that users appreciate seeing internal states, which helped them expose
biased behavior and increased their sense of control. Participants also made
valuable suggestions that point to future directions for both design and
machine learning research. The project page and video demo of our TalkTuner
system are available at https://bit.ly/talktuner-project-page | cs.CL | Designing a Dashboard for Transparency and Control of Conversational AI |
2024-06-12T05:13:09Z | http://arxiv.org/pdf/2406.10281v1 | http://arxiv.org/abs/2406.10281v1 | Patrick Chao, Edgar Dobriban, Hamed Hassani | 2024-06-12T05:13:09Z | cs.CR, cs.CL, cs.LG | Recent progress in large language models enables the creation of realistic
machine-generated content. Watermarking is a promising approach to distinguish
machine-generated text from human text, embedding statistical signals in the
output that are ideally undetectable to humans. We propose a watermarking
framework that encodes such signals through an error correcting code. Our
method, termed robust binary code (RBC) watermark, introduces no distortion
compared to the original probability distribution, and no noticeable
degradation in quality. We evaluate our watermark on base and instruction
fine-tuned models and find our watermark is robust to edits, deletions, and
translations. We provide an information-theoretic perspective on watermarking,
a powerful statistical test for detection and for generating p-values, and
theoretical guarantees. Our empirical findings suggest our watermark is fast,
powerful, and robust, comparing favorably to the state-of-the-art. | cs.CR | Watermarking Language Models with Error Correcting Codes |
2024-06-12T05:09:58Z | http://arxiv.org/pdf/2406.10280v1 | http://arxiv.org/abs/2406.10280v1 | Yu-Hsiang Huang, Yuche Tsai, Hsiang Hsiao, Hong-Yi Lin, Shou-De Lin | 2024-06-12T05:09:58Z | cs.CR, cs.CL, cs.LG | This study investigates the privacy risks associated with text embeddings,
focusing on the scenario where attackers cannot access the original embedding
model. Contrary to previous research requiring direct model access, we explore
a more realistic threat model by developing a transfer attack method. This
approach uses a surrogate model to mimic the victim model's behavior, allowing
the attacker to infer sensitive information from text embeddings without direct
access. Our experiments across various embedding models and a clinical dataset
demonstrate that our transfer attack significantly outperforms traditional
methods, revealing the potential privacy vulnerabilities in embedding
technologies and emphasizing the need for enhanced security measures. | cs.CR | Transferable Embedding Inversion Attack: Uncovering Privacy Risks in Text Embeddings without Model Queries |
2024-06-12T04:22:27Z | http://arxiv.org/pdf/2406.07860v1 | http://arxiv.org/abs/2406.07860v1 | Rahul Kumar, Amar Raja Dibbu, Shrutendra Harsola, Vignesh Subrahmaniam, Ashutosh Modi | 2024-06-12T04:22:27Z | cs.CL, cs.AI, cs.LG | Several large-scale datasets (e.g., WikiSQL, Spider) for developing natural
language interfaces to databases have recently been proposed. These datasets
cover a wide breadth of domains but fall short on some essential domains, such
as finance and accounting. Given that accounting databases are used worldwide,
particularly by non-technical people, there is an imminent need to develop
models that could help extract information from accounting databases via
natural language queries. In this resource paper, we aim to fill this gap by
proposing a new large-scale Text-to-SQL dataset for the accounting and
financial domain: BookSQL. The dataset consists of 100k natural language
queries-SQL pairs, and accounting databases of 1 million records. We experiment
with and analyze existing state-of-the-art models (including GPT-4) for the
Text-to-SQL task on BookSQL. We find significant performance gaps, thus
pointing towards developing more focused models for this domain. | cs.CL | BookSQL: A Large Scale Text-to-SQL Dataset for Accounting Domain |
2024-06-12T04:09:44Z | http://arxiv.org/pdf/2406.07855v1 | http://arxiv.org/abs/2406.07855v1 | Bing Han, Long Zhou, Shujie Liu, Sanyuan Chen, Lingwei Meng, Yanming Qian, Yanqing Liu, Sheng Zhao, Jinyu Li, Furu Wei | 2024-06-12T04:09:44Z | cs.CL, cs.SD, eess.AS | With the help of discrete neural audio codecs, large language models (LLM)
have increasingly been recognized as a promising methodology for zero-shot
Text-to-Speech (TTS) synthesis. However, sampling based decoding strategies
bring astonishing diversity to generation, but also pose robustness issues such
as typos, omissions and repetition. In addition, the high sampling rate of
audio also brings huge computational overhead to the inference process of
autoregression. To address these issues, we propose VALL-E R, a robust and
efficient zero-shot TTS system, building upon the foundation of VALL-E.
Specifically, we introduce a phoneme monotonic alignment strategy to strengthen
the connection between phonemes and acoustic sequence, ensuring a more precise
alignment by constraining the acoustic tokens to match their associated
phonemes. Furthermore, we employ a codec-merging approach to downsample the
discrete codes in shallow quantization layer, thereby accelerating the decoding
speed while preserving the high quality of speech output. Benefiting from these
strategies, VALL-E R obtains controllablity over phonemes and demonstrates its
strong robustness by approaching the WER of ground truth. In addition, it
requires fewer autoregressive steps, with over 60% time reduction during
inference. This research has the potential to be applied to meaningful
projects, including the creation of speech for those affected by aphasia. Audio
samples will be available at: https://aka.ms/valler. | cs.CL | VALL-E R: Robust and Efficient Zero-Shot Text-to-Speech Synthesis via Monotonic Alignment |
2024-06-12T03:38:45Z | http://arxiv.org/pdf/2406.07850v1 | http://arxiv.org/abs/2406.07850v1 | Yiwei Li, Fei Mi, Yitong Li, Yasheng Wang, Bin Sun, Shaoxiong Feng, Kan Li | 2024-06-12T03:38:45Z | cs.CL, cs.AI | Stochastic sampling strategies such as top-k and top-p have been widely used
in dialogue generation task. However, as an open-domain chatting system, there
will be two different conversation scenarios, i.e. chit-chat and
knowledge-based question answering. In the former situation, responses
diversity is essential due to the one-to-many nature in dialogue. The latter,
on the other hand, requires less randomness given that stochastic decoding
strategy entails the risk of generating incorrect information. As a result, an
adaptive and flexible decoding strategy is needed to cope with these two
scenarios simultaneously. To this end, we propose the dynamic decoding strategy
(DDS), which can adjust the decoding space w.r.t. different contexts. In DDS,
both sequence-level and token-level adaptive search can be achieved to adjust
the decoding process in a unified framework. Besides, our adaptive algorithm
can not only be used during model inference, but it can also be applied during
the model training stage to further enhance the performance. Comprehensive
experiments indicate that the proposed decoding strategy can consistently
improve the performance of pre-trained dialogue models when coupled with four
well-used stochastic decoding algorithms. | cs.CL | Dynamic Stochastic Decoding Strategy for Open-Domain Dialogue Generation |
2024-06-12T03:17:57Z | http://arxiv.org/pdf/2406.07842v1 | http://arxiv.org/abs/2406.07842v1 | Yerbolat Khassanov, Zhipeng Chen, Tianfeng Chen, Tze Yuang Chong, Wei Li, Jun Zhang, Lu Lu, Yuxuan Wang | 2024-06-12T03:17:57Z | eess.AS, cs.CL | This paper addresses challenges in integrating new languages into a
pre-trained multilingual automatic speech recognition (mASR) system,
particularly in scenarios where training data for existing languages is limited
or unavailable. The proposed method employs a dual-pipeline with low-rank
adaptation (LoRA). It maintains two data flow pipelines-one for existing
languages and another for new languages. The primary pipeline follows the
standard flow through the pre-trained parameters of mASR, while the secondary
pipeline additionally utilizes language-specific parameters represented by LoRA
and a separate output decoder module. Importantly, the proposed approach
minimizes the performance degradation of existing languages and enables a
language-agnostic operation mode, facilitated by a decoder selection strategy.
We validate the effectiveness of the proposed method by extending the
pre-trained Whisper model to 19 new languages from the FLEURS dataset | eess.AS | Dual-Pipeline with Low-Rank Adaptation for New Language Integration in Multilingual ASR |
2024-06-12T03:16:45Z | http://arxiv.org/pdf/2406.07841v1 | http://arxiv.org/abs/2406.07841v1 | Elaheh Baharlouei, Mahsa Shafaei, Yigeng Zhang, Hugo Jair Escalante, Thamar Solorio | 2024-06-12T03:16:45Z | cs.CV, cs.CL | We address the challenge of detecting questionable content in online media,
specifically the subcategory of comic mischief. This type of content combines
elements such as violence, adult content, or sarcasm with humor, making it
difficult to detect. Employing a multimodal approach is vital to capture the
subtle details inherent in comic mischief content. To tackle this problem, we
propose a novel end-to-end multimodal system for the task of comic mischief
detection. As part of this contribution, we release a novel dataset for the
targeted task consisting of three modalities: video, text (video captions and
subtitles), and audio. We also design a HIerarchical Cross-attention model with
CAPtions (HICCAP) to capture the intricate relationships among these
modalities. The results show that the proposed approach makes a significant
improvement over robust baselines and state-of-the-art models for comic
mischief detection and its type classification. This emphasizes the potential
of our system to empower users, to make informed decisions about the online
content they choose to see. In addition, we conduct experiments on the UCF101,
HMDB51, and XD-Violence datasets, comparing our model against other
state-of-the-art approaches showcasing the outstanding performance of our
proposed model in various scenarios. | cs.CV | Labeling Comic Mischief Content in Online Videos with a Multimodal Hierarchical-Cross-Attention Model |
2024-06-12T02:46:17Z | http://arxiv.org/pdf/2406.07823v1 | http://arxiv.org/abs/2406.07823v1 | Trang Le, Daniel Lazar, Suyoun Kim, Shan Jiang, Duc Le, Adithya Sagar, Aleksandr Livshits, Ahmed Aly, Akshat Shrivastava | 2024-06-12T02:46:17Z | cs.CL, cs.SD, eess.AS | Spoken Language Understanding (SLU) is a critical component of voice
assistants; it consists of converting speech to semantic parses for task
execution. Previous works have explored end-to-end models to improve the
quality and robustness of SLU models with Deliberation, however these models
have remained autoregressive, resulting in higher latencies. In this work we
introduce PRoDeliberation, a novel method leveraging a Connectionist Temporal
Classification-based decoding strategy as well as a denoising objective to
train robust non-autoregressive deliberation models. We show that
PRoDeliberation achieves the latency reduction of parallel decoding (2-10x
improvement over autoregressive models) while retaining the ability to correct
Automatic Speech Recognition (ASR) mistranscriptions of autoregressive
deliberation systems. We further show that the design of the denoising training
allows PRoDeliberation to overcome the limitations of small ASR devices, and we
provide analysis on the necessity of each component of the system. | cs.CL | PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding |
2024-06-12T02:43:19Z | http://arxiv.org/pdf/2406.07822v1 | http://arxiv.org/abs/2406.07822v1 | Andrea Burns, Kate Saenko, Bryan A. Plummer | 2024-06-12T02:43:19Z | cs.CV, cs.CL | Mobile app user interfaces (UIs) are rich with action, text, structure, and
image content that can be utilized to learn generic UI representations for
tasks like automating user commands, summarizing content, and evaluating the
accessibility of user interfaces. Prior work has learned strong visual
representations with local or global captioning losses, but fails to retain
both granularities. To combat this, we propose Textual Foresight, a novel
pretraining objective for learning UI screen representations. Textual Foresight
generates global text descriptions of future UI states given a current UI and
local action taken. Our approach requires joint reasoning over elements and
entire screens, resulting in improved UI features: on generation tasks, UI
agents trained with Textual Foresight outperform state-of-the-art by 2% with
28x fewer images. We train with our newly constructed mobile app dataset,
OpenApp, which results in the first public dataset for app UI representation
learning. OpenApp enables new baselines, and we find Textual Foresight improves
average task performance over them by 5.7% while having access to 2x less data. | cs.CV | Tell Me What's Next: Textual Foresight for Generic UI Representations |
2024-06-12T02:23:57Z | http://arxiv.org/pdf/2406.07816v1 | http://arxiv.org/abs/2406.07816v1 | Lin Zhang, Xin Wang, Erica Cooper, Mireia Diez, Federico Landini, Nicholas Evans, Junichi Yamagishi | 2024-06-12T02:23:57Z | eess.AS, cs.CL, cs.SD | This paper defines Spoof Diarization as a novel task in the Partial Spoof
(PS) scenario. It aims to determine what spoofed when, which includes not only
locating spoof regions but also clustering them according to different spoofing
methods. As a pioneering study in spoof diarization, we focus on defining the
task, establishing evaluation metrics, and proposing a benchmark model, namely
the Countermeasure-Condition Clustering (3C) model. Utilizing this model, we
first explore how to effectively train countermeasures to support spoof
diarization using three labeling schemes. We then utilize spoof localization
predictions to enhance the diarization performance. This first study reveals
the high complexity of the task, even in restricted scenarios where only a
single speaker per audio file and an oracle number of spoofing methods are
considered. Our code is available at
https://github.com/nii-yamagishilab/PartialSpoof. | eess.AS | Spoof Diarization: "What Spoofed When" in Partially Spoofed Audio |
2024-06-12T02:23:51Z | http://arxiv.org/pdf/2406.07815v1 | http://arxiv.org/abs/2406.07815v1 | Yizhang Zhu, Shiyin Du, Boyan Li, Yuyu Luo, Nan Tang | 2024-06-12T02:23:51Z | cs.CL, cs.AI | Large Language Models (LLMs) have demonstrated impressive capabilities across
a range of scientific tasks including mathematics, physics, and chemistry.
Despite their successes, the effectiveness of LLMs in handling complex
statistical tasks remains systematically under-explored. To bridge this gap, we
introduce StatQA, a new benchmark designed for statistical analysis tasks.
StatQA comprises 11,623 examples tailored to evaluate LLMs' proficiency in
specialized statistical tasks and their applicability assessment capabilities,
particularly for hypothesis testing methods. We systematically experiment with
representative LLMs using various prompting strategies and show that even
state-of-the-art models such as GPT-4o achieve a best performance of only
64.83%, indicating significant room for improvement. Notably, while open-source
LLMs (e.g. LLaMA-3) show limited capability, those fine-tuned ones exhibit
marked improvements, outperforming all in-context learning-based methods (e.g.
GPT-4o). Moreover, our comparative human experiments highlight a striking
contrast in error types between LLMs and humans: LLMs primarily make
applicability errors, whereas humans mostly make statistical task confusion
errors. This divergence highlights distinct areas of proficiency and
deficiency, suggesting that combining LLM and human expertise could lead to
complementary strengths, inviting further investigation into their
collaborative potential. | cs.CL | Are Large Language Models Good Statisticians? |
2024-06-12T02:20:46Z | http://arxiv.org/pdf/2406.07814v1 | http://arxiv.org/abs/2406.07814v1 | Saffron Huang, Divya Siddarth, Liane Lovitt, Thomas I. Liao, Esin Durmus, Alex Tamkin, Deep Ganguli | 2024-06-12T02:20:46Z | cs.AI, cs.CL, cs.HC, I.2.7; K.4.2 | There is growing consensus that language model (LM) developers should not be
the sole deciders of LM behavior, creating a need for methods that enable the
broader public to collectively shape the behavior of LM systems that affect
them. To address this need, we present Collective Constitutional AI (CCAI): a
multi-stage process for sourcing and integrating public input into LMs-from
identifying a target population to sourcing principles to training and
evaluating a model. We demonstrate the real-world practicality of this approach
by creating what is, to our knowledge, the first LM fine-tuned with
collectively sourced public input and evaluating this model against a baseline
model trained with established principles from a LM developer. Our quantitative
evaluations demonstrate several benefits of our approach: the CCAI-trained
model shows lower bias across nine social dimensions compared to the baseline
model, while maintaining equivalent performance on language, math, and
helpful-harmless evaluations. Qualitative comparisons of the models suggest
that the models differ on the basis of their respective constitutions, e.g.,
when prompted with contentious topics, the CCAI-trained model tends to generate
responses that reframe the matter positively instead of a refusal. These
results demonstrate a promising, tractable pathway toward publicly informed
development of language models. | cs.AI | Collective Constitutional AI: Aligning a Language Model with Public Input |
2024-06-12T02:08:45Z | http://arxiv.org/pdf/2406.07812v1 | http://arxiv.org/abs/2406.07812v1 | Yiran Wang, Masao Utiyama | 2024-06-12T02:08:45Z | cs.CL, cs.AI, cs.LG | Recently, binary representation has been proposed as a novel representation
that lies between continuous and discrete representations. It exhibits
considerable information-preserving capability when being used to replace
continuous input vectors. In this paper, we investigate the feasibility of
further introducing it to the output side, aiming to allow models to output
binary labels instead. To preserve the structural information on the output
side along with label information, we extend the previous contrastive hashing
method as structured contrastive hashing. More specifically, we upgrade CKY
from label-level to bit-level, define a new similarity function with span
marginal probabilities, and introduce a novel contrastive loss function with a
carefully designed instance selection strategy. Our model achieves competitive
performance on various structured prediction tasks, and demonstrates that
binary representation can be considered a novel representation that further
bridges the gap between the continuous nature of deep learning and the discrete
intrinsic property of natural languages. | cs.CL | To be Continuous, or to be Discrete, Those are Bits of Questions |
2024-06-12T01:49:54Z | http://arxiv.org/pdf/2406.10278v1 | http://arxiv.org/abs/2406.10278v1 | Renlong Jie, Xiaojun Meng, Lifeng Shang, Xin Jiang, Qun Liu | 2024-06-12T01:49:54Z | cs.CL, cs.AI | Large language models (LLMs) have attracted great attention given their
strong performance on a wide range of NLP tasks. In practice, users often
expect generated texts to fall within a specific length range, making length
controlled generation an important topic, especially for GPT-style models.
Existing length control methods mostly focus on a simple control type of "equal
to" a target length. Different from them, we propose a prompt-based method to
achieve length controlled generation under different control types with high
accuracy. In particular, we adopt reinforcement learning (RL) and sample
filtering with the reward signal given by rule-based reward models, which
enhances the length control ability of models by rewarding outputs that follow
certain control instructions. In addition, we introduce a standard prompt
extractor to parse arbitrary users' input into standard control instructions.
Experiments show that our method significantly improves the accuracy of
prompt-based length control on popular summarization datasets like CNNDM and
NYT under multiple control types. Moreover, both the standard prompt extractor
and RL-tuned model show strong generalization to unseen control prompt
templates. | cs.CL | Prompt-Based Length Controlled Generation with Multiple Control Types |
2024-06-12T01:35:46Z | http://arxiv.org/pdf/2406.07801v1 | http://arxiv.org/abs/2406.07801v1 | Runyan Yang, Huibao Yang, Xiqing Zhang, Tiantian Ye, Ying Liu, Yingying Gao, Shilei Zhang, Chao Deng, Junlan Feng | 2024-06-12T01:35:46Z | cs.CL, cs.SD, eess.AS | Recently, there have been attempts to integrate various speech processing
tasks into a unified model. However, few previous works directly demonstrated
that joint optimization of diverse tasks in multitask speech models has
positive influence on the performance of individual tasks. In this paper we
present a multitask speech model -- PolySpeech, which supports speech
recognition, speech synthesis, and two speech classification tasks. PolySpeech
takes multi-modal language model as its core structure and uses semantic
representations as speech inputs. We introduce semantic speech embedding
tokenization and speech reconstruction methods to PolySpeech, enabling
efficient generation of high-quality speech for any given speaker. PolySpeech
shows competitiveness across various tasks compared to single-task models. In
our experiments, multitask optimization achieves performance comparable to
single-task optimization and is especially beneficial for specific tasks. | cs.CL | PolySpeech: Exploring Unified Multitask Speech Models for Competitiveness with Single-task Models |
2024-06-12T01:25:00Z | http://arxiv.org/pdf/2406.09444v2 | http://arxiv.org/abs/2406.09444v2 | Yingying Gao, Shilei Zhang, Chao Deng, Junlan Feng | 2024-06-21T08:48:41Z | eess.AS, cs.CL, cs.SD | Pre-trained speech language models such as HuBERT and WavLM leverage
unlabeled speech data for self-supervised learning and offer powerful
representations for numerous downstream tasks. Despite the success of these
models, their high requirements for memory and computing resource hinder their
application on resource restricted devices. Therefore, this paper introduces
GenDistiller, a novel knowledge distillation framework which generates the
hidden representations of the pre-trained teacher model directly by a much
smaller student network. The proposed method takes the previous hidden layer as
history and implements a layer-by-layer prediction of the teacher model
autoregressively. Experiments on SUPERB reveal the advantage of GenDistiller
over the baseline distilling method without an autoregressive framework, with
33% fewer parameters, similar time consumption and better performance on most
of the SUPERB tasks. Ultimately, the proposed GenDistiller reduces the size of
WavLM by 82%. | eess.AS | GenDistiller: Distilling Pre-trained Language Models based on an Autoregressive Generative Model |
2024-06-12T01:18:04Z | http://arxiv.org/pdf/2406.07794v2 | http://arxiv.org/abs/2406.07794v2 | Amogh Mannekote, Jinseok Nam, Ziming Li, Jian Gao, Kristy Elizabeth Boyer, Bonnie J. Dorr | 2024-06-16T21:20:34Z | cs.CL, cs.AI | Indirect User Requests (IURs), such as "It's cold in here" instead of "Could
you please increase the temperature?" are common in human-human task-oriented
dialogue and require world knowledge and pragmatic reasoning from the listener.
While large language models (LLMs) can handle these requests effectively,
smaller models deployed on virtual assistants often struggle due to resource
constraints. Moreover, existing task-oriented dialogue benchmarks lack
sufficient examples of complex discourse phenomena such as indirectness. To
address this, we propose a set of linguistic criteria along with an LLM-based
pipeline for generating realistic IURs to test natural language understanding
(NLU) and dialogue state tracking (DST) models before deployment in a new
domain. We also release IndirectRequests, a dataset of IURs based on the Schema
Guided Dialog (SGD) corpus, as a comparative testbed for evaluating the
performance of smaller models in handling indirect requests. | cs.CL | Making Task-Oriented Dialogue Datasets More Natural by Synthetically Generating Indirect User Requests |
2024-06-12T01:12:28Z | http://arxiv.org/pdf/2406.07791v1 | http://arxiv.org/abs/2406.07791v1 | Lin Shi, Weicheng Ma, Soroush Vosoughi | 2024-06-12T01:12:28Z | cs.CL, cs.AI | LLM-as-a-Judge offers a promising alternative to human judges across various
tasks, yet inherent biases, particularly position bias - a systematic
preference for answers based on their position in the prompt - compromise its
effectiveness. Our study investigates this issue by developing a framework to
systematically study and quantify position bias using metrics such as
repetitional consistency, positional consistency, and positional fairness. We
conduct experiments with 9 judge models across 22 tasks from the MTBench and
DevBench benchmarks and nearly 40 answer-generating models, generating
approximately 80,000 evaluation instances. This comprehensive assessment
reveals significant variations in bias across judges and tasks. Although GPT-4
often excels in positional consistency and fairness, some more cost-effective
models perform comparably or even better in specific tasks, highlighting
essential trade-offs between consistency, fairness, and cost. Our results also
demonstrate high consistency of judgment across repetitions, confirming that
position bias is not due to random variations. This research significantly
contributes to the field by introducing new concepts for understanding position
bias and providing a multi-dimensional framework for evaluation. These insights
guide the selection of optimal judge models, enhance benchmark design, and lay
the foundation for future research into effective debiasing strategies,
ultimately enhancing the reliability of LLM evaluators. | cs.CL | Judging the Judges: A Systematic Investigation of Position Bias in Pairwise Comparative Assessments by LLMs |
2024-06-12T00:19:40Z | http://arxiv.org/pdf/2406.07780v1 | http://arxiv.org/abs/2406.07780v1 | Ahmad Rashid, Ruotian Wu, Julia Grosse, Agustinus Kristiadi, Pascal Poupart | 2024-06-12T00:19:40Z | cs.LG, cs.CL | Large language models (LLMs) can significantly be improved by aligning to
human preferences -- the so-called reinforcement learning from human feedback
(RLHF). However, the cost of fine-tuning an LLM is prohibitive for many users.
Due to their ability to bypass LLM finetuning, tokenwise reward-guided text
generation (RGTG) methods have recently been proposed. They use a reward model
trained on full sequences to score partial sequences during a tokenwise
decoding, in a bid to steer the generation towards sequences with high rewards.
However, these methods have so far been only heuristically motivated and poorly
analyzed. In this work, we show that reward models trained on full sequences
are not compatible with scoring partial sequences. To alleviate this issue, we
propose to explicitly train a Bradley-Terry reward model on partial sequences,
and autoregressively sample from the implied tokenwise policy during decoding
time. We study the property of this reward model and the implied policy. In
particular, we show that this policy is proportional to the ratio of two
distinct RLHF policies. We show that our simple approach outperforms previous
RGTG methods and achieves similar performance as strong offline baselines but
without large-scale LLM finetuning. | cs.LG | A Critical Look At Tokenwise Reward-Guided Text Generation |
2024-06-12T00:01:32Z | http://arxiv.org/pdf/2406.07778v1 | http://arxiv.org/abs/2406.07778v1 | Jayaram Raghuram, George Kesidis, David J. Miller | 2024-06-12T00:01:32Z | cs.CR, cs.AI, cs.CL, cs.LG | A Trojan in a language model can be inserted when the model is refined for a
particular application such as determining the sentiment of product reviews. In
this paper, we clarify and empirically explore variations of the data-poisoning
threat model. We then empirically assess two simple defenses each for a
different defense scenario. Finally, we provide a brief survey of related
attacks and defenses. | cs.CR | On Trojans in Refined Language Models |
2024-06-12T00:00:39Z | http://arxiv.org/pdf/2406.10276v1 | http://arxiv.org/abs/2406.10276v1 | Peidong Wang, Jian Xue, Jinyu Li, Junkun Chen, Aswin Shanmugam Subramanian | 2024-06-12T00:00:39Z | cs.CL, cs.SD, eess.AS | Language-agnostic many-to-one end-to-end speech translation models can
convert audio signals from different source languages into text in a target
language. These models do not need source language identification, which
improves user experience. In some cases, the input language can be given or
estimated. Our goal is to use this additional language information while
preserving the quality of the other languages. We accomplish this by
introducing a simple and effective linear input network. The linear input
network is initialized as an identity matrix, which ensures that the model can
perform as well as, or better than, the original model. Experimental results
show that the proposed method can successfully enhance the specified language,
while keeping the language-agnostic ability of the many-to-one ST models. | cs.CL | Soft Language Identification for Language-Agnostic Many-to-One End-to-End Speech Translation |
2024-06-11T22:48:18Z | http://arxiv.org/pdf/2406.07759v1 | http://arxiv.org/abs/2406.07759v1 | Dasun Athukoralage, Thushari Atapattu, Menasha Thilakaratne, Katrina Falkner | 2024-06-11T22:48:18Z | cs.CL | This paper presents our approaches for the SMM4H24 Shared Task 5 on the
binary classification of English tweets reporting children's medical disorders.
Our first approach involves fine-tuning a single RoBERTa-large model, while the
second approach entails ensembling the results of three fine-tuned
BERTweet-large models. We demonstrate that although both approaches exhibit
identical performance on validation data, the BERTweet-large ensemble excels on
test data. Our best-performing system achieves an F1-score of 0.938 on test
data, outperforming the benchmark classifier by 1.18%. | cs.CL | LT4SG@SMM4H24: Tweets Classification for Digital Epidemiology of Childhood Health Outcomes Using Pre-Trained Language Models |
2024-06-11T22:26:20Z | http://arxiv.org/pdf/2406.07753v1 | http://arxiv.org/abs/2406.07753v1 | Shahin Amiriparian, Lukas Christ, Alexander Kathan, Maurice Gerczuk, Niklas Müller, Steffen Klug, Lukas Stappen, Andreas König, Erik Cambria, Björn Schuller, Simone Eulitz | 2024-06-11T22:26:20Z | cs.AI, cs.CL, 68T10, I.2 | The Multimodal Sentiment Analysis Challenge (MuSe) 2024 addresses two
contemporary multimodal affect and sentiment analysis problems: In the Social
Perception Sub-Challenge (MuSe-Perception), participants will predict 16
different social attributes of individuals such as assertiveness, dominance,
likability, and sincerity based on the provided audio-visual data. The
Cross-Cultural Humor Detection Sub-Challenge (MuSe-Humor) dataset expands upon
the Passau Spontaneous Football Coach Humor (Passau-SFCH) dataset, focusing on
the detection of spontaneous humor in a cross-lingual and cross-cultural
setting. The main objective of MuSe 2024 is to unite a broad audience from
various research domains, including multimodal sentiment analysis, audio-visual
affective computing, continuous signal processing, and natural language
processing. By fostering collaboration and exchange among experts in these
fields, the MuSe 2024 endeavors to advance the understanding and application of
sentiment analysis and affective computing across multiple modalities. This
baseline paper provides details on each sub-challenge and its corresponding
dataset, extracted features from each data modality, and discusses challenge
baselines. For our baseline system, we make use of a range of Transformers and
expert-designed features and train Gated Recurrent Unit (GRU)-Recurrent Neural
Network (RNN) models on them, resulting in a competitive baseline system. On
the unseen test datasets of the respective sub-challenges, it achieves a mean
Pearson's Correlation Coefficient ($\rho$) of 0.3573 for MuSe-Perception and an
Area Under the Curve (AUC) value of 0.8682 for MuSe-Humor. | cs.AI | The MuSe 2024 Multimodal Sentiment Analysis Challenge: Social Perception and Humor Recognition |
2024-06-11T21:53:46Z | http://arxiv.org/pdf/2406.07739v1 | http://arxiv.org/abs/2406.07739v1 | Jason Wu, Eldon Schoop, Alan Leung, Titus Barik, Jeffrey P. Bigham, Jeffrey Nichols | 2024-06-11T21:53:46Z | cs.CL, cs.HC, cs.SE | Large language models (LLMs) struggle to consistently generate UI code that
compiles and produces visually relevant designs. Existing approaches to improve
generation rely on expensive human feedback or distilling a proprietary model.
In this paper, we explore the use of automated feedback (compilers and
multi-modal models) to guide LLMs to generate high-quality UI code. Our method
starts with an existing LLM and iteratively produces improved models by
self-generating a large synthetic dataset using an original model, applying
automated tools to aggressively filter, score, and de-duplicate the data into a
refined higher quality dataset. The original LLM is improved by finetuning on
this refined dataset. We applied our approach to several open-source LLMs and
compared the resulting performance to baseline models with both automated
metrics and human preferences. Our evaluation shows the resulting models
outperform all other downloadable baselines and approach the performance of
larger proprietary models. | cs.CL | UICoder: Finetuning Large Language Models to Generate User Interface Code through Automated Feedback |
2024-06-11T21:46:03Z | http://arxiv.org/pdf/2406.07736v1 | http://arxiv.org/abs/2406.07736v1 | Dojun Park, Jiwoo Lee, Seohyun Park, Hyeyun Jeong, Youngeun Koo, Soonha Hwang, Seonwoo Park, Sungeun Lee | 2024-06-11T21:46:03Z | cs.CL | As the capabilities of LLMs expand, it becomes increasingly important to
evaluate them beyond basic knowledge assessment, focusing on higher-level
language understanding. This study introduces MultiPragEval, a robust test
suite designed for the multilingual pragmatic evaluation of LLMs across
English, German, Korean, and Chinese. Comprising 1200 question units
categorized according to Grice's Cooperative Principle and its four
conversational maxims, MultiPragEval enables an in-depth assessment of LLMs'
contextual awareness and their ability to infer implied meanings. Our findings
demonstrate that Claude3-Opus significantly outperforms other models in all
tested languages, establishing a state-of-the-art in the field. Among
open-source models, Solar-10.7B and Qwen1.5-14B emerge as strong competitors.
This study not only leads the way in the multilingual evaluation of LLMs in
pragmatic inference but also provides valuable insights into the nuanced
capabilities necessary for advanced language comprehension in AI systems. | cs.CL | MultiPragEval: Multilingual Pragmatic Evaluation of Large Language Models |
2024-06-11T21:44:49Z | http://arxiv.org/pdf/2406.07735v1 | http://arxiv.org/abs/2406.07735v1 | Haw-Shiuan Chang, Nanyun Peng, Mohit Bansal, Anil Ramakrishna, Tagyoung Chung | 2024-06-11T21:44:49Z | cs.CL, cs.LG | Decoding methods for large language models (LLMs) usually struggle with the
tradeoff between ensuring factuality and maintaining diversity. For example, a
higher p threshold in the nucleus (top-p) sampling increases the diversity but
decreases the factuality, and vice versa. In this paper, we propose REAL
(Residual Entropy from Asymptotic Line) sampling, a decoding method that
achieves improved factuality and diversity over nucleus sampling by predicting
an adaptive threshold of $p$. Specifically, REAL sampling predicts the
step-wise likelihood of an LLM to hallucinate, and lowers the p threshold when
an LLM is likely to hallucinate. Otherwise, REAL sampling increases the p
threshold to boost the diversity. To predict the step-wise hallucination
likelihood without supervision, we construct a Token-level Hallucination
Forecasting (THF) model to predict the asymptotic entropy (i.e., inherent
uncertainty) of the next token by extrapolating the next-token entropies from a
series of LLMs with different sizes. If a LLM's entropy is higher than the
asymptotic entropy (i.e., the LLM is more uncertain than it should be), the THF
model predicts a high hallucination hazard, which leads to a lower p threshold
in REAL sampling. In the FactualityPrompts benchmark, we demonstrate that REAL
sampling based on a 70M THF model can substantially improve the factuality and
diversity of 7B LLMs simultaneously, judged by both retrieval-based metrics and
human evaluation. After combined with contrastive decoding, REAL sampling
outperforms 9 sampling methods, and generates texts that are more factual than
the greedy sampling and more diverse than the nucleus sampling with $p=0.5$.
Furthermore, the predicted asymptotic entropy is also a useful unsupervised
signal for hallucination detection tasks. | cs.CL | REAL Sampling: Boosting Factuality and Diversity of Open-Ended Generation via Asymptotic Entropy |
2024-06-11T21:30:15Z | http://arxiv.org/pdf/2406.10275v1 | http://arxiv.org/abs/2406.10275v1 | Shahin Amiriparian, Filip Packań, Maurice Gerczuk, Björn W. Schuller | 2024-06-11T21:30:15Z | cs.CL, 68T10, I.2 | Foundation models have shown great promise in speech emotion recognition
(SER) by leveraging their pre-trained representations to capture emotion
patterns in speech signals. To further enhance SER performance across various
languages and domains, we propose a novel twofold approach. First, we gather
EmoSet++, a comprehensive multi-lingual, multi-cultural speech emotion corpus
with 37 datasets, 150,907 samples, and a total duration of 119.5 hours. Second,
we introduce ExHuBERT, an enhanced version of HuBERT achieved by backbone
extension and fine-tuning on EmoSet++. We duplicate each encoder layer and its
weights, then freeze the first duplicate, integrating an extra zero-initialized
linear layer and skip connections to preserve functionality and ensure its
adaptability for subsequent fine-tuning. Our evaluation on unseen datasets
shows the efficacy of ExHuBERT, setting a new benchmark for various SER tasks.
Model and details on EmoSet++: https://huggingface.co/amiriparian/ExHuBERT. | cs.CL | ExHuBERT: Enhancing HuBERT Through Block Extension and Fine-Tuning on 37 Emotion Datasets |
2024-06-11T20:27:45Z | http://arxiv.org/pdf/2406.08520v1 | http://arxiv.org/abs/2406.08520v1 | Mohammad Tami, Huthaifa I. Ashqar, Mohammed Elhenawy | 2024-06-11T20:27:45Z | cs.CL, cs.CY | Question generation for education assessments is a growing field within
artificial intelligence applied to education. These question-generation tools
have significant importance in the educational technology domain, such as
intelligent tutoring systems and dialogue-based platforms. The automatic
generation of assessment questions, which entail clear-cut answers, usually
relies on syntactical and semantic indications within declarative sentences,
which are then transformed into questions. Recent research has explored the
generation of assessment educational questions in Arabic. The reported
performance has been adversely affected by inherent errors, including sentence
parsing inaccuracies, name entity recognition issues, and errors stemming from
rule-based question transformation. Furthermore, the complexity of lengthy
Arabic sentences has contributed to these challenges. This research presents an
innovative Arabic question-generation system built upon a three-stage process:
keywords and key phrases extraction, question generation, and subsequent
ranking. The aim is to tackle the difficulties associated with automatically
generating assessment questions in the Arabic language. The proposed approach
and results show a precision of 83.50%, a recall of 78.68%, and an Fl score of
80.95%, indicating the framework high efficiency. Human evaluation further
confirmed the model efficiency, receiving an average rating of 84%. | cs.CL | Automated Question Generation for Science Tests in Arabic Language Using NLP Techniques |
2024-06-11T20:23:31Z | http://arxiv.org/pdf/2406.08519v1 | http://arxiv.org/abs/2406.08519v1 | Mohammad Sammoudi, Ahmad Habaybeh, Huthaifa I. Ashqar, Mohammed Elhenawy | 2024-06-11T20:23:31Z | cs.CL, cs.CY | This paper describes the creation, optimization, and assessment of a
question-answering (QA) model for a personalized learning assistant that uses
BERT transformers customized for the Arabic language. The model was
particularly finetuned on science textbooks in Palestinian curriculum. Our
approach uses BERT's brilliant capabilities to automatically produce correct
answers to questions in the field of science education. The model's ability to
understand and extract pertinent information is improved by finetuning it using
11th and 12th grade biology book in Palestinian curriculum. This increases the
model's efficacy in producing enlightening responses. Exact match (EM) and F1
score metrics are used to assess the model's performance; the results show an
EM score of 20% and an F1 score of 51%. These findings show that the model can
comprehend and react to questions in the context of Palestinian science book.
The results demonstrate the potential of BERT-based QA models to support
learning and understanding Arabic students questions. | cs.CL | Question-Answering (QA) Model for a Personalized Learning Assistant for Arabic Language |
2024-06-11T20:21:36Z | http://arxiv.org/pdf/2406.07696v1 | http://arxiv.org/abs/2406.07696v1 | Luis Lugo, Valentin Vielzeuf | 2024-06-11T20:21:36Z | cs.CL | Sustainable artificial intelligence focuses on data, hardware, and algorithms
to make machine learning models more environmentally responsible. In
particular, machine learning models for speech representations are
computationally expensive, generating environmental concerns because of their
high energy consumption. Thus, we propose a sustainable self-supervised model
to learn speech representation, combining optimizations in neural layers and
training to reduce computing costs. The proposed model improves over a
resource-efficient baseline, reducing both memory usage and computing cost
estimations. It pretrains using a single GPU in less than a day. On top of
that, it improves the error rate performance of the baseline in downstream task
evaluations. When comparing it to large speech representation approaches, there
is an order of magnitude reduction in memory usage, while computing cost
reductions represent almost three orders of magnitude improvement. | cs.CL | Sustainable self-supervised learning for speech representations |
2024-06-11T20:15:57Z | http://arxiv.org/pdf/2406.10274v1 | http://arxiv.org/abs/2406.10274v1 | Patrick D. F. Ion, Stephen M. Watt | 2024-06-11T20:15:57Z | cs.IR, cs.CL, cs.DL | In this article we report on an initial exploration to assess the viability
of using the general large language models (LLMs), recently made public, to
classify mathematical documents. Automated classification would be useful from
the applied perspective of improving the navigation of the literature and the
more open-ended goal of identifying relations among mathematical results. The
Mathematical Subject Classification MSC 2020, from MathSciNet and zbMATH, is
widely used and there is a significant corpus of ground truth material in the
open literature. We have evaluated the classification of preprint articles from
arXiv.org according to MSC 2020. The experiment used only the title and
abstract alone -- not the entire paper. Since this was early in the use of
chatbots and the development of their APIs, we report here on what was carried
out by hand. Of course, the automation of the process will have to follow if it
is to be generally useful. We found that in about 60% of our sample the LLM
produced a primary classification matching that already reported on arXiv. In
about half of those instances, there were additional primary classifications
that were not detected. In about 40% of our sample, the LLM suggested a
different classification than what was provided. A detailed examination of
these cases, however, showed that the LLM-suggested classifications were in
most cases better than those provided. | cs.IR | Using General Large Language Models to Classify Mathematical Documents |
2024-06-11T20:14:22Z | http://arxiv.org/pdf/2406.07693v2 | http://arxiv.org/abs/2406.07693v2 | Nirmalya Thakur, Vanessa Su, Mingchen Shao, Kesha A. Patel, Hongseok Jeong, Victoria Knieling, Andrew Bian | 2024-06-16T21:10:55Z | cs.CY, cs.AI, cs.CL, cs.LG, cs.SI, I.2.7; I.2.8; I.5.4; K.4.2; H.2.8; I.2.6 | The work of this paper presents a dataset that contains the data of 4011
videos about the ongoing outbreak of measles published on 264 websites on the
internet between January 1, 2024, and May 31, 2024. The dataset is available at
https://dx.doi.org/10.21227/40s8-xf63. These websites primarily include YouTube
and TikTok, which account for 48.6% and 15.2% of the videos, respectively. The
remainder of the websites include Instagram and Facebook as well as the
websites of various global and local news organizations. For each of these
videos, the URL of the video, title of the post, description of the post, and
the date of publication of the video are presented as separate attributes in
the dataset. After developing this dataset, sentiment analysis (using VADER),
subjectivity analysis (using TextBlob), and fine-grain sentiment analysis
(using DistilRoBERTa-base) of the video titles and video descriptions were
performed. This included classifying each video title and video description
into (i) one of the sentiment classes i.e. positive, negative, or neutral, (ii)
one of the subjectivity classes i.e. highly opinionated, neutral opinionated,
or least opinionated, and (iii) one of the fine-grain sentiment classes i.e.
fear, surprise, joy, sadness, anger, disgust, or neutral. These results are
presented as separate attributes in the dataset for the training and testing of
machine learning algorithms for performing sentiment analysis or subjectivity
analysis in this field as well as for other applications. Finally, this paper
also presents a list of open research questions that may be investigated using
this dataset. | cs.CY | A Labelled Dataset for Sentiment Analysis of Videos on YouTube, TikTok, and Other Sources about the 2024 Outbreak of Measles |
2024-06-11T20:14:09Z | http://arxiv.org/pdf/2406.07692v1 | http://arxiv.org/abs/2406.07692v1 | Sari Masri, Yaqeen Raddad, Fidaa Khandaqji, Huthaifa I. Ashqar, Mohammed Elhenawy | 2024-06-11T20:14:09Z | cs.CL, cs.ET | Recently, with the rapid development in the fields of technology and the
increasing amount of text t available on the internet, it has become urgent to
develop effective tools for processing and understanding texts in a way that
summaries the content without losing the fundamental essence of the
information. Given this challenge, we have developed an advanced text
summarization system targeting Arabic textbooks. Relying on modern natu-ral
language processing models such as MT5, AraBART, AraT5, and mBART50, this
system evaluates and extracts the most important sentences found in biology
textbooks for the 11th and 12th grades in the Palestinian curriculum, which
enables students and teachers to obtain accurate and useful summaries that help
them easily understand the content. We utilized the Rouge metric to evaluate
the performance of the trained models. Moreover, experts in education Edu
textbook authoring assess the output of the trained models. This approach aims
to identify the best solutions and clarify areas needing improvement. This
research provides a solution for summarizing Arabic text. It enriches the field
by offering results that can open new horizons for research and development in
the technologies for understanding and generating the Arabic language.
Additionally, it contributes to the field with Arabic texts through creating
and compiling schoolbook texts and building a dataset. | cs.CL | Transformer Models in Education: Summarizing Science Textbooks with AraBART, MT5, AraT5, and mBART |
2024-06-11T20:07:39Z | http://arxiv.org/pdf/2406.09438v1 | http://arxiv.org/abs/2406.09438v1 | Shadi Jaradat, Taqwa I. Alhadidi, Huthaifa I. Ashqar, Ahmed Hossain, Mohammed Elhenawy | 2024-06-11T20:07:39Z | cs.CL, cs.IR | This study explores traffic crash narratives in an attempt to inform and
enhance effective traffic safety policies using text-mining analytics. Text
mining techniques are employed to unravel key themes and trends within the
narratives, aiming to provide a deeper understanding of the factors
contributing to traffic crashes. This study collected crash data from five
major freeways in Jordan that cover narratives of 7,587 records from 2018-2022.
An unsupervised learning method was adopted to learn the pattern from crash
data. Various text mining techniques, such as topic modeling, keyword
extraction, and Word Co-Occurrence Network, were also used to reveal the
co-occurrence of crash patterns. Results show that text mining analytics is a
promising method and underscore the multifactorial nature of traffic crashes,
including intertwining human decisions and vehicular conditions. The recurrent
themes across all analyses highlight the need for a balanced approach to road
safety, merging both proactive and reactive measures. Emphasis on driver
education and awareness around animal-related incidents is paramount. | cs.CL | Exploring Traffic Crash Narratives in Jordan Using Text Mining Analytics |
2024-06-11T20:05:15Z | http://arxiv.org/pdf/2406.07685v1 | http://arxiv.org/abs/2406.07685v1 | Leonardo Cotta, Chris J. Maddison | 2024-06-11T20:05:15Z | cs.CL, cs.AI | Frontier Large Language Models (LLMs) are increasingly being deployed for
high-stakes decision-making. On the other hand, these models are still
consistently making predictions that contradict users' or society's
expectations, e.g., hallucinating, or discriminating. Thus, it is important
that we develop test-time strategies to improve their trustworthiness. Inspired
by prior work, we leverage causality as a tool to formally encode two aspects
of trustworthiness in LLMs: fairness and robustness. Under this perspective,
existing test-time solutions explicitly instructing the model to be fair or
robust implicitly depend on the LLM's causal reasoning capabilities. In this
work, we explore the opposite approach. Instead of explicitly asking the LLM
for trustworthiness, we design prompts to encode the underlying causal
inference algorithm that will, by construction, result in more trustworthy
predictions. Concretely, we propose out-of-context prompting as a test-time
solution to encourage fairness and robustness in LLMs. Out-of-context prompting
leverages the user's prior knowledge of the task's causal model to apply
(random) counterfactual transformations and improve the model's
trustworthiness. Empirically, we show that out-of-context prompting
consistently improves the fairness and robustness of frontier LLMs across five
different benchmark datasets without requiring additional data, finetuning or
pre-training. | cs.CL | Out-Of-Context Prompting Boosts Fairness and Robustness in Large Language Model Predictions |
2024-06-11T19:20:27Z | http://arxiv.org/pdf/2406.10273v1 | http://arxiv.org/abs/2406.10273v1 | Matteo Esposito, Francesco Palagiano, Valentina Lenarduzzi | 2024-06-11T19:20:27Z | cs.CL, cs.AI, cs.CR, cs.HC | Context. Risk analysis assesses potential risks in specific scenarios. Risk
analysis principles are context-less; the same methodology can be applied to a
risk connected to health and information technology security. Risk analysis
requires a vast knowledge of national and international regulations and
standards and is time and effort-intensive. A large language model can quickly
summarize information in less time than a human and can be fine-tuned to
specific tasks. Aim. Our empirical study aims to investigate the effectiveness
of Retrieval-Augmented Generation and fine-tuned LLM in Risk analysis. To our
knowledge, no prior study has explored its capabilities in risk analysis.
Method. We manually curated \totalscenarios unique scenarios leading to
\totalsamples representative samples from over 50 mission-critical analyses
archived by the industrial context team in the last five years. We compared the
base GPT-3.5 and GPT-4 models versus their Retrieval-Augmented Generation and
fine-tuned counterparts. We employ two human experts as competitors of the
models and three other three human experts to review the models and the former
human expert's analysis. The reviewers analyzed 5,000 scenario analyses.
Results and Conclusions. HEs demonstrated higher accuracy, but LLMs are quicker
and more actionable. Moreover, our findings show that RAG-assisted LLMs have
the lowest hallucination rates, effectively uncovering hidden risks and
complementing human expertise. Thus, the choice of model depends on specific
needs, with FTMs for accuracy, RAG for hidden risks discovery, and base models
for comprehensiveness and actionability. Therefore, experts can leverage LLMs
for an effective complementing companion in risk analysis within a condensed
timeframe. They can also save costs by averting unnecessary expenses associated
with implementing unwarranted countermeasures. | cs.CL | Beyond Words: On Large Language Models Actionability in Mission-Critical Risk Analysis |
2024-06-11T19:04:29Z | http://arxiv.org/pdf/2406.10272v2 | http://arxiv.org/abs/2406.10272v2 | Saturnino Luz, Sofia De La Fuente Garcia, Fasih Haider, Davida Fromm, Brian MacWhinney, Alyssa Lanzi, Ya-Ning Chang, Chia-Ju Chou, Yi-Chien Liu | 2024-06-18T10:41:48Z | cs.CL, cs.LG, cs.SD, eess.AS, J.3; I.5.4 | We present a novel benchmark dataset and prediction tasks for investigating
approaches to assess cognitive function through analysis of connected speech.
The dataset consists of speech samples and clinical information for speakers of
Mandarin Chinese and English with different levels of cognitive impairment as
well as individuals with normal cognition. These data have been carefully
matched by age and sex by propensity score analysis to ensure balance and
representativity in model training. The prediction tasks encompass mild
cognitive impairment diagnosis and cognitive test score prediction. This
framework was designed to encourage the development of approaches to
speech-based cognitive assessment which generalise across languages. We
illustrate it by presenting baseline prediction models that employ
language-agnostic and comparable features for diagnosis and cognitive test
score prediction. The models achieved unweighted average recall was 59.2% in
diagnosis, and root mean squared error of 2.89 in score prediction. | cs.CL | Connected Speech-Based Cognitive Assessment in Chinese and English |
2024-06-11T18:55:04Z | http://arxiv.org/pdf/2406.07657v1 | http://arxiv.org/abs/2406.07657v1 | Lichang Chen, Jiuhai Chen, Chenxi Liu, John Kirchenbauer, Davit Soselia, Chen Zhu, Tom Goldstein, Tianyi Zhou, Heng Huang | 2024-06-11T18:55:04Z | cs.LG, cs.CL | Reinforcement learning with human feedback~(RLHF) is critical for aligning
Large Language Models (LLMs) with human preference. Compared to the widely
studied offline version of RLHF, \emph{e.g.} direct preference optimization
(DPO), recent works have shown that the online variants achieve even better
alignment. However, online alignment requires on-the-fly generation of new
training data, which is costly, hard to parallelize, and suffers from varying
quality and utility. In this paper, we propose a more efficient data
exploration strategy for online preference tuning (OPTune), which does not rely
on human-curated or pre-collected teacher responses but dynamically samples
informative responses for on-policy preference alignment. During data
generation, OPTune only selects prompts whose (re)generated responses can
potentially provide more informative and higher-quality training signals than
the existing responses. In the training objective, OPTune reweights each
generated response (pair) by its utility in improving the alignment so that
learning can be focused on the most helpful samples. Throughout our
evaluations, OPTune'd LLMs maintain the instruction-following benefits provided
by standard preference tuning whilst enjoying 1.27-1.56x faster training speed
due to the efficient data exploration strategy. | cs.LG | OPTune: Efficient Online Preference Tuning |
2024-06-11T17:59:48Z | http://arxiv.org/pdf/2406.07546v1 | http://arxiv.org/abs/2406.07546v1 | Xingyu Fu, Muyu He, Yujie Lu, William Yang Wang, Dan Roth | 2024-06-11T17:59:48Z | cs.CV, cs.AI, cs.CL | We present a novel task and benchmark for evaluating the ability of
text-to-image(T2I) generation models to produce images that fit commonsense in
real life, which we call Commonsense-T2I. Given two adversarial text prompts
containing an identical set of action words with minor differences, such as "a
lightbulb without electricity" v.s. "a lightbulb with electricity", we evaluate
whether T2I models can conduct visual-commonsense reasoning, e.g. produce
images that fit "the lightbulb is unlit" vs. "the lightbulb is lit"
correspondingly. Commonsense-T2I presents an adversarial challenge, providing
pairwise text prompts along with expected outputs. The dataset is carefully
hand-curated by experts and annotated with fine-grained labels, such as
commonsense type and likelihood of the expected outputs, to assist analyzing
model behavior. We benchmark a variety of state-of-the-art (sota) T2I models
and surprisingly find that, there is still a large gap between image synthesis
and real life photos--even the DALL-E 3 model could only achieve 48.92% on
Commonsense-T2I, and the stable diffusion XL model only achieves 24.92%
accuracy. Our experiments show that GPT-enriched prompts cannot solve this
challenge, and we include a detailed analysis about possible reasons for such
deficiency. We aim for Commonsense-T2I to serve as a high-quality evaluation
benchmark for T2I commonsense checking, fostering advancements in real life
image generation. | cs.CV | Commonsense-T2I Challenge: Can Text-to-Image Generation Models Understand Commonsense? |
2024-06-11T17:59:47Z | http://arxiv.org/pdf/2406.07545v1 | http://arxiv.org/abs/2406.07545v1 | Aidar Myrzakhan, Sondos Mahmoud Bsharat, Zhiqiang Shen | 2024-06-11T17:59:47Z | cs.CL, cs.AI | Multiple-choice questions (MCQ) are frequently used to assess large language
models (LLMs). Typically, an LLM is given a question and selects the answer
deemed most probable after adjustments for factors like length. Unfortunately,
LLMs may inherently favor certain answer choice IDs, such as A/B/C/D, due to
inherent biases of priori unbalanced probabilities, influencing the prediction
of answers based on these IDs. Previous research has introduced methods to
reduce this ''selection bias'' by simply permutating options on a few test
samples and applying to new ones. Another problem of MCQ is the lottery ticket
choice by ''random guessing''. The LLM does not learn particular knowledge, but
the option is guessed correctly. This situation is especially serious for those
small-scale LLMs. To address them, a more thorough approach involves shifting
from MCQ to open-style questions, which can fundamentally eliminate selection
bias and random guessing issues. However, transitioning causes its own set of
challenges in (1) identifying suitable open-style questions and (2) validating
the correctness of LLM open-style responses against human-annotated
ground-truths. This work aims to tackle these significant difficulties, and
establish a new LLM evaluation benchmark through entirely open-style questions.
Consequently, we introduce the Open-LLM-Leaderboard to track various LLMs'
performance and reflect true capability of them, such as GPT-4o/4/3.5, Claude
3, Gemini, etc. Our code and dataset are available at
https://github.com/VILA-Lab/Open-LLM-Leaderboard. | cs.CL | Open-LLM-Leaderboard: From Multi-choice to Open-style Questions for LLMs Evaluation, Benchmark, and Arena |
2024-06-11T17:59:45Z | http://arxiv.org/pdf/2406.07544v2 | http://arxiv.org/abs/2406.07544v2 | Yunze Man, Liang-Yan Gui, Yu-Xiong Wang | 2024-06-26T17:59:50Z | cs.CV, cs.AI, cs.CL, cs.LG | Being able to carry out complicated vision language reasoning tasks in 3D
space represents a significant milestone in developing household robots and
human-centered embodied AI. In this work, we demonstrate that a critical and
distinct challenge in 3D vision language reasoning is situational awareness,
which incorporates two key components: (1) The autonomous agent grounds its
self-location based on a language prompt. (2) The agent answers open-ended
questions from the perspective of its calculated position. To address this
challenge, we introduce SIG3D, an end-to-end Situation-Grounded model for 3D
vision language reasoning. We tokenize the 3D scene into sparse voxel
representation and propose a language-grounded situation estimator, followed by
a situated question answering module. Experiments on the SQA3D and ScanQA
datasets show that SIG3D outperforms state-of-the-art models in situation
estimation and question answering by a large margin (e.g., an enhancement of
over 30% on situation estimation accuracy). Subsequent analysis corroborates
our architectural design choices, explores the distinct functions of visual and
textual tokens, and highlights the importance of situational awareness in the
domain of 3D question answering. | cs.CV | Situational Awareness Matters in 3D Vision Language Reasoning |
2024-06-11T17:51:40Z | http://arxiv.org/pdf/2406.07524v1 | http://arxiv.org/abs/2406.07524v1 | Subham Sekhar Sahoo, Marianne Arriola, Yair Schiff, Aaron Gokaslan, Edgar Marroquin, Justin T Chiu, Alexander Rush, Volodymyr Kuleshov | 2024-06-11T17:51:40Z | cs.CL, cs.AI, cs.LG | While diffusion models excel at generating high-quality images, prior work
reports a significant performance gap between diffusion and autoregressive (AR)
methods in language modeling. In this work, we show that simple masked discrete
diffusion is more performant than previously thought. We apply an effective
training recipe that improves the performance of masked diffusion models and
derive a simplified, Rao-Blackwellized objective that results in additional
improvements. Our objective has a simple form -- it is a mixture of classical
masked language modeling losses -- and can be used to train encoder-only
language models that admit efficient samplers, including ones that can generate
arbitrary lengths of text semi-autoregressively like a traditional language
model. On language modeling benchmarks, a range of masked diffusion models
trained with modern engineering practices achieves a new state-of-the-art among
diffusion models, and approaches AR perplexity. We release our code at:
https://github.com/kuleshov-group/mdlm | cs.CL | Simple and Effective Masked Diffusion Language Models |
2024-06-11T17:50:51Z | http://arxiv.org/pdf/2406.07522v1 | http://arxiv.org/abs/2406.07522v1 | Liliang Ren, Yang Liu, Yadong Lu, Yelong Shen, Chen Liang, Weizhu Chen | 2024-06-11T17:50:51Z | cs.CL, cs.LG | Efficiently modeling sequences with infinite context length has been a
long-standing problem. Past works suffer from either the quadratic computation
complexity or the limited extrapolation ability on length generalization. In
this work, we present Samba, a simple hybrid architecture that layer-wise
combines Mamba, a selective State Space Model (SSM), with Sliding Window
Attention (SWA). Samba selectively compresses a given sequence into recurrent
hidden states while still maintaining the ability to precisely recall memories
with the attention mechanism. We scale Samba up to 3.8B parameters with 3.2T
training tokens and show that Samba substantially outperforms the
state-of-the-art models based on pure attention or SSMs on a wide range of
benchmarks. When trained on 4K length sequences, Samba can be efficiently
extrapolated to 256K context length with perfect memory recall and show
improved token predictions up to 1M context length. As a linear-time sequence
model, Samba enjoys a 3.73x higher throughput compared to Transformers with
grouped-query attention when processing user prompts of 128K length, and 3.64x
speedup when generating 64K tokens with unlimited streaming. A sample
implementation of Samba is publicly available in
https://github.com/microsoft/Samba. | cs.CL | Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling |
2024-06-11T17:40:00Z | http://arxiv.org/pdf/2406.07505v1 | http://arxiv.org/abs/2406.07505v1 | KBTG Labs, Danupat Khamnuansin, Atthakorn Petchsod, Anuruth Lertpiya, Pornchanan Balee, Thanawat Lodkaew, Tawunrat Chalothorn, Thadpong Pongthawornkamol, Monchai Lertsutthiwong | 2024-06-11T17:40:00Z | cs.CL | Recent advancements in Large Language Models (LLMs) have revealed new
capabilities and opportunities across the technological landscape. However, the
practicality of very large LLMs is challenged by their high compute cost, which
does not justify the benefits given their limited capability compared to
humans. While smaller, more practical LLMs have shown potential in financial
analysis, though they are not yet fully proficient, as evidenced by their
near-passing performance on the Chartered Financial Analyst (CFA) exam. In this
work, we present Financial Analyst Extension to our Text Hyperlocally Augmented
Large Language Extension (THaLLE), a series of 8B LLMs consistently achieving
highest performance on mock CFA exams against models of comparable size. We
thoroughly document the fine-tuning techniques used to facilitate future
research. Additionally, we introduce the use of Flare CFA, a publicly available
dataset for evaluating LLMs as a financial advisor. | cs.CL | THaLLE: Text Hyperlocally Augmented Large Language Extension -- Technical Report |
2024-06-11T17:39:46Z | http://arxiv.org/pdf/2406.07504v1 | http://arxiv.org/abs/2406.07504v1 | Atli Sigurgeirsson, Eddie L. Ungless | 2024-06-11T17:39:46Z | cs.CL | Modern voice cloning models claim to be able to capture a diverse range of
voices. We test the ability of a typical pipeline to capture the style known
colloquially as "gay voice" and notice a homogenisation effect: synthesised
speech is rated as sounding significantly "less gay" (by LGBTQ+ participants)
than its corresponding ground-truth for speakers with "gay voice", but ratings
actually increase for control speakers. Loss of "gay voice" has implications
for accessibility. We also find that for speakers with "gay voice", loss of
"gay voice" corresponds to lower similarity ratings.
However, we caution that improving the ability of such models to synthesise
``gay voice'' comes with a great number of risks. We use this pipeline as a
starting point for a discussion on the ethics of modelling queer voices more
broadly. Collecting "clean" queer data has safety and fairness ramifications,
and the resulting technology may cause harms from mockery to death. | cs.CL | Just Because We Camp, Doesn't Mean We Should: The Ethics of Modelling Queer Voices |
2024-06-11T17:37:45Z | http://arxiv.org/pdf/2406.07502v1 | http://arxiv.org/abs/2406.07502v1 | Renjie Pi, Jianshu Zhang, Jipeng Zhang, Rui Pan, Zhekai Chen, Tong Zhang | 2024-06-11T17:37:45Z | cs.CV, cs.CL | Image description datasets play a crucial role in the advancement of various
applications such as image understanding, text-to-image generation, and
text-image retrieval. Currently, image description datasets primarily originate
from two sources. One source is the scraping of image-text pairs from the web.
Despite their abundance, these descriptions are often of low quality and noisy.
Another is through human labeling. Datasets such as COCO are generally very
short and lack details. Although detailed image descriptions can be annotated
by humans, the high annotation cost limits the feasibility. These limitations
underscore the need for more efficient and scalable methods to generate
accurate and detailed image descriptions. In this paper, we propose an
innovative framework termed Image Textualization (IT), which automatically
produces high-quality image descriptions by leveraging existing multi-modal
large language models (MLLMs) and multiple vision expert models in a
collaborative manner, which maximally convert the visual information into text.
To address the current lack of benchmarks for detailed descriptions, we propose
several benchmarks for comprehensive evaluation, which verifies the quality of
image descriptions created by our framework. Furthermore, we show that
LLaVA-7B, benefiting from training on IT-curated descriptions, acquire improved
capability to generate richer image descriptions, substantially increasing the
length and detail of their output with less hallucination. | cs.CV | Image Textualization: An Automatic Framework for Creating Accurate and Detailed Image Descriptions |
2024-06-11T17:32:21Z | http://arxiv.org/pdf/2406.07496v1 | http://arxiv.org/abs/2406.07496v1 | Mert Yuksekgonul, Federico Bianchi, Joseph Boen, Sheng Liu, Zhi Huang, Carlos Guestrin, James Zou | 2024-06-11T17:32:21Z | cs.CL, cs.AI, cs.LG | AI is undergoing a paradigm shift, with breakthroughs achieved by systems
orchestrating multiple large language models (LLMs) and other complex
components. As a result, developing principled and automated optimization
methods for compound AI systems is one of the most important new challenges.
Neural networks faced a similar challenge in its early days until
backpropagation and automatic differentiation transformed the field by making
optimization turn-key. Inspired by this, we introduce TextGrad, a powerful
framework performing automatic ``differentiation'' via text. TextGrad
backpropagates textual feedback provided by LLMs to improve individual
components of a compound AI system. In our framework, LLMs provide rich,
general, natural language suggestions to optimize variables in computation
graphs, ranging from code snippets to molecular structures. TextGrad follows
PyTorch's syntax and abstraction and is flexible and easy-to-use. It works
out-of-the-box for a variety of tasks, where the users only provide the
objective function without tuning components or prompts of the framework. We
showcase TextGrad's effectiveness and generality across a diverse range of
applications, from question answering and molecule optimization to radiotherapy
treatment planning. Without modifying the framework, TextGrad improves the
zero-shot accuracy of GPT-4o in Google-Proof Question Answering from $51\%$ to
$55\%$, yields $20\%$ relative performance gain in optimizing LeetCode-Hard
coding problem solutions, improves prompts for reasoning, designs new druglike
small molecules with desirable in silico binding, and designs radiation
oncology treatment plans with high specificity. TextGrad lays a foundation to
accelerate the development of the next-generation of AI systems. | cs.CL | TextGrad: Automatic "Differentiation" via Text |
2024-06-11T17:30:22Z | http://arxiv.org/pdf/2406.07494v2 | http://arxiv.org/abs/2406.07494v2 | Frederic Kirstein, Jan Philip Wahle, Bela Gipp, Terry Ruas | 2024-06-12T10:47:09Z | cs.CL, cs.AI | Abstractive dialogue summarization is the task of distilling conversations
into informative and concise summaries. Although reviews have been conducted on
this topic, there is a lack of comprehensive work detailing the challenges of
dialogue summarization, unifying the differing understanding of the task, and
aligning proposed techniques, datasets, and evaluation metrics with the
challenges. This article summarizes the research on Transformer-based
abstractive summarization for English dialogues by systematically reviewing
1262 unique research papers published between 2019 and 2024, relying on the
Semantic Scholar and DBLP databases. We cover the main challenges present in
dialog summarization (i.e., language, structure, comprehension, speaker,
salience, and factuality) and link them to corresponding techniques such as
graph-based approaches, additional training tasks, and planning strategies,
which typically overly rely on BART-based encoder-decoder models. We find that
while some challenges, like language, have seen considerable progress, mainly
due to training methods, others, such as comprehension, factuality, and
salience, remain difficult and hold significant research opportunities. We
investigate how these approaches are typically assessed, covering the datasets
for the subdomains of dialogue (e.g., meeting, medical), the established
automatic metrics and human evaluation approaches for assessing scores and
annotator agreement. We observe that only a few datasets span across all
subdomains. The ROUGE metric is the most used, while human evaluation is
frequently reported without sufficient detail on inner-annotator agreement and
annotation guidelines. Additionally, we discuss the possible implications of
the recently explored large language models and conclude that despite a
potential shift in relevance and difficulty, our described challenge taxonomy
remains relevant. | cs.CL | CADS: A Systematic Literature Review on the Challenges of Abstractive Dialogue Summarization |
2024-06-11T17:30:03Z | http://arxiv.org/pdf/2406.07492v1 | http://arxiv.org/abs/2406.07492v1 | MohammadHossein Rezaei, Eduardo Blanco | 2024-06-11T17:30:03Z | cs.CL | Negation is a common linguistic phenomenon. Yet language models face
challenges with negation in many natural language understanding tasks such as
question answering and natural language inference. In this paper, we experiment
with seamless strategies that incorporate affirmative interpretations (i.e.,
paraphrases without negation) to make models more robust against negation.
Crucially, our affirmative interpretations are obtained automatically. We show
improvements with CondaQA, a large corpus requiring reasoning with negation,
and five natural language understanding tasks. | cs.CL | Paraphrasing in Affirmative Terms Improves Negation Understanding |
2024-06-11T17:26:07Z | http://arxiv.org/pdf/2406.07483v1 | http://arxiv.org/abs/2406.07483v1 | Mao Li, Frederick Conrad | 2024-06-11T17:26:07Z | cs.CL | In the rapidly evolving landscape of Natural Language Processing (NLP), the
use of Large Language Models (LLMs) for automated text annotation in social
media posts has garnered significant interest. Despite the impressive
innovations in developing LLMs like ChatGPT, their efficacy, and accuracy as
annotation tools are not well understood. In this paper, we analyze the
performance of eight open-source and proprietary LLMs for annotating the stance
expressed in social media posts, benchmarking their performance against human
annotators' (i.e., crowd-sourced) judgments. Additionally, we investigate the
conditions under which LLMs are likely to disagree with human judgment. A
significant finding of our study is that the explicitness of text expressing a
stance plays a critical role in how faithfully LLMs' stance judgments match
humans'. We argue that LLMs perform well when human annotators do, and when
LLMs fail, it often corresponds to situations in which human annotators
struggle to reach an agreement. We conclude with recommendations for a
comprehensive approach that combines the precision of human expertise with the
scalability of LLM predictions. This study highlights the importance of
improving the accuracy and comprehensiveness of automated stance detection,
aiming to advance these technologies for more efficient and unbiased analysis
of social media. | cs.CL | Advancing Annotation of Stance in Social Media Posts: A Comparative Analysis of Large Language Models and Crowd Sourcing |
2024-06-11T17:22:23Z | http://arxiv.org/pdf/2406.07476v2 | http://arxiv.org/abs/2406.07476v2 | Zesen Cheng, Sicong Leng, Hang Zhang, Yifei Xin, Xin Li, Guanzheng Chen, Yongxin Zhu, Wenqi Zhang, Ziyang Luo, Deli Zhao, Lidong Bing | 2024-06-17T16:40:43Z | cs.CV, cs.CL | In this paper, we present the VideoLLaMA 2, a set of Video Large Language
Models (Video-LLMs) designed to enhance spatial-temporal modeling and audio
understanding in video and audio-oriented tasks. Building upon its predecessor,
VideoLLaMA 2 incorporates a tailor-made Spatial-Temporal Convolution (STC)
connector, which effectively captures the intricate spatial and temporal
dynamics of video data. Additionally, we integrate an Audio Branch into the
model through joint training, thereby enriching the multimodal understanding
capabilities of the model by seamlessly incorporating audio cues. Comprehensive
evaluations on multiple-choice video question answering (MC-VQA), open-ended
video question answering (OE-VQA), and video captioning (VC) tasks demonstrate
that VideoLLaMA 2 consistently achieves competitive results among open-source
models and even gets close to some proprietary models on several benchmarks.
Furthermore, VideoLLaMA 2 exhibits reasonable improvements in audio-only and
audio-video question-answering (AQA & OE-AVQA) benchmarks over existing models.
These advancements underline VideoLLaMA 2's superior performance in multimodal
comprehension, setting a new standard for intelligent video analysis systems.
All models are public to facilitate further research. | cs.CV | VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs |
2024-06-11T17:12:41Z | http://arxiv.org/pdf/2406.07466v1 | http://arxiv.org/abs/2406.07466v1 | John Murzaku, Adil Soubki, Owen Rambow | 2024-06-11T17:12:41Z | cs.CL, cs.LG, cs.SD, eess.AS | Recognizing a speaker's level of commitment to a belief is a difficult task;
humans do not only interpret the meaning of the words in context, but also
understand cues from intonation and other aspects of the audio signal. Many
papers and corpora in the NLP community have approached the belief prediction
task using text-only approaches. We are the first to frame and present results
on the multimodal belief prediction task. We use the CB-Prosody corpus (CBP),
containing aligned text and audio with speaker belief annotations. We first
report baselines and significant features using acoustic-prosodic features and
traditional machine learning methods. We then present text and audio baselines
for the CBP corpus fine-tuning on BERT and Whisper respectively. Finally, we
present our multimodal architecture which fine-tunes on BERT and Whisper and
uses multiple fusion methods, improving on both modalities alone. | cs.CL | Multimodal Belief Prediction |
2024-06-11T17:08:16Z | http://arxiv.org/pdf/2406.10270v1 | http://arxiv.org/abs/2406.10270v1 | Temitayo Adefemi | 2024-06-11T17:08:16Z | cs.CL, cs.LG | Trie-Augmented Neural Networks (TANNs) combine trie structures with neural
networks, forming a hierarchical design that enhances decision-making
transparency and efficiency in machine learning. This paper investigates the
use of TANNs for text and document classification, applying Recurrent Neural
Networks (RNNs) and Feed forward Neural Networks (FNNs). We evaluated TANNs on
the 20 NewsGroup and SMS Spam Collection datasets, comparing their performance
with traditional RNN and FFN Networks with and without dropout regularization.
The results show that TANNs achieve similar or slightly better performance in
text classification. The primary advantage of TANNs is their structured
decision-making process, which improves interpretability. We discuss
implementation challenges and practical limitations. Future work will aim to
refine the TANNs architecture for more complex classification tasks. | cs.CL | A Conceptual Framework For Trie-Augmented Neural Networks (TANNS) |
2024-06-11T16:51:14Z | http://arxiv.org/pdf/2406.07444v1 | http://arxiv.org/abs/2406.07444v1 | Shiao Meng, Xuming Hu, Aiwei Liu, Fukun Ma, Yawen Yang, Shuang Li, Lijie Wen | 2024-06-11T16:51:14Z | cs.CL, 68T50, I.2.7 | Driven by the demand for cross-sentence and large-scale relation extraction,
document-level relation extraction (DocRE) has attracted increasing research
interest. Despite the continuous improvement in performance, we find that
existing DocRE models which initially perform well may make more mistakes when
merely changing the entity names in the document, hindering the generalization
to novel entity names. To this end, we systematically investigate the
robustness of DocRE models to entity name variations in this work. We first
propose a principled pipeline to generate entity-renamed documents by replacing
the original entity names with names from Wikidata. By applying the pipeline to
DocRED and Re-DocRED datasets, we construct two novel benchmarks named
Env-DocRED and Env-Re-DocRED for robustness evaluation. Experimental results
show that both three representative DocRE models and two in-context learned
large language models consistently lack sufficient robustness to entity name
variations, particularly on cross-sentence relation instances and documents
with more entities. Finally, we propose an entity variation robust training
method which not only improves the robustness of DocRE models but also enhances
their understanding and reasoning capabilities. We further verify that the
basic idea of this method can be effectively transferred to in-context learning
for DocRE as well. | cs.CL | On the Robustness of Document-Level Relation Extraction Models to Entity Name Variations |
2024-06-11T16:48:17Z | http://arxiv.org/pdf/2406.07440v2 | http://arxiv.org/abs/2406.07440v2 | Kun Sun, Rong Wang | 2024-07-01T09:30:34Z | cs.CL, cs.AI | Machine Translation (MT) Quality Estimation (QE) assesses translation
reliability without reference texts. This study introduces "textual similarity"
as a new metric for QE, using sentence transformers and cosine similarity to
measure semantic closeness. Analyzing data from the MLQE-PE dataset, we found
that textual similarity exhibits stronger correlations with human scores than
traditional metrics (hter, model evaluation, sentence probability etc.).
Employing GAMMs as a statistical tool, we demonstrated that textual similarity
consistently outperforms other metrics across multiple language pairs in
predicting human scores. We also found that "hter" actually failed to predict
human scores in QE. Our findings highlight the effectiveness of textual
similarity as a robust QE metric, recommending its integration with other
metrics into QE frameworks and MT system training for improved accuracy and
usability. | cs.CL | Textual Similarity as a Key Metric in Machine Translation Quality Estimation |
2024-06-11T16:34:02Z | http://arxiv.org/pdf/2406.07430v1 | http://arxiv.org/abs/2406.07430v1 | Yimeng Gu, Mengqi Zhang, Ignacio Castro, Shu Wu, Gareth Tyson | 2024-06-11T16:34:02Z | cs.CL, cs.MM | Multimodal out-of-context news is a common type of misinformation on online
media platforms. This involves posting a caption, alongside an invalid
out-of-context news image. Reflecting its importance, researchers have
developed models to detect such misinformation. However, a common limitation of
these models is that they only consider the scenario where pre-labeled data is
available for each domain, failing to address the out-of-context news detection
on unlabeled domains (e.g., unverified news on new topics or agencies). In this
work, we therefore focus on domain adaptive out-of-context news detection. In
order to effectively adapt the detection model to unlabeled news topics or
agencies, we propose ConDA-TTA (Contrastive Domain Adaptation with Test-Time
Adaptation) which applies contrastive learning and maximum mean discrepancy
(MMD) to learn the domain-invariant feature. In addition, it leverages target
domain statistics during test-time to further assist domain adaptation.
Experimental results show that our approach outperforms baselines in 5 out of 7
domain adaptation settings on two public datasets, by as much as 2.93% in F1
and 2.08% in accuracy. | cs.CL | Learning Domain-Invariant Features for Out-of-Context News Detection |
2024-06-11T16:26:18Z | http://arxiv.org/pdf/2406.07424v2 | http://arxiv.org/abs/2406.07424v2 | Genta Indra Winata, Ruochen Zhang, David Ifeoluwa Adelani | 2024-06-19T08:29:51Z | cs.CL | Words have been represented in a high-dimensional vector space that encodes
their semantic similarities, enabling downstream applications such as
retrieving synonyms, antonyms, and relevant contexts. However, despite recent
advances in multilingual language models (LMs), the effectiveness of these
models' representations in semantic retrieval contexts has not been
comprehensively explored. To fill this gap, this paper introduces the MINERS, a
benchmark designed to evaluate the ability of multilingual LMs in semantic
retrieval tasks, including bitext mining and classification via
retrieval-augmented contexts. We create a comprehensive framework to assess the
robustness of LMs in retrieving samples across over 200 diverse languages,
including extremely low-resource languages in challenging cross-lingual and
code-switching settings. Our results demonstrate that by solely retrieving
semantically similar embeddings yields performance competitive with
state-of-the-art approaches, without requiring any fine-tuning. | cs.CL | MINERS: Multilingual Language Models as Semantic Retrievers |
2024-06-11T16:15:06Z | http://arxiv.org/pdf/2406.07411v1 | http://arxiv.org/abs/2406.07411v1 | Tongtong Wu, Weigang Wu, Xingyu Wang, Kang Xu, Suyu Ma, Bo Jiang, Ping Yang, Zhenchang Xing, Yuan-Fang Li, Gholamreza Haffari | 2024-06-11T16:15:06Z | cs.SE, cs.CL | Significant research has focused on improving the performance of large
language model on code-related tasks due to their practical importance.
Although performance is typically evaluated using public benchmark datasets,
the existing datasets do not account for the concept of \emph{version}, which
is crucial in professional software development. In this paper, we introduce
VersiCode, the first comprehensive dataset designed to assess the ability of
large language models to generate verifiable code for specific library
versions. VersiCode encompasses 300 libraries across more than 2,000 versions
spanning 9 years. We design two dedicated evaluation tasks: version-specific
code completion (VSCC) and version-aware code editing (VACE). Comprehensive
experiments are conducted to benchmark the performance of LLMs, revealing the
challenging nature of these tasks and VersiCode, that even state-of-the-art
LLMs struggle to generate version-correct code. This dataset, together with the
proposed tasks, sheds light on LLMs' capabilities and limitations in handling
version-specific code generation, and opens up an important new area of
research for further investigation. The resources can be found at
https://github.com/wutong8023/VersiCode. | cs.SE | VersiCode: Towards Version-controllable Code Generation |
2024-06-11T16:09:53Z | http://arxiv.org/pdf/2406.10269v1 | http://arxiv.org/abs/2406.10269v1 | Alexandre Bonlarron, Jean-Charles Régin | 2024-06-11T16:09:53Z | cs.CL, cs.AI, cs.LG | This paper presents NgramMarkov, a variant of the Markov constraints. It is
dedicated to text generation in constraint programming (CP). It involves a set
of n-grams (i.e., sequence of n words) associated with probabilities given by a
large language model (LLM). It limits the product of the probabilities of the
n-gram of a sentence. The propagator of this constraint can be seen as an
extension of the ElementaryMarkov constraint propagator, incorporating the LLM
distribution instead of the maximum likelihood estimation of n-grams. It uses a
gliding threshold, i.e., it rejects n-grams whose local probabilities are too
low, to guarantee balanced solutions. It can also be combined with a
"look-ahead" approach to remove n-grams that are very unlikely to lead to
acceptable sentences for a fixed-length horizon. This idea is based on the
MDDMarkovProcess constraint propagator, but without explicitly using an MDD
(Multi-Valued Decision Diagram). The experimental results show that the
generated text is valued in a similar way to the LLM perplexity function. Using
this new constraint dramatically reduces the number of candidate sentences
produced, improves computation times, and allows larger corpora or smaller
n-grams to be used. A real-world problem has been solved for the first time
using 4-grams instead of 5-grams. | cs.CL | Markov Constraint as Large Language Model Surrogate |
2024-06-11T15:58:59Z | http://arxiv.org/pdf/2406.07393v2 | http://arxiv.org/abs/2406.07393v2 | Peng Hu, Changjiang Gao, Ruiqi Gao, Jiajun Chen, Shujian Huang | 2024-06-24T14:59:54Z | cs.CL | Large Language Models (LLMs) have demonstrated strong capabilities as
knowledge bases and significant in-context reasoning capabilities. However,
previous work challenges their out-of-context reasoning ability, i.e., the
ability to infer information from their training data, instead of from the
context or prompt. This paper focuses on a significant facet of out-of-context
reasoning: Out-of-Context Knowledge Reasoning (OCKR), which is to combine
multiple knowledge to infer new knowledge. We designed a synthetic dataset with
seven representative OCKR tasks to systematically assess the OCKR capabilities
of LLMs. Using this dataset, we evaluated the LLaMA2-13B-chat model and
discovered that its proficiency in this aspect is limited, regardless of
whether the knowledge is trained in a separate or adjacent training settings.
Moreover, training the model to reason with complete reasoning data did not
result in significant improvement. Training the model to perform explicit
knowledge retrieval helps in only one of the tasks, indicating that the model's
limited OCKR capabilities are due to difficulties in retrieving relevant
knowledge. Furthermore, we treat cross-lingual knowledge transfer as a distinct
form of OCKR, and evaluate this ability. Our results show that the evaluated
model also exhibits limited ability in transferring knowledge across languages.
The dataset used in this study is available at
https://github.com/NJUNLP/ID-OCKR. | cs.CL | Limited Out-of-Context Knowledge Reasoning in Large Language Models |
2024-06-11T15:45:24Z | http://arxiv.org/pdf/2406.07378v1 | http://arxiv.org/abs/2406.07378v1 | Kai-Hendrik Cohrs, Gherardo Varando, Emiliano Diaz, Vasileios Sitokonstantinou, Gustau Camps-Valls | 2024-06-11T15:45:24Z | cs.AI, cs.CL | Causality is essential for understanding complex systems, such as the
economy, the brain, and the climate. Constructing causal graphs often relies on
either data-driven or expert-driven approaches, both fraught with challenges.
The former methods, like the celebrated PC algorithm, face issues with data
requirements and assumptions of causal sufficiency, while the latter demand
substantial time and domain knowledge. This work explores the capabilities of
Large Language Models (LLMs) as an alternative to domain experts for causal
graph generation. We frame conditional independence queries as prompts to LLMs
and employ the PC algorithm with the answers. The performance of the LLM-based
conditional independence oracle on systems with known causal graphs shows a
high degree of variability. We improve the performance through a proposed
statistical-inspired voting schema that allows some control over false-positive
and false-negative rates. Inspecting the chain-of-thought argumentation, we
find causal reasoning to justify its answer to a probabilistic query. We show
evidence that knowledge-based CIT could eventually become a complementary tool
for data-driven causal discovery. | cs.AI | Large Language Models for Constrained-Based Causal Discovery |
2024-06-11T15:34:43Z | http://arxiv.org/pdf/2406.07368v1 | http://arxiv.org/abs/2406.07368v1 | Haoran You, Yichao Fu, Zheng Wang, Amir Yazdanbakhsh, Yingyan, Lin | 2024-06-11T15:34:43Z | cs.CL, cs.AI, cs.LG | Autoregressive Large Language Models (LLMs) have achieved impressive
performance in language tasks but face two significant bottlenecks: (1)
quadratic complexity in the attention module as the number of tokens increases,
and (2) limited efficiency due to the sequential processing nature of
autoregressive LLMs during generation. While linear attention and speculative
decoding offer potential solutions, their applicability and synergistic
potential for enhancing autoregressive LLMs remain uncertain. We conduct the
first comprehensive study on the efficacy of existing linear attention methods
for autoregressive LLMs, integrating them with speculative decoding. We
introduce an augmentation technique for linear attention that ensures
compatibility with speculative decoding, enabling more efficient training and
serving of LLMs. Extensive experiments and ablation studies involving seven
existing linear attention models and five encoder/decoder-based LLMs
consistently validate the effectiveness of our augmented linearized LLMs.
Notably, our approach achieves up to a 6.67 reduction in perplexity on the
LLaMA model and up to a 2$\times$ speedup during generation compared to prior
linear attention methods. Codes and models are available at
https://github.com/GATECH-EIC/Linearized-LLM. | cs.CL | When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Models |
2024-06-11T15:32:32Z | http://arxiv.org/pdf/2406.07365v1 | http://arxiv.org/abs/2406.07365v1 | Yinhao Bai, Yalan Xie, Xiaoyi Liu, Yuhua Zhao, Zhixin Han, Mengting Hu, Hang Gao, Renhong Cheng | 2024-06-11T15:32:32Z | cs.CL, cs.AI | Aspect sentiment quad prediction (ASQP) aims to predict four aspect-based
elements, including aspect term, opinion term, aspect category, and sentiment
polarity. In practice, unseen aspects, due to distinct data distribution,
impose many challenges for a trained neural model. Motivated by this, this work
formulates ASQP into the few-shot scenario, which aims for fast adaptation in
real applications. Therefore, we first construct a few-shot ASQP dataset (FSQP)
that contains richer categories and is more balanced for the few-shot study.
Moreover, recent methods extract quads through a generation paradigm, which
involves converting the input sentence into a templated target sequence.
However, they primarily focus on the utilization of a single template or the
consideration of different template orders, thereby overlooking the
correlations among various templates. To tackle this issue, we further propose
a Broadview Soft Prompting (BvSP) method that aggregates multiple templates
with a broader view by taking into account the correlation between the
different templates. Specifically, BvSP uses the pre-trained language model to
select the most relevant k templates with Jensen-Shannon divergence. BvSP
further introduces soft prompts to guide the pre-trained language model using
the selected templates. Then, we aggregate the results of multi-templates by
voting mechanism. Empirical results demonstrate that BvSP significantly
outperforms the stateof-the-art methods under four few-shot settings and other
public datasets. Our code and dataset are available at
https://github.com/byinhao/BvSP. | cs.CL | BvSP: Broad-view Soft Prompting for Few-Shot Aspect Sentiment Quad Prediction |
2024-06-11T15:30:26Z | http://arxiv.org/pdf/2406.10268v1 | http://arxiv.org/abs/2406.10268v1 | Chenyan Zhao, Mariana Silva, Seth Poulsen | 2024-06-11T15:30:26Z | cs.AI, cs.CL, cs.HC | In mathematical proof education, there remains a need for interventions that
help students learn to write mathematical proofs. Research has shown that
timely feedback can be very helpful to students learning new skills. While for
many years natural language processing models have struggled to perform well on
tasks related to mathematical texts, recent developments in natural language
processing have created the opportunity to complete the task of giving students
instant feedback on their mathematical proofs. In this paper, we present a set
of training methods and models capable of autograding freeform mathematical
proofs by leveraging existing large language models and other machine learning
techniques. The models are trained using proof data collected from four
different proof by induction problems. We use four different robust large
language models to compare their performances, and all achieve satisfactory
performances to various degrees. Additionally, we recruit human graders to
grade the same proofs as the training data, and find that the best grading
model is also more accurate than most human graders. With the development of
these grading models, we create and deploy an autograder for proof by induction
problems and perform a user study with students. Results from the study shows
that students are able to make significant improvements to their proofs using
the feedback from the autograder, but students still do not trust the AI
autograders as much as they trust human graders. Future work can improve on the
autograder feedback and figure out ways to help students trust AI autograders. | cs.AI | Autograding Mathematical Induction Proofs with Natural Language Processing |
2024-06-11T15:27:01Z | http://arxiv.org/pdf/2406.07359v1 | http://arxiv.org/abs/2406.07359v1 | Maxime Darrin, Ines Arous, Pablo Piantanida, Jackie CK Cheung | 2024-06-11T15:27:01Z | cs.CL | Scientific peer review is essential for the quality of academic publications.
However, the increasing number of paper submissions to conferences has strained
the reviewing process. This surge poses a burden on area chairs who have to
carefully read an ever-growing volume of reviews and discern each reviewer's
main arguments as part of their decision process. In this paper, we introduce
\sys, a summarization method designed to offer a concise yet comprehensive
overview of scholarly reviews. Unlike traditional consensus-based methods, \sys
extracts both common and unique opinions from the reviews. We introduce novel
uniqueness scores based on the Rational Speech Act framework to identify
relevant sentences in the reviews. Our method aims to provide a pragmatic
glimpse into all reviews, offering a balanced perspective on their opinions.
Our experimental results with both automatic metrics and human evaluation show
that \sys generates more discriminative summaries than baseline methods in
terms of human evaluation while achieving comparable performance with these
methods in terms of automatic metrics. | cs.CL | GLIMPSE: Pragmatically Informative Multi-Document Summarization for Scholarly Reviews |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.