publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-11-29T16:11:45Z | http://arxiv.org/pdf/2311.17771v1 | http://arxiv.org/abs/2311.17771v1 | Simão Gonçalves, Gonçalo Correia, Diogo Pernes, Afonso Mendes | 2023-11-29T16:11:45Z | cs.CL | The centroid method is a simple approach for extractive multi-document
summarization and many improvements to its pipeline have been proposed. We
further refine it by adding a beam search process to the sentence selection and
also a centroid estimation attention model that leads to improved results. We
demonstrate this in several multi-document summarization datasets, including in
a multilingual scenario. | cs.CL | Supervising the Centroid Baseline for Extractive Multi-Document Summarization |
2023-11-29T16:02:06Z | http://arxiv.org/pdf/2311.17972v3 | http://arxiv.org/abs/2311.17972v3 | Lin Zheng, Jianbo Yuan, Zhi Zhang, Hongxia Yang, Lingpeng Kong | 2024-05-26T18:15:39Z | cs.PL, cs.CL, cs.LG | This work introduces self-infilling code generation, a general framework that
incorporates infilling operations into auto-regressive decoding. Our approach
capitalizes on the observation that recent infilling-capable code language
models can self-infill: whereas infilling operations aim to fill in the middle
based on a predefined prefix and suffix, self-infilling sequentially generates
both such surrounding context and the infilled content. We utilize this
capability to introduce novel interruption and looping mechanisms in
conventional decoding, evolving it into a non-monotonic process. Interruptions
allow for postponing the generation of specific code until a definitive suffix
is established, enhancing control over the output. Meanwhile, the looping
mechanism, which leverages the complementary nature of self-infilling and
left-to-right decoding, can iteratively update and synchronize each piece of
generation cyclically. Extensive experiments are conducted to demonstrate that
our proposed decoding process is effective in enhancing both regularity and
quality across several code generation benchmarks. | cs.PL | Self-Infilling Code Generation |
2023-11-29T15:49:24Z | http://arxiv.org/pdf/2311.17743v1 | http://arxiv.org/abs/2311.17743v1 | Lokesh Madasu, Gopichand Kanumolu, Nirmal Surange, Manish Shrivastava | 2023-11-29T15:49:24Z | cs.CL, cs.AI | The task of headline generation within the realm of Natural Language
Processing (NLP) holds immense significance, as it strives to distill the true
essence of textual content into concise and attention-grabbing summaries. While
noteworthy progress has been made in headline generation for widely spoken
languages like English, there persist numerous challenges when it comes to
generating headlines in low-resource languages, such as the rich and diverse
Indian languages. A prominent obstacle that specifically hinders headline
generation in Indian languages is the scarcity of high-quality annotated data.
To address this crucial gap, we proudly present Mukhyansh, an extensive
multilingual dataset, tailored for Indian language headline generation.
Comprising an impressive collection of over 3.39 million article-headline
pairs, Mukhyansh spans across eight prominent Indian languages, namely Telugu,
Tamil, Kannada, Malayalam, Hindi, Bengali, Marathi, and Gujarati. We present a
comprehensive evaluation of several state-of-the-art baseline models.
Additionally, through an empirical analysis of existing works, we demonstrate
that Mukhyansh outperforms all other models, achieving an impressive average
ROUGE-L score of 31.43 across all 8 languages. | cs.CL | Mukhyansh: A Headline Generation Dataset for Indic Languages |
2023-11-29T15:44:39Z | http://arxiv.org/pdf/2311.17741v1 | http://arxiv.org/abs/2311.17741v1 | Can Cui, Imran Ahamad Sheikh, Mostafa Sadeghi, Emmanuel Vincent | 2023-11-29T15:44:39Z | cs.CL, cs.SD, eess.AS | Joint rich and normalized automatic speech recognition (ASR), that produces
transcriptions both with and without punctuation and capitalization, remains a
challenge. End-to-end (E2E) ASR models offer both convenience and the ability
to perform such joint transcription of speech. Training such models requires
paired speech and rich text data, which is not widely available. In this paper,
we compare two different approaches to train a stateless Transducer-based E2E
joint rich and normalized ASR system, ready for streaming applications, with a
limited amount of rich labeled data. The first approach uses a language model
to generate pseudo-rich transcriptions of normalized training data. The second
approach uses a single decoder conditioned on the type of the output. The first
approach leads to E2E rich ASR which perform better on out-of-domain data, with
up to 9% relative reduction in errors. The second approach demonstrates the
feasibility of an E2E joint rich and normalized ASR system using as low as 5%
rich training data with moderate (2.42% absolute) increase in errors. | cs.CL | End-to-end Joint Rich and Normalized ASR with a limited amount of rich training data |
2023-11-29T15:21:35Z | http://arxiv.org/pdf/2311.17722v1 | http://arxiv.org/abs/2311.17722v1 | Tanmay Chavan, Shantanu Patankar, Aditya Kane, Omkar Gokhale, Geetanjali Kale, Raviraj Joshi | 2023-11-29T15:21:35Z | cs.CL, cs.LG | Contrastive learning has proven to be an effective method for pre-training
models using weakly labeled data in the vision domain. Sentence transformers
are the NLP counterparts to this architecture, and have been growing in
popularity due to their rich and effective sentence representations. Having
effective sentence representations is paramount in multiple tasks, such as
information retrieval, retrieval augmented generation (RAG), and sentence
comparison. Keeping in mind the deployability factor of transformers,
evaluating the robustness of sentence transformers is of utmost importance.
This work focuses on evaluating the robustness of the sentence encoders. We
employ several adversarial attacks to evaluate its robustness. This system uses
character-level attacks in the form of random character substitution,
word-level attacks in the form of synonym replacement, and sentence-level
attacks in the form of intra-sentence word order shuffling. The results of the
experiments strongly undermine the robustness of sentence encoders. The models
produce significantly different predictions as well as embeddings on perturbed
datasets. The accuracy of the models can fall up to 15 percent on perturbed
datasets as compared to unperturbed datasets. Furthermore, the experiments
demonstrate that these embeddings does capture the semantic and syntactic
structure (sentence order) of sentences. However, existing supervised
classification strategies fail to leverage this information, and merely
function as n-gram detectors. | cs.CL | SenTest: Evaluating Robustness of Sentence Encoders |
2023-11-29T15:02:46Z | http://arxiv.org/pdf/2311.17696v3 | http://arxiv.org/abs/2311.17696v3 | Chenxi Dong | 2024-06-21T09:29:07Z | cs.CL | This paper proposes a low-code solution to build an AI tutor that leverages
advanced AI techniques to provide accurate and contextually relevant responses
in a personalized learning environment. The OpenAI Assistants API allows AI
Tutor to easily embed, store, retrieve, and manage files and chat history,
enabling a low-code solution. Large Language Models (LLMs) and
Retrieval-Augmented Generation (RAG) technology generate sophisticated answers
based on course-specific materials. The application efficiently organizes and
retrieves relevant information through vector embedding and similarity-based
retrieval algorithms. The AI Tutor prototype demonstrates its ability to
generate relevant, accurate answers with source citations. It represents a
significant advancement in technology-enhanced tutoring systems, democratizing
access to high-quality, customized educational support in higher education. | cs.CL | How to Build an AI Tutor that Can Adapt to Any Course and Provide Accurate Answers Using Large Language Model and Retrieval-Augmented Generation |
2023-11-29T14:49:31Z | http://arxiv.org/pdf/2311.17686v1 | http://arxiv.org/abs/2311.17686v1 | Liya Wang, Jason Chou, Xin Zhou, Alex Tien, Diane M Baumgartner | 2023-11-29T14:49:31Z | cs.CL, cs.AI | The advent of ChatGPT and GPT-4 has captivated the world with large language
models (LLMs), demonstrating exceptional performance in question-answering,
summarization, and content generation. The aviation industry is characterized
by an abundance of complex, unstructured text data, replete with technical
jargon and specialized terminology. Moreover, labeled data for model building
are scarce in this domain, resulting in low usage of aviation text data. The
emergence of LLMs presents an opportunity to transform this situation, but
there is a lack of LLMs specifically designed for the aviation domain. To
address this gap, we propose AviationGPT, which is built on open-source LLaMA-2
and Mistral architectures and continuously trained on a wealth of carefully
curated aviation datasets. Experimental results reveal that AviationGPT offers
users multiple advantages, including the versatility to tackle diverse natural
language processing (NLP) problems (e.g., question-answering, summarization,
document writing, information extraction, report querying, data cleaning, and
interactive data exploration). It also provides accurate and contextually
relevant responses within the aviation domain and significantly improves
performance (e.g., over a 40% performance gain in tested cases). With
AviationGPT, the aviation industry is better equipped to address more complex
research problems and enhance the efficiency and safety of National Airspace
System (NAS) operations. | cs.CL | AviationGPT: A Large Language Model for the Aviation Domain |
2023-11-29T14:39:38Z | http://arxiv.org/pdf/2311.17676v1 | http://arxiv.org/abs/2311.17676v1 | Jonathan Ivey, Susan Gauch | 2023-11-29T14:39:38Z | cs.CL, cs.AI | Psychological stress detection is an important task for mental healthcare
research, but there has been little prior work investigating the effectiveness
of psychological stress models on minority individuals, who are especially
vulnerable to poor mental health outcomes. In this work, we use the related
task of minority stress detection to evaluate the ability of psychological
stress models to understand the language of sexual and gender minorities. We
find that traditional psychological stress models underperform on minority
stress detection, and we propose using emotion-infused models to reduce that
performance disparity. We further demonstrate that multi-task psychological
stress models outperform the current state-of-the-art for minority stress
detection without directly training on minority stress data. We provide
explanatory analysis showing that minority communities have different
distributions of emotions than the general population and that emotion-infused
models improve the performance of stress models on underrepresented groups
because of their effectiveness in low-data environments, and we propose that
integrating emotions may benefit underrepresented groups in other mental health
detection tasks. | cs.CL | Improving Minority Stress Detection with Emotions |
2023-11-29T14:30:16Z | http://arxiv.org/pdf/2311.17667v2 | http://arxiv.org/abs/2311.17667v2 | Zheng Chu, Jingchang Chen, Qianglong Chen, Weijiang Yu, Haotian Wang, Ming Liu, Bing Qin | 2024-06-28T10:40:26Z | cs.CL, cs.AI | Grasping the concept of time is a fundamental facet of human cognition,
indispensable for truly comprehending the intricacies of the world. Previous
studies typically focus on specific aspects of time, lacking a comprehensive
temporal reasoning benchmark. To address this, we propose TimeBench, a
comprehensive hierarchical temporal reasoning benchmark that covers a broad
spectrum of temporal reasoning phenomena. TimeBench provides a thorough
evaluation for investigating the temporal reasoning capabilities of large
language models. We conduct extensive experiments on GPT-4, LLaMA2, and other
popular LLMs under various settings. Our experimental results indicate a
significant performance gap between the state-of-the-art LLMs and humans,
highlighting that there is still a considerable distance to cover in temporal
reasoning. Besides, LLMs exhibit capability discrepancies across different
reasoning categories. Furthermore, we thoroughly analyze the impact of multiple
aspects on temporal reasoning and emphasize the associated challenges. We
aspire for TimeBench to serve as a comprehensive benchmark, fostering research
in temporal reasoning. Resources are available at:
https://github.com/zchuz/TimeBench | cs.CL | TimeBench: A Comprehensive Evaluation of Temporal Reasoning Abilities in Large Language Models |
2023-11-29T14:08:53Z | http://arxiv.org/pdf/2311.17647v2 | http://arxiv.org/abs/2311.17647v2 | Xiujun Li, Yujie Lu, Zhe Gan, Jianfeng Gao, William Yang Wang, Yejin Choi | 2024-06-10T23:39:24Z | cs.CV, cs.AI, cs.CL | Recent multimodal large language models (MLLMs) have shown promising
instruction following capabilities on vision-language tasks. In this work, we
introduce VISUAL MODALITY INSTRUCTION (VIM), and investigate how well
multimodal models can understand textual instructions provided in pixels,
despite not being explicitly trained on such data during pretraining or
fine-tuning. We adapt VIM to eight benchmarks, including OKVQA, MM-Vet,
MathVista, MMMU, and probe diverse MLLMs in both the text-modality instruction
(TEM) setting and VIM setting. Notably, we observe a significant performance
disparity between the original TEM and VIM settings for open-source MLLMs,
indicating that open-source MLLMs face greater challenges when text instruction
is presented solely in image form. To address this issue, we train v-MLLM, a
generalizable model that is capable to conduct robust instruction following in
both text-modality and visual-modality instructions. | cs.CV | Text as Images: Can Multimodal Large Language Models Follow Printed Instructions in Pixels? |
2023-11-29T13:51:04Z | http://arxiv.org/pdf/2311.17633v1 | http://arxiv.org/abs/2311.17633v1 | Tong Xiao, Jingbo Zhu | 2023-11-29T13:51:04Z | cs.CL, cs.AI, cs.LG | Transformers have dominated empirical machine learning models of natural
language processing. In this paper, we introduce basic concepts of Transformers
and present key techniques that form the recent advances of these models. This
includes a description of the standard Transformer architecture, a series of
model refinements, and common applications. Given that Transformers and related
deep learning techniques might be evolving in ways we have never seen, we
cannot dive into all the model details or cover all the technical areas.
Instead, we focus on just those concepts that are helpful for gaining a good
understanding of Transformers and their variants. We also summarize the key
ideas that impact this field, thereby yielding some insights into the strengths
and limitations of these models. | cs.CL | Introduction to Transformers: an NLP Perspective |
2023-11-29T12:41:55Z | http://arxiv.org/pdf/2311.17593v1 | http://arxiv.org/abs/2311.17593v1 | Rudra P. K. Poudel, Harit Pandya, Chao Zhang, Roberto Cipolla | 2023-11-29T12:41:55Z | cs.LG, cs.AI, cs.CL, cs.CV, cs.RO | Recent advances in deep reinforcement learning have showcased its potential
in tackling complex tasks. However, experiments on visual control tasks have
revealed that state-of-the-art reinforcement learning models struggle with
out-of-distribution generalization. Conversely, expressing higher-level
concepts and global contexts is relatively easy using language.
Building upon recent success of the large language models, our main objective
is to improve the state abstraction technique in reinforcement learning by
leveraging language for robust action selection. Specifically, we focus on
learning language-grounded visual features to enhance the world model learning,
a model-based reinforcement learning technique.
To enforce our hypothesis explicitly, we mask out the bounding boxes of a few
objects in the image observation and provide the text prompt as descriptions
for these masked objects. Subsequently, we predict the masked objects along
with the surrounding regions as pixel reconstruction, similar to the
transformer-based masked autoencoder approach.
Our proposed LanGWM: Language Grounded World Model achieves state-of-the-art
performance in out-of-distribution test at the 100K interaction steps
benchmarks of iGibson point navigation tasks. Furthermore, our proposed
technique of explicit language-grounded visual representation learning has the
potential to improve models for human-robot interaction because our extracted
visual features are language grounded. | cs.LG | LanGWM: Language Grounded World Model |
2023-11-29T10:38:16Z | http://arxiv.org/pdf/2311.17514v1 | http://arxiv.org/abs/2311.17514v1 | Swaroop Nath, Harshad Khadilkar, Pushpak Bhattacharyya | 2023-11-29T10:38:16Z | cs.CL, cs.AI | Query-focused Summarization (QfS) deals with systems that generate summaries
from document(s) based on a query. Motivated by the insight that Reinforcement
Learning (RL) provides a generalization to Supervised Learning (SL) for Natural
Language Generation, and thereby performs better (empirically) than SL, we use
an RL-based approach for this task of QfS. Additionally, we also resolve the
conflict of employing RL in Transformers with Teacher Forcing. We develop
multiple Policy Gradient networks, trained on various reward signals: ROUGE,
BLEU, and Semantic Similarity, which lead to a 10-point improvement over the
State-of-the-Art approach on the ROUGE-L metric for a benchmark dataset (ELI5).
We also show performance of our approach in zero-shot setting for another
benchmark dataset (DebatePedia) -- our approach leads to results comparable to
baselines, which were specifically trained on DebatePedia. To aid the RL
training, we propose a better semantic similarity reward, enabled by a novel
Passage Embedding scheme developed using Cluster Hypothesis. Lastly, we
contribute a gold-standard test dataset to further research in QfS and
Long-form Question Answering (LfQA). | cs.CL | Reinforcement Replaces Supervision: Query focused Summarization using Deep Reinforcement Learning |
2023-11-29T10:24:50Z | http://arxiv.org/pdf/2311.17502v1 | http://arxiv.org/abs/2311.17502v1 | Xinghang Hu | 2023-11-29T10:24:50Z | cs.CL | Community Question Answering (CQA) becomes increasingly prevalent in recent
years. However, there are a large number of answers, which is difficult for
users to select the relevant answers. Therefore, answer selection is a very
significant subtask of CQA. In this paper, we first propose the Question-Answer
cross attention networks (QAN) with pre-trained models for answer selection and
utilize large language model (LLM) to perform answer selection with knowledge
augmentation. Specifically, we apply the BERT model as the encoder layer to do
pre-training for question subjects, question bodies and answers, respectively,
then the cross attention mechanism selects the most relevant answer for
different questions. Experiments show that the QAN model achieves
state-of-the-art performance on two datasets, SemEval2015 and SemEval2017.
Moreover, we use the LLM to generate external knowledge from questions and
correct answers to achieve knowledge augmentation for the answer selection task
by LLM, while optimizing the prompt of LLM in different aspects. The results
show that the introduction of external knowledge can improve the correct answer
selection rate of LLM on datasets SemEval2015 and SemEval2017. Meanwhile, LLM
can also select the correct answer on more questions by optimized prompt. | cs.CL | Enhancing Answer Selection in Community Question Answering with Pre-trained and Large Language Models |
2023-11-29T10:05:58Z | http://arxiv.org/pdf/2312.03743v1 | http://arxiv.org/abs/2312.03743v1 | Alwan Wirawan, Hasan Dwi Cahyono, Winarno | 2023-11-29T10:05:58Z | cs.CL, cs.LG | Instagram, a social media platform, has in the vicinity of 2 billion active
users in 2023. The platform allows users to post photos and videos with one
another. However, cyberbullying remains a significant problem for about 50% of
young Indonesians. To address this issue, sentiment analysis for comment
filtering uses a Support Vector Machine (SVM) and Easy Data Augmentation (EDA).
EDA will augment the dataset, enabling robust prediction and analysis of
cyberbullying by introducing more variation. Based on the tests, SVM
combination with EDA results in a 2.52% increase in the k-Fold Cross Validation
score. Our proposed approach shows an improved accuracy of 92.5%, 2.5% higher
than that of the existing state-of-the-art method. To maintain the
reproducibility and replicability of this research, the source code can be
accessed at uns.id/eda_svm. | cs.CL | Easy Data Augmentation in Sentiment Analysis of Cyberbullying |
2023-11-29T10:01:48Z | http://arxiv.org/pdf/2311.17492v2 | http://arxiv.org/abs/2311.17492v2 | Jean Seo, Sungjoo Byun, Minha Kang, Sangah Lee | 2024-01-12T14:18:03Z | cs.CL | The Manchu language, with its roots in the historical Manchurian region of
Northeast China, is now facing a critical threat of extinction, as there are
very few speakers left. In our efforts to safeguard the Manchu language, we
introduce Mergen, the first-ever attempt at a Manchu-Korean Machine Translation
(MT) model. To develop this model, we utilize valuable resources such as the
Manwen Laodang(a historical book) and a Manchu-Korean dictionary. Due to the
scarcity of a Manchu-Korean parallel dataset, we expand our data by employing
word replacement guided by GloVe embeddings, trained on both monolingual and
parallel texts. Our approach is built around an encoder-decoder neural machine
translation model, incorporating a bi-directional Gated Recurrent Unit (GRU)
layer. The experiments have yielded promising results, showcasing a significant
enhancement in Manchu-Korean translation, with a remarkable 20-30 point
increase in the BLEU score. | cs.CL | Mergen: The First Manchu-Korean Machine Translation Model Trained on Augmented Data |
2023-11-29T09:48:34Z | http://arxiv.org/pdf/2311.17487v1 | http://arxiv.org/abs/2311.17487v1 | Yen-Ting Lin, Yun-Nung Chen | 2023-11-29T09:48:34Z | cs.CL, cs.AI | In the realm of language models, the nuanced linguistic and cultural
intricacies of Traditional Chinese, as spoken in Taiwan, have been largely
overlooked. This paper introduces Taiwan LLM, a pioneering Large Language Model
that specifically caters to the Traditional Chinese language, with a focus on
the variant used in Taiwan. Leveraging a comprehensive pretraining corpus and
instruction-finetuning datasets, we have developed a model that not only
understands the complexities of Traditional Chinese but also embodies the
cultural context of Taiwan. Taiwan LLM represents the first of its kind, a
model that is not only linguistically accurate but also culturally resonant
with its user base. Our evaluations demonstrate that Taiwan LLM achieves
superior performance in understanding and generating Traditional Chinese text,
outperforming existing models that are predominantly trained on Simplified
Chinese or English. The open-source release of Taiwan LLM invites collaboration
and further innovation, ensuring that the linguistic diversity of Chinese
speakers is embraced and well-served. The model, datasets, and further
resources are made publicly available to foster ongoing research and
development in this field. | cs.CL | Taiwan LLM: Bridging the Linguistic Divide with a Culturally Aligned Language Model |
2023-11-29T08:29:54Z | http://arxiv.org/pdf/2311.17438v4 | http://arxiv.org/abs/2311.17438v4 | Yinya Huang, Ruixin Hong, Hongming Zhang, Wei Shao, Zhicheng Yang, Dong Yu, Changshui Zhang, Xiaodan Liang, Linqi Song | 2024-06-07T04:27:54Z | cs.CL, cs.AI | In this study, we delve into the realm of counterfactual reasoning
capabilities of large language models (LLMs). Our primary objective is to
cultivate the counterfactual thought processes within LLMs and rigorously
assess these processes for their validity. Specifically, we introduce a novel
task, Counterfactual Logical Modification (CLOMO), and a high-quality
human-annotated benchmark. In this task, LLMs must adeptly alter a given
argumentative text to uphold a predetermined logical relationship. To
effectively evaluate a generation model's counterfactual capabilities, we
propose an innovative evaluation metric, the decomposed Self-Evaluation Score
(SES) to directly evaluate the natural language output of LLMs instead of
modeling the task as a multiple-choice problem. Analysis shows that the
proposed automatic metric aligns well with human preference. Our experimental
results show that while LLMs demonstrate a notable capacity for logical
counterfactual thinking, there remains a discernible gap between their current
abilities and human performance. Code and data are available at
https://github.com/Eleanor-H/CLOMO. | cs.CL | CLOMO: Counterfactual Logical Modification with Large Language Models |
2023-11-29T08:12:09Z | http://arxiv.org/pdf/2311.17429v1 | http://arxiv.org/abs/2311.17429v1 | Zihao Tan, Qingliang Chen, Yongjian Huang, Chen Liang | 2023-11-29T08:12:09Z | cs.CL, cs.AI | Prompt-based learning has been widely applied in many low-resource NLP tasks
such as few-shot scenarios. However, this paradigm has been shown to be
vulnerable to backdoor attacks. Most of the existing attack methods focus on
inserting manually predefined templates as triggers in the pre-training phase
to train the victim model and utilize the same triggers in the downstream task
to perform inference, which tends to ignore the transferability and
stealthiness of the templates. In this work, we propose a novel approach of
TARGET (Template-trAnsfeRable backdoor attack aGainst prompt-basEd NLP models
via GPT4), which is a data-independent attack method. Specifically, we first
utilize GPT4 to reformulate manual templates to generate tone-strong and normal
templates, and the former are injected into the model as a backdoor trigger in
the pre-training phase. Then, we not only directly employ the above templates
in the downstream task, but also use GPT4 to generate templates with similar
tone to the above templates to carry out transferable attacks. Finally we have
conducted extensive experiments on five NLP datasets and three BERT series
models, with experimental results justifying that our TARGET method has better
attack performance and stealthiness compared to the two-external baseline
methods on direct attacks, and in addition achieves satisfactory attack
capability in the unseen tone-similar templates. | cs.CL | TARGET: Template-Transferable Backdoor Attack Against Prompt-based NLP Models via GPT4 |
2023-11-29T07:27:15Z | http://arxiv.org/pdf/2312.02181v1 | http://arxiv.org/abs/2312.02181v1 | Zeteng Lin | 2023-11-29T07:27:15Z | cs.CL, cs.AI, cs.LG, econ.GN, q-fin.EC | With the rapid development of artificial intelligence and breakthroughs in
machine learning and natural language processing, intelligent
question-answering robots have become widely used in government affairs. This
paper conducts a horizontal comparison between Guangdong Province's government
chatbots, ChatGPT, and Wenxin Ernie, two large language models, to analyze the
strengths and weaknesses of existing government chatbots and AIGC technology.
The study finds significant differences between government chatbots and large
language models. China's government chatbots are still in an exploratory stage
and have a gap to close to achieve "intelligence." To explore the future
direction of government chatbots more deeply, this research proposes targeted
optimization paths to help generative AI be effectively applied in government
chatbot conversations. | cs.CL | How Generative-AI can be Effectively used in Government Chatbots |
2023-11-29T07:15:34Z | http://arxiv.org/pdf/2311.17404v1 | http://arxiv.org/abs/2311.17404v1 | Shicheng Li, Lei Li, Shuhuai Ren, Yuanxin Liu, Yi Liu, Rundong Gao, Xu Sun, Lu Hou | 2023-11-29T07:15:34Z | cs.CV, cs.AI, cs.CL | The ability to perceive how objects change over time is a crucial ingredient
in human intelligence. However, current benchmarks cannot faithfully reflect
the temporal understanding abilities of video-language models (VidLMs) due to
the existence of static visual shortcuts. To remedy this issue, we present
VITATECS, a diagnostic VIdeo-Text dAtaset for the evaluation of TEmporal
Concept underStanding. Specifically, we first introduce a fine-grained taxonomy
of temporal concepts in natural language in order to diagnose the capability of
VidLMs to comprehend different temporal aspects. Furthermore, to disentangle
the correlation between static and temporal information, we generate
counterfactual video descriptions that differ from the original one only in the
specified temporal aspect. We employ a semi-automatic data collection framework
using large language models and human-in-the-loop annotation to obtain
high-quality counterfactual descriptions efficiently. Evaluation of
representative video-language understanding models confirms their deficiency in
temporal understanding, revealing the need for greater emphasis on the temporal
elements in video-language research. | cs.CV | VITATECS: A Diagnostic Dataset for Temporal Concept Understanding of Video-Language Models |
2023-11-29T07:09:13Z | http://arxiv.org/pdf/2311.17400v2 | http://arxiv.org/abs/2311.17400v2 | Lujia Shen, Yuwen Pu, Shouling Ji, Changjiang Li, Xuhong Zhang, Chunpeng Ge, Ting Wang | 2023-11-30T02:08:24Z | cs.CL, cs.CR, cs.LG | Transformer-based models, such as BERT and GPT, have been widely adopted in
natural language processing (NLP) due to their exceptional performance.
However, recent studies show their vulnerability to textual adversarial attacks
where the model's output can be misled by intentionally manipulating the text
inputs. Despite various methods that have been proposed to enhance the model's
robustness and mitigate this vulnerability, many require heavy consumption
resources (e.g., adversarial training) or only provide limited protection
(e.g., defensive dropout). In this paper, we propose a novel method called
dynamic attention, tailored for the transformer architecture, to enhance the
inherent robustness of the model itself against various adversarial attacks.
Our method requires no downstream task knowledge and does not incur additional
costs. The proposed dynamic attention consists of two modules: (I) attention
rectification, which masks or weakens the attention value of the chosen tokens,
and (ii) dynamic modeling, which dynamically builds the set of candidate
tokens. Extensive experiments demonstrate that dynamic attention significantly
mitigates the impact of adversarial attacks, improving up to 33\% better
performance than previous methods against widely-used adversarial attacks. The
model-level design of dynamic attention enables it to be easily combined with
other defense methods (e.g., adversarial training) to further enhance the
model's robustness. Furthermore, we demonstrate that dynamic attention
preserves the state-of-the-art robustness space of the original model compared
to other dynamic modeling methods. | cs.CL | Improving the Robustness of Transformer-based Large Language Models with Dynamic Attention |
2023-11-29T06:42:36Z | http://arxiv.org/pdf/2311.17391v1 | http://arxiv.org/abs/2311.17391v1 | Jiaxin Wen, Pei Ke, Hao Sun, Zhexin Zhang, Chengfei Li, Jinfeng Bai, Minlie Huang | 2023-11-29T06:42:36Z | cs.CL | The open-endedness of large language models (LLMs) combined with their
impressive capabilities may lead to new safety issues when being exploited for
malicious use. While recent studies primarily focus on probing toxic outputs
that can be easily detected with existing toxicity classifiers, we show that
LLMs can generate diverse implicit toxic outputs that are exceptionally
difficult to detect via simply zero-shot prompting. Moreover, we propose a
reinforcement learning (RL) based attacking method to further induce the
implicit toxicity in LLMs. Specifically, we optimize the language model with a
reward that prefers implicit toxic outputs to explicit toxic and non-toxic
ones. Experiments on five widely-adopted toxicity classifiers demonstrate that
the attack success rate can be significantly improved through RL fine-tuning.
For instance, the RL-finetuned LLaMA-13B model achieves an attack success rate
of 90.04% on BAD and 62.85% on Davinci003. Our findings suggest that LLMs pose
a significant threat in generating undetectable implicit toxic outputs. We
further show that fine-tuning toxicity classifiers on the annotated examples
from our attacking method can effectively enhance their ability to detect
LLM-generated implicit toxic language. The code is publicly available at
https://github.com/thu-coai/Implicit-Toxicity. | cs.CL | Unveiling the Implicit Toxicity in Large Language Models |
2023-11-29T06:02:16Z | http://arxiv.org/pdf/2311.17376v1 | http://arxiv.org/abs/2311.17376v1 | Taha Aksu, Devamanyu Hazarika, Shikib Mehri, Seokhwan Kim, Dilek Hakkani-Tür, Yang Liu, Mahdi Namazifar | 2023-11-29T06:02:16Z | cs.CL | Instruction-based multitasking has played a critical role in the success of
large language models (LLMs) in multi-turn dialog applications. While publicly
available LLMs have shown promising performance, when exposed to complex
instructions with multiple constraints, they lag against state-of-the-art
models like ChatGPT. In this work, we hypothesize that the availability of
large-scale complex demonstrations is crucial in bridging this gap. Focusing on
dialog applications, we propose a novel framework, CESAR, that unifies a large
number of dialog tasks in the same format and allows programmatic induction of
complex instructions without any manual effort.
We apply CESAR on InstructDial, a benchmark for instruction-based dialog
tasks. We further enhance InstructDial with new datasets and tasks and utilize
CESAR to induce complex tasks with compositional instructions. This results in
a new benchmark called InstructDial++, which includes 63 datasets with 86 basic
tasks and 68 composite tasks. Through rigorous experiments, we demonstrate the
scalability of CESAR in providing rich instructions. Models trained on
InstructDial++ can follow compositional prompts, such as prompts that ask for
multiple stylistic constraints. | cs.CL | CESAR: Automatic Induction of Compositional Instructions for Multi-turn Dialogs |
2023-11-29T05:54:41Z | http://arxiv.org/pdf/2311.17371v2 | http://arxiv.org/abs/2311.17371v2 | Andries Smit, Paul Duckworth, Nathan Grinsztajn, Thomas D. Barrett, Arnu Pretorius | 2024-03-14T10:56:50Z | cs.CL, cs.AI | Recent advancements in large language models (LLMs) underscore their
potential for responding to inquiries in various domains. However, ensuring
that generative agents provide accurate and reliable answers remains an ongoing
challenge. In this context, multi-agent debate (MAD) has emerged as a promising
strategy for enhancing the truthfulness of LLMs. We benchmark a range of
debating and prompting strategies to explore the trade-offs between cost, time,
and accuracy. Importantly, we find that multi-agent debating systems, in their
current form, do not reliably outperform other proposed prompting strategies,
such as self-consistency and ensembling using multiple reasoning paths.
However, when performing hyperparameter tuning, several MAD systems, such as
Multi-Persona, perform better. This suggests that MAD protocols might not be
inherently worse than other approaches, but that they are more sensitive to
different hyperparameter settings and difficult to optimize. We build on these
results to offer insights into improving debating strategies, such as adjusting
agent agreement levels, which can significantly enhance performance and even
surpass all other non-debate protocols we evaluated. We provide an open-source
repository to the community with several state-of-the-art protocols together
with evaluation scripts to benchmark across popular research datasets. | cs.CL | Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs |
2023-11-29T05:04:52Z | http://arxiv.org/pdf/2311.17355v1 | http://arxiv.org/abs/2311.17355v1 | Han Cao, Lingwei Wei, Mengyang Chen, Wei Zhou, Songlin Hu | 2023-11-29T05:04:52Z | cs.CL | Recently, Large Language Models (LLMs) have drawn significant attention due
to their outstanding reasoning capabilities and extensive knowledge repository,
positioning them as superior in handling various natural language processing
tasks compared to other language models. In this paper, we present a
preliminary investigation into the potential of LLMs in fact-checking. This
study aims to comprehensively evaluate various LLMs in tackling specific
fact-checking subtasks, systematically evaluating their capabilities, and
conducting a comparative analysis of their performance against pre-trained and
state-of-the-art low-parameter models. Experiments demonstrate that LLMs
achieve competitive performance compared to other small models in most
scenarios. However, they encounter challenges in effectively handling Chinese
fact verification and the entirety of the fact-checking pipeline due to
language inconsistencies and hallucinations. These findings underscore the need
for further exploration and research to enhance the proficiency of LLMs as
reliable fact-checkers, unveiling the potential capability of LLMs and the
possible challenges in fact-checking tasks. | cs.CL | Are Large Language Models Good Fact Checkers: A Preliminary Study |
2023-11-29T04:32:19Z | http://arxiv.org/pdf/2312.03742v1 | http://arxiv.org/abs/2312.03742v1 | Angeela Acharya, Sulabh Shrestha, Anyi Chen, Joseph Conte, Sanja Avramovic, Siddhartha Sikdar, Antonios Anastasopoulos, Sanmay Das | 2023-11-29T04:32:19Z | cs.CL, cs.LG | The utilization of Electronic Health Records (EHRs) for clinical risk
prediction is on the rise. However, strict privacy regulations limit access to
comprehensive health records, making it challenging to apply standard machine
learning algorithms in practical real-world scenarios. Previous research has
addressed this data limitation by incorporating medical ontologies and
employing transfer learning methods. In this study, we investigate the
potential of leveraging language models (LMs) as a means to incorporate
supplementary domain knowledge for improving the performance of various
EHR-based risk prediction tasks. Unlike applying LMs to unstructured EHR data
such as clinical notes, this study focuses on using textual descriptions within
structured EHR to make predictions exclusively based on that information. We
extensively compare against previous approaches across various data types and
sizes. We find that employing LMs to represent structured EHRs, such as
diagnostic histories, leads to improved or at least comparable performance in
diverse risk prediction tasks. Furthermore, LM-based approaches offer numerous
advantages, including few-shot learning, the capability to handle previously
unseen medical concepts, and adaptability to various medical vocabularies.
Nevertheless, we underscore, through various experiments, the importance of
being cautious when employing such models, as concerns regarding the
reliability of LMs persist. | cs.CL | Clinical Risk Prediction Using Language Models: Benefits And Considerations |
2023-11-29T04:31:35Z | http://arxiv.org/pdf/2311.17352v1 | http://arxiv.org/abs/2311.17352v1 | Haoyu He, Zizheng Pan, Jing Liu, Jianfei Cai, Bohan Zhuang | 2023-11-29T04:31:35Z | cs.LG, cs.CL, cs.CV | The paradigm of pre-training and fine-tuning has laid the foundation for
deploying deep learning models. However, most fine-tuning methods are designed
to meet a specific resource budget. Recently, considering diverse deployment
scenarios with various resource budgets, stitchable neural network (SN-Net) is
introduced to quickly obtain numerous new networks (stitches) from the
pre-trained models (anchors) in a model family via model stitching. Although
promising, SN-Net confronts new challenges when adapting it to new target
domains, including huge memory and storage requirements and a long and
sub-optimal multistage adaptation process. In this work, we present a novel
framework, Efficient Stitchable Task Adaptation (ESTA), to efficiently produce
a palette of fine-tuned models that adhere to diverse resource constraints.
Specifically, we first tailor parameter-efficient fine-tuning to share low-rank
updates among the stitches while maintaining independent bias terms. In this
way, we largely reduce fine-tuning memory burdens and mitigate the interference
among stitches that arises in task adaptation. Furthermore, we streamline a
simple yet effective one-stage deployment pipeline, which estimates the
important stitches to deploy with training-time gradient statistics. By
assigning higher sampling probabilities to important stitches, we also get a
boosted Pareto frontier. Extensive experiments on 25 downstream visual
recognition tasks demonstrate that our ESTA is capable of generating stitches
with smooth accuracy-efficiency trade-offs and surpasses the direct SN-Net
adaptation by remarkable margins with significantly lower training time and
fewer trainable parameters. Furthermore, we demonstrate the flexibility and
scalability of our ESTA framework by stitching LLMs from LLaMA family,
obtaining chatbot stitches of assorted sizes. | cs.LG | Efficient Stitchable Task Adaptation |
2023-11-29T04:25:15Z | http://arxiv.org/pdf/2311.17351v1 | http://arxiv.org/abs/2311.17351v1 | Yuebing Liang, Yichao Liu, Xiaohan Wang, Zhan Zhao | 2023-11-29T04:25:15Z | cs.AI, cs.CL | Public events, such as concerts and sports games, can be major attractors for
large crowds, leading to irregular surges in travel demand. Accurate human
mobility prediction for public events is thus crucial for event planning as
well as traffic or crowd management. While rich textual descriptions about
public events are commonly available from online sources, it is challenging to
encode such information in statistical or machine learning models. Existing
methods are generally limited in incorporating textual information, handling
data sparsity, or providing rationales for their predictions. To address these
challenges, we introduce a framework for human mobility prediction under public
events (LLM-MPE) based on Large Language Models (LLMs), leveraging their
unprecedented ability to process textual data, learn from minimal examples, and
generate human-readable explanations. Specifically, LLM-MPE first transforms
raw, unstructured event descriptions from online sources into a standardized
format, and then segments historical mobility data into regular and
event-related components. A prompting strategy is designed to direct LLMs in
making and rationalizing demand predictions considering historical mobility and
event features. A case study is conducted for Barclays Center in New York City,
based on publicly available event information and taxi trip data. Results show
that LLM-MPE surpasses traditional models, particularly on event days, with
textual data significantly enhancing its accuracy. Furthermore, LLM-MPE offers
interpretable insights into its predictions. Despite the great potential of
LLMs, we also identify key challenges including misinformation and high costs
that remain barriers to their broader adoption in large-scale human mobility
analysis. | cs.AI | Exploring Large Language Models for Human Mobility Prediction under Public Events |
2023-11-29T03:42:16Z | http://arxiv.org/pdf/2311.17946v1 | http://arxiv.org/abs/2311.17946v1 | Jiao Sun, Deqing Fu, Yushi Hu, Su Wang, Royi Rassin, Da-Cheng Juan, Dana Alon, Charles Herrmann, Sjoerd van Steenkiste, Ranjay Krishna, Cyrus Rashtchian | 2023-11-29T03:42:16Z | cs.CV, cs.AI, cs.CL | Despite their wide-spread success, Text-to-Image models (T2I) still struggle
to produce images that are both aesthetically pleasing and faithful to the
user's input text. We introduce DreamSync, a model-agnostic training algorithm
by design that improves T2I models to be faithful to the text input. DreamSync
builds off a recent insight from TIFA's evaluation framework -- that large
vision-language models (VLMs) can effectively identify the fine-grained
discrepancies between generated images and the text inputs. DreamSync uses this
insight to train T2I models without any labeled data; it improves T2I models
using its own generations. First, it prompts the model to generate several
candidate images for a given input text. Then, it uses two VLMs to select the
best generation: a Visual Question Answering model that measures the alignment
of generated images to the text, and another that measures the generation's
aesthetic quality. After selection, we use LoRA to iteratively finetune the T2I
model to guide its generation towards the selected best generations. DreamSync
does not need any additional human annotation. model architecture changes, or
reinforcement learning. Despite its simplicity, DreamSync improves both the
semantic alignment and aesthetic appeal of two diffusion-based T2I models,
evidenced by multiple benchmarks (+1.7% on TIFA, +2.9% on DSG1K, +3.4% on VILA
aesthetic) and human evaluation. | cs.CV | DreamSync: Aligning Text-to-Image Generation with Image Understanding Feedback |
2023-11-29T03:07:00Z | http://arxiv.org/pdf/2311.17330v2 | http://arxiv.org/abs/2311.17330v2 | Karthik Soman, Peter W Rose, John H Morris, Rabia E Akbas, Brett Smith, Braian Peetoom, Catalina Villouta-Reyes, Gabriel Cerono, Yongmei Shi, Angela Rizk-Jackson, Sharat Israni, Charlotte A Nelson, Sui Huang, Sergio E Baranzini | 2024-05-13T22:48:04Z | cs.CL | Large Language Models (LLMs) are being adopted at an unprecedented rate, yet
still face challenges in knowledge-intensive domains like biomedicine.
Solutions such as pre-training and domain-specific fine-tuning add substantial
computational overhead, requiring further domain expertise. Here, we introduce
a token-optimized and robust Knowledge Graph-based Retrieval Augmented
Generation (KG-RAG) framework by leveraging a massive biomedical KG (SPOKE)
with LLMs such as Llama-2-13b, GPT-3.5-Turbo and GPT-4, to generate meaningful
biomedical text rooted in established knowledge. Compared to the existing RAG
technique for Knowledge Graphs, the proposed method utilizes minimal graph
schema for context extraction and uses embedding methods for context pruning.
This optimization in context extraction results in more than 50% reduction in
token consumption without compromising the accuracy, making a cost-effective
and robust RAG implementation on proprietary LLMs. KG-RAG consistently enhanced
the performance of LLMs across diverse biomedical prompts by generating
responses rooted in established knowledge, accompanied by accurate provenance
and statistical evidence (if available) to substantiate the claims. Further
benchmarking on human curated datasets, such as biomedical true/false and
multiple-choice questions (MCQ), showed a remarkable 71% boost in the
performance of the Llama-2 model on the challenging MCQ dataset, demonstrating
the framework's capacity to empower open-source models with fewer parameters
for domain specific questions. Furthermore, KG-RAG enhanced the performance of
proprietary GPT models, such as GPT-3.5 and GPT-4. In summary, the proposed
framework combines explicit and implicit knowledge of KG and LLM in a token
optimized fashion, thus enhancing the adaptability of general-purpose LLMs to
tackle domain-specific questions in a cost-effective fashion. | cs.CL | Biomedical knowledge graph-optimized prompt generation for large language models |
2023-11-29T02:07:09Z | http://arxiv.org/pdf/2311.17311v1 | http://arxiv.org/abs/2311.17311v1 | Xinyun Chen, Renat Aksitov, Uri Alon, Jie Ren, Kefan Xiao, Pengcheng Yin, Sushant Prakash, Charles Sutton, Xuezhi Wang, Denny Zhou | 2023-11-29T02:07:09Z | cs.CL, cs.AI | Self-consistency with chain-of-thought prompting (CoT) has demonstrated
remarkable performance gains on various challenging tasks, by utilizing
multiple reasoning paths sampled from large language models (LLMs). However,
self-consistency relies on the answer extraction process to aggregate multiple
solutions, which is not applicable to free-form answers. In this work, we
propose Universal Self-Consistency (USC), which leverages LLMs themselves to
select the most consistent answer among multiple candidates. We evaluate USC on
a variety of benchmarks, including mathematical reasoning, code generation,
long-context summarization, and open-ended question answering. On open-ended
generation tasks where the original self-consistency method is not applicable,
USC effectively utilizes multiple samples and improves the performance. For
mathematical reasoning, USC matches the standard self-consistency performance
without requiring the answer formats to be similar. Finally, without access to
execution results, USC also matches the execution-based voting performance on
code generation. | cs.CL | Universal Self-Consistency for Large Language Model Generation |
2023-11-29T01:59:38Z | http://arxiv.org/pdf/2311.17307v1 | http://arxiv.org/abs/2311.17307v1 | Hua Pu, Jiacong Mi, Shan Lu, Jieyue He | 2023-11-29T01:59:38Z | cs.CL, cs.AI | Traditional Chinese medicine (TCM) prescription is the most critical form of
TCM treatment, and uncovering the complex nonlinear relationship between
symptoms and TCM is of great significance for clinical practice and assisting
physicians in diagnosis and treatment. Although there have been some studies on
TCM prescription generation, these studies consider a single factor and
directly model the symptom-prescription generation problem mainly based on
symptom descriptions, lacking guidance from TCM knowledge. To this end, we
propose a RoBERTa and Knowledge Enhancement model for Prescription Generation
of Traditional Chinese Medicine (RoKEPG). RoKEPG is firstly pre-trained by our
constructed TCM corpus, followed by fine-tuning the pre-trained model, and the
model is guided to generate TCM prescriptions by introducing four classes of
knowledge of TCM through the attention mask matrix. Experimental results on the
publicly available TCM prescription dataset show that RoKEPG improves the F1
metric by about 2% over the baseline model with the best results. | cs.CL | RoKEPG: RoBERTa and Knowledge Enhancement for Prescription Generation of Traditional Chinese Medicine |
2023-11-29T01:19:02Z | http://arxiv.org/pdf/2311.17301v1 | http://arxiv.org/abs/2311.17301v1 | Sofia Serrano, Zander Brumbaugh, Noah A. Smith | 2023-11-29T01:19:02Z | cs.CL, cs.AI, cs.CY, cs.LG | Given the growing importance of AI literacy, we decided to write this
tutorial to help narrow the gap between the discourse among those who study
language models -- the core technology underlying ChatGPT and similar products
-- and those who are intrigued and want to learn more about them. In short, we
believe the perspective of researchers and educators can add some clarity to
the public's understanding of the technologies beyond what's currently
available, which tends to be either extremely technical or promotional material
generated about products by their purveyors.
Our approach teases apart the concept of a language model from products built
on them, from the behaviors attributed to or desired from those products, and
from claims about similarity to human cognition. As a starting point, we (1)
offer a scientific viewpoint that focuses on questions amenable to study
through experimentation; (2) situate language models as they are today in the
context of the research that led to their development; and (3) describe the
boundaries of what is known about the models at this writing. | cs.CL | Language Models: A Guide for the Perplexed |
2023-11-29T00:45:23Z | http://arxiv.org/pdf/2311.17295v1 | http://arxiv.org/abs/2311.17295v1 | Meriem Boubdir, Edward Kim, Beyza Ermis, Sara Hooker, Marzieh Fadaee | 2023-11-29T00:45:23Z | cs.CL, cs.AI | In Natural Language Processing (NLP), the Elo rating system, originally
designed for ranking players in dynamic games such as chess, is increasingly
being used to evaluate Large Language Models (LLMs) through "A vs B" paired
comparisons. However, while popular, the system's suitability for assessing
entities with constant skill levels, such as LLMs, remains relatively
unexplored. We study two fundamental axioms that evaluation methods should
adhere to: reliability and transitivity. We conduct extensive evaluation of Elo
behaviour, illustrating that individual Elo computations exhibit volatility and
delving into the impact of varying the Elo rating system's hyperparameters. We
show that these axioms are not always satisfied raising questions about the
reliability of current comparative evaluations of LLMs. If the current use of
Elo scores is intended to substitute the costly head-to-head comparison of
LLMs, it is crucial to ensure the ranking is as robust as possible. Guided by
the axioms, our findings offer concrete guidelines for enhancing the
reliability of LLM evaluation methods, suggesting a need for reassessment of
existing comparative approaches. | cs.CL | Elo Uncovered: Robustness and Best Practices in Language Model Evaluation |
2023-11-28T23:40:13Z | http://arxiv.org/pdf/2311.17280v4 | http://arxiv.org/abs/2311.17280v4 | Wang Zhu, Ishika Singh, Yuan Huang, Robin Jia, Jesse Thomason | 2023-12-23T06:12:37Z | cs.CL, cs.CV | Data augmentation via back-translation is common when pretraining
Vision-and-Language Navigation (VLN) models, even though the generated
instructions are noisy. But: does that noise matter? We find that nonsensical
or irrelevant language instructions during pretraining can have little effect
on downstream performance for both HAMT and VLN-BERT on R2R, and is still
better than only using clean, human data. To underscore these results, we
concoct an efficient augmentation method, Unigram + Object, which generates
nonsensical instructions that nonetheless improve downstream performance. Our
findings suggest that what matters for VLN R2R pretraining is the quantity of
visual trajectories, not the quality of instructions. | cs.CL | Does VLN Pretraining Work with Nonsensical or Irrelevant Instructions? |
2023-11-28T22:54:33Z | http://arxiv.org/pdf/2311.17264v1 | http://arxiv.org/abs/2311.17264v1 | Marina Zhang, Owen Vallis, Aysegul Bumin, Tanay Vakharia, Elie Bursztein | 2023-11-28T22:54:33Z | cs.CL | This paper introduces RETSim (Resilient and Efficient Text Similarity), a
lightweight, multilingual deep learning model trained to produce robust metric
embeddings for near-duplicate text retrieval, clustering, and dataset
deduplication tasks. We demonstrate that RETSim is significantly more robust
and accurate than MinHash and neural text embeddings, achieving new
state-of-the-art performance on dataset deduplication, adversarial text
retrieval benchmarks, and spam clustering tasks. We also introduce the W4NT3D
benchmark (Wiki-40B 4dversarial Near-T3xt Dataset) for evaluating multilingual,
near-duplicate text retrieval capabilities under adversarial settings. RETSim
and the W4NT3D benchmark are open-sourced under the MIT License at
https://github.com/google/unisim. | cs.CL | RETSim: Resilient and Efficient Text Similarity |
2023-11-28T21:15:24Z | http://arxiv.org/pdf/2311.17233v1 | http://arxiv.org/abs/2311.17233v1 | Lukas Wolf, Tiago Pimentel, Evelina Fedorenko, Ryan Cotterell, Alex Warstadt, Ethan Wilcox, Tamar Regev | 2023-11-28T21:15:24Z | cs.CL, cs.AI, cs.IT, cs.LG, math.IT | Prosody -- the suprasegmental component of speech, including pitch, loudness,
and tempo -- carries critical aspects of meaning. However, the relationship
between the information conveyed by prosody vs. by the words themselves remains
poorly understood. We use large language models (LLMs) to estimate how much
information is redundant between prosody and the words themselves. Using a
large spoken corpus of English audiobooks, we extract prosodic features aligned
to individual words and test how well they can be predicted from LLM
embeddings, compared to non-contextual word embeddings. We find a high degree
of redundancy between the information carried by the words and prosodic
information across several prosodic features, including intensity, duration,
pauses, and pitch contours. Furthermore, a word's prosodic information is
redundant with both the word itself and the context preceding as well as
following it. Still, we observe that prosodic features can not be fully
predicted from text, suggesting that prosody carries information above and
beyond the words. Along with this paper, we release a general-purpose data
processing pipeline for quantifying the relationship between linguistic
information and extra-linguistic features. | cs.CL | Quantifying the redundancy between prosody and text |
2023-11-28T20:59:49Z | http://arxiv.org/pdf/2311.17227v2 | http://arxiv.org/abs/2311.17227v2 | Wenyue Hua, Lizhou Fan, Lingyao Li, Kai Mei, Jianchao Ji, Yingqiang Ge, Libby Hemphill, Yongfeng Zhang | 2024-01-30T18:53:30Z | cs.AI, cs.CL, cs.CY | Can we avoid wars at the crossroads of history? This question has been
pursued by individuals, scholars, policymakers, and organizations throughout
human history. In this research, we attempt to answer the question based on the
recent advances of Artificial Intelligence (AI) and Large Language Models
(LLMs). We propose \textbf{WarAgent}, an LLM-powered multi-agent AI system, to
simulate the participating countries, their decisions, and the consequences, in
historical international conflicts, including the World War I (WWI), the World
War II (WWII), and the Warring States Period (WSP) in Ancient China. By
evaluating the simulation effectiveness, we examine the advancements and
limitations of cutting-edge AI systems' abilities in studying complex
collective human behaviors such as international conflicts under diverse
settings. In these simulations, the emergent interactions among agents also
offer a novel perspective for examining the triggers and conditions that lead
to war. Our findings offer data-driven and AI-augmented insights that can
redefine how we approach conflict resolution and peacekeeping strategies. The
implications stretch beyond historical analysis, offering a blueprint for using
AI to understand human history and possibly prevent future international
conflicts. Code and data are available at
\url{https://github.com/agiresearch/WarAgent}. | cs.AI | War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars |
2023-11-28T20:34:40Z | http://arxiv.org/pdf/2311.17213v3 | http://arxiv.org/abs/2311.17213v3 | Ali H. Dhanaliwala, Rikhiya Ghosh, Sanjeev Kumar Karn, Poikavila Ullaskrishnan, Oladimeji Farri, Dorin Comaniciu, Charles E. Kahn | 2024-04-09T13:46:46Z | cs.CL, eess.IV | Radiologists produce unstructured data that can be valuable for clinical care
when consumed by information systems. However, variability in style limits
usage. Study compares system using domain-adapted language model (RadLing) and
general-purpose LLM (GPT-4) in extracting relevant features from chest
radiology reports and standardizing them to common data elements (CDEs). Three
radiologists annotated a retrospective dataset of 1399 chest XR reports (900
training, 499 test) and mapped to 44 pre-selected relevant CDEs. GPT-4 system
was prompted with report, feature set, value set, and dynamic few-shots to
extract values and map to CDEs. Output key:value pairs were compared to
reference standard at both stages and an identical match was considered TP. F1
score for extraction was 97% for RadLing-based system and 78% for GPT-4 system.
F1 score for mapping was 98% for RadLing and 94% for GPT-4; difference was
statistically significant (P<.001). RadLing's domain-adapted embeddings were
better in feature extraction and its light-weight mapper had better f1 score in
CDE assignment. RadLing system also demonstrated higher capabilities in
differentiating between absent (99% vs 64%) and unspecified (99% vs 89%).
RadLing system's domain-adapted embeddings helped improve performance of GPT-4
system to 92% by giving more relevant few-shot prompts. RadLing system offers
operational advantages including local deployment and reduced runtime costs. | cs.CL | General-Purpose vs. Domain-Adapted Large Language Models for Extraction of Structured Data from Chest Radiology Reports |
2023-11-28T19:00:03Z | http://arxiv.org/pdf/2311.17154v1 | http://arxiv.org/abs/2311.17154v1 | Dang Nguyen, Chacha Chen, He He, Chenhao Tan | 2023-11-28T19:00:03Z | cs.CL, cs.AI, cs.CY, cs.LG | When pneumonia is not found on a chest X-ray, should the report describe this
negative observation or omit it? We argue that this question cannot be answered
from the X-ray alone and requires a pragmatic perspective, which captures the
communicative goal that radiology reports serve between radiologists and
patients. However, the standard image-to-text formulation for radiology report
generation fails to incorporate such pragmatic intents. Following this
pragmatic perspective, we demonstrate that the indication, which describes why
a patient comes for an X-ray, drives the mentions of negative observations and
introduce indications as additional input to report generation. With respect to
the output, we develop a framework to identify uninferable information from the
image as a source of model hallucinations, and limit them by cleaning
groundtruth reports. Finally, we use indications and cleaned groundtruth
reports to develop pragmatic models, and show that they outperform existing
methods not only in new pragmatics-inspired metrics (+4.3 Negative F1) but also
in standard metrics (+6.3 Positive F1 and +11.0 BLEU-2). | cs.CL | Pragmatic Radiology Report Generation |
2023-11-28T18:55:52Z | http://arxiv.org/pdf/2311.17136v1 | http://arxiv.org/abs/2311.17136v1 | Cong Wei, Yang Chen, Haonan Chen, Hexiang Hu, Ge Zhang, Jie Fu, Alan Ritter, Wenhu Chen | 2023-11-28T18:55:52Z | cs.CV, cs.AI, cs.CL, cs.IR | Existing information retrieval (IR) models often assume a homogeneous format,
limiting their applicability to diverse user needs, such as searching for
images with text descriptions, searching for a news article with a headline
image, or finding a similar photo with a query image. To approach such
different information-seeking demands, we introduce UniIR, a unified
instruction-guided multimodal retriever capable of handling eight distinct
retrieval tasks across modalities. UniIR, a single retrieval system jointly
trained on ten diverse multimodal-IR datasets, interprets user instructions to
execute various retrieval tasks, demonstrating robust performance across
existing datasets and zero-shot generalization to new tasks. Our experiments
highlight that multi-task training and instruction tuning are keys to UniIR's
generalization ability. Additionally, we construct the M-BEIR, a multimodal
retrieval benchmark with comprehensive results, to standardize the evaluation
of universal multimodal information retrieval. | cs.CV | UniIR: Training and Benchmarking Universal Multimodal Information Retrievers |
2023-11-28T18:55:42Z | http://arxiv.org/pdf/2311.17049v2 | http://arxiv.org/abs/2311.17049v2 | Pavan Kumar Anasosalu Vasu, Hadi Pouransari, Fartash Faghri, Raviteja Vemulapalli, Oncel Tuzel | 2024-04-01T13:06:06Z | cs.CV, cs.CL, cs.LG | Contrastive pretraining of image-text foundation models, such as CLIP,
demonstrated excellent zero-shot performance and improved robustness on a wide
range of downstream tasks. However, these models utilize large
transformer-based encoders with significant memory and latency overhead which
pose challenges for deployment on mobile devices. In this work, we introduce
MobileCLIP -- a new family of efficient image-text models optimized for runtime
performance along with a novel and efficient training approach, namely
multi-modal reinforced training. The proposed training approach leverages
knowledge transfer from an image captioning model and an ensemble of strong
CLIP encoders to improve the accuracy of efficient models. Our approach avoids
train-time compute overhead by storing the additional knowledge in a reinforced
dataset. MobileCLIP sets a new state-of-the-art latency-accuracy tradeoff for
zero-shot classification and retrieval tasks on several datasets. Our
MobileCLIP-S2 variant is 2.3$\times$ faster while more accurate compared to
previous best CLIP model based on ViT-B/16. We further demonstrate the
effectiveness of our multi-modal reinforced training by training a CLIP model
based on ViT-B/16 image backbone and achieving +2.9% average performance
improvement on 38 evaluation benchmarks compared to the previous best.
Moreover, we show that the proposed approach achieves 10$\times$-1000$\times$
improved learning efficiency when compared with non-reinforced CLIP training.
Code and models are available at https://github.com/apple/ml-mobileclip . | cs.CV | MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training |
2023-11-28T18:53:43Z | http://arxiv.org/pdf/2311.17043v1 | http://arxiv.org/abs/2311.17043v1 | Yanwei Li, Chengyao Wang, Jiaya Jia | 2023-11-28T18:53:43Z | cs.CV, cs.CL | In this work, we present a novel method to tackle the token generation
challenge in Vision Language Models (VLMs) for video and image understanding,
called LLaMA-VID. Current VLMs, while proficient in tasks like image captioning
and visual question answering, face computational burdens when processing long
videos due to the excessive visual tokens. LLaMA-VID addresses this issue by
representing each frame with two distinct tokens, namely context token and
content token. The context token encodes the overall image context based on
user input, whereas the content token encapsulates visual cues in each frame.
This dual-token strategy significantly reduces the overload of long videos
while preserving critical information. Generally, LLaMA-VID empowers existing
frameworks to support hour-long videos and pushes their upper limit with an
extra context token. It is proved to surpass previous methods on most of video-
or image-based benchmarks. Code is available
https://github.com/dvlab-research/LLaMA-VID}{https://github.com/dvlab-research/LLaMA-VID | cs.CV | LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models |
2023-11-28T18:53:06Z | http://arxiv.org/pdf/2311.17041v2 | http://arxiv.org/abs/2311.17041v2 | Keunwoo Peter Yu, Zheyuan Zhang, Fengyuan Hu, Joyce Chai | 2023-11-29T15:52:55Z | cs.CV, cs.AI, cs.CL | Recent advancements in text-only large language models (LLMs) have
highlighted the benefit of in-context learning for adapting to new tasks with a
few demonstrations. However, extending in-context learning to large
vision-language models (VLMs) using a huge amount of naturalistic
vision-language data has shown limited success, particularly for egocentric
videos, due to high data collection costs. We propose a novel training method
$\mathbb{E}$fficient $\mathbb{I}$n-context $\mathbb{L}$earning on
$\mathbb{E}$gocentric $\mathbb{V}$ideos ($\mathbb{EILEV}$), which elicits
in-context learning in VLMs for egocentric videos without requiring massive,
naturalistic egocentric video datasets. $\mathbb{EILEV}$ involves architectural
and training data adaptations to allow the model to process contexts
interleaved with video clips and narrations, sampling of in-context examples
with clusters of similar verbs and nouns, use of data with skewed marginal
distributions with a long tail of infrequent verbs and nouns, as well as
homonyms and synonyms. Our evaluations show that $\mathbb{EILEV}$-trained
models outperform larger VLMs trained on a huge amount of naturalistic data in
in-context learning. Furthermore, they can generalize to not only
out-of-distribution, but also novel, rare egocentric videos and texts via
in-context learning, demonstrating potential for applications requiring
cost-effective training, and rapid post-deployment adaptability. Our code and
demo are available at \url{https://github.com/yukw777/EILEV}. | cs.CV | Efficient In-Context Learning in Vision-Language Models for Egocentric Videos |
2023-11-28T18:47:03Z | http://arxiv.org/pdf/2311.17035v1 | http://arxiv.org/abs/2311.17035v1 | Milad Nasr, Nicholas Carlini, Jonathan Hayase, Matthew Jagielski, A. Feder Cooper, Daphne Ippolito, Christopher A. Choquette-Choo, Eric Wallace, Florian Tramèr, Katherine Lee | 2023-11-28T18:47:03Z | cs.LG, cs.CL, cs.CR | This paper studies extractable memorization: training data that an adversary
can efficiently extract by querying a machine learning model without prior
knowledge of the training dataset. We show an adversary can extract gigabytes
of training data from open-source language models like Pythia or GPT-Neo,
semi-open models like LLaMA or Falcon, and closed models like ChatGPT. Existing
techniques from the literature suffice to attack unaligned models; in order to
attack the aligned ChatGPT, we develop a new divergence attack that causes the
model to diverge from its chatbot-style generations and emit training data at a
rate 150x higher than when behaving properly. Our methods show practical
attacks can recover far more data than previously thought, and reveal that
current alignment techniques do not eliminate memorization. | cs.LG | Scalable Extraction of Training Data from (Production) Language Models |
2023-11-28T18:32:19Z | http://arxiv.org/pdf/2311.17030v2 | http://arxiv.org/abs/2311.17030v2 | Aleksandar Makelov, Georg Lange, Neel Nanda | 2023-12-06T14:28:46Z | cs.LG, cs.AI, cs.CL | Mechanistic interpretability aims to understand model behaviors in terms of
specific, interpretable features, often hypothesized to manifest as
low-dimensional subspaces of activations. Specifically, recent studies have
explored subspace interventions (such as activation patching) as a way to
simultaneously manipulate model behavior and attribute the features behind it
to given subspaces.
In this work, we demonstrate that these two aims diverge, potentially leading
to an illusory sense of interpretability. Counterintuitively, even if a
subspace intervention makes the model's output behave as if the value of a
feature was changed, this effect may be achieved by activating a dormant
parallel pathway leveraging another subspace that is causally disconnected from
model outputs. We demonstrate this phenomenon in a distilled mathematical
example, in two real-world domains (the indirect object identification task and
factual recall), and present evidence for its prevalence in practice. In the
context of factual recall, we further show a link to rank-1 fact editing,
providing a mechanistic explanation for previous work observing an
inconsistency between fact editing performance and fact localization.
However, this does not imply that activation patching of subspaces is
intrinsically unfit for interpretability. To contextualize our findings, we
also show what a success case looks like in a task (indirect object
identification) where prior manual circuit analysis informs an understanding of
the location of a feature. We explore the additional evidence needed to argue
that a patched subspace is faithful. | cs.LG | Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patching |
2023-11-28T18:25:22Z | http://arxiv.org/pdf/2312.03741v1 | http://arxiv.org/abs/2312.03741v1 | Luis Fernando Lins, Nathalia Nascimento, Paulo Alencar, Toacy Oliveira, Donald Cowan | 2023-11-28T18:25:22Z | cs.CL, cs.AI, cs.SE | Business processes are commonly represented by modelling languages, such as
Event-driven Process Chain (EPC), Yet Another Workflow Language (YAWL), and the
most popular standard notation for modelling business processes, the Business
Process Model and Notation (BPMN). Most recently, chatbots, programs that allow
users to interact with a machine using natural language, have been increasingly
used for business process execution support. A recent category of chatbots
worth mentioning is generative-based chatbots, powered by Large Language Models
(LLMs) such as OpenAI's Generative Pre-Trained Transformer (GPT) model and
Google's Pathways Language Model (PaLM), which are trained on billions of
parameters and support conversational intelligence. However, it is not clear
whether generative-based chatbots are able to understand and meet the
requirements of constructs such as those provided by BPMN for process execution
support. This paper presents a case study to compare the performance of
prominent generative models, GPT and PaLM, in the context of process execution
support. The research sheds light into the challenging problem of using
conversational approaches supported by generative chatbots as a means to
understand process-aware modelling notations and support users to execute their
tasks. | cs.CL | Comparing Generative Chatbots Based on Process Requirements |
2023-11-28T17:56:34Z | http://arxiv.org/pdf/2312.03740v2 | http://arxiv.org/abs/2312.03740v2 | Prabin Bhandari | 2024-04-16T22:27:39Z | cs.CL, cs.AI | Autoregressive Large Language Models have transformed the landscape of
Natural Language Processing. Pre-train and prompt paradigm has replaced the
conventional approach of pre-training and fine-tuning for many downstream NLP
tasks. This shift has been possible largely due to LLMs and innovative
prompting techniques. LLMs have shown great promise for a variety of downstream
tasks owing to their vast parameters and huge datasets that they are
pre-trained on. However, in order to fully realize their potential, their
outputs must be guided towards the desired outcomes. Prompting, in which a
specific input or instruction is provided to guide the LLMs toward the intended
output, has become a tool for achieving this goal. In this paper, we discuss
the various prompting techniques that have been applied to fully harness the
power of LLMs. We present a taxonomy of existing literature on prompting
techniques and provide a concise survey based on this taxonomy. Further, we
identify some open problems in the realm of prompting in autoregressive LLMs
which could serve as a direction for future research. | cs.CL | A Survey on Prompting Techniques in LLMs |
2023-11-28T17:47:32Z | http://arxiv.org/pdf/2312.02179v1 | http://arxiv.org/abs/2312.02179v1 | Du Phan, Matthew D. Hoffman, David Dohan, Sholto Douglas, Tuan Anh Le, Aaron Parisi, Pavel Sountsov, Charles Sutton, Sharad Vikram, Rif A. Saurous | 2023-11-28T17:47:32Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) solve problems more accurately and interpretably
when instructed to work out the answer step by step using a
``chain-of-thought'' (CoT) prompt. One can also improve LLMs' performance on a
specific task by supervised fine-tuning, i.e., by using gradient ascent on some
tunable parameters to maximize the average log-likelihood of correct answers
from a labeled training set. Naively combining CoT with supervised tuning
requires supervision not just of the correct answers, but also of detailed
rationales that lead to those answers; these rationales are expensive to
produce by hand. Instead, we propose a fine-tuning strategy that tries to
maximize the \emph{marginal} log-likelihood of generating a correct answer
using CoT prompting, approximately averaging over all possible rationales. The
core challenge is sampling from the posterior over rationales conditioned on
the correct answer; we address it using a simple Markov-chain Monte Carlo
(MCMC) expectation-maximization (EM) algorithm inspired by the self-taught
reasoner (STaR), memoized wake-sleep, Markovian score climbing, and persistent
contrastive divergence. This algorithm also admits a novel control-variate
technique that drives the variance of our gradient estimates to zero as the
model improves. Applying our technique to GSM8K and the tasks in BIG-Bench
Hard, we find that this MCMC-EM fine-tuning technique typically improves the
model's accuracy on held-out examples more than STaR or prompt-tuning with or
without CoT. | cs.LG | Training Chain-of-Thought via Latent-Variable Inference |
2023-11-28T17:44:51Z | http://arxiv.org/pdf/2311.16989v4 | http://arxiv.org/abs/2311.16989v4 | Hailin Chen, Fangkai Jiao, Xingxuan Li, Chengwei Qin, Mathieu Ravaut, Ruochen Zhao, Caiming Xiong, Shafiq Joty | 2024-01-15T09:55:05Z | cs.CL | Upon its release in late 2022, ChatGPT has brought a seismic shift in the
entire landscape of AI, both in research and commerce. Through
instruction-tuning a large language model (LLM) with supervised fine-tuning and
reinforcement learning from human feedback, it showed that a model could answer
human questions and follow instructions on a broad panel of tasks. Following
this success, interests in LLMs have intensified, with new LLMs flourishing at
frequent interval across academia and industry, including many start-ups
focused on LLMs. While closed-source LLMs (e.g., OpenAI's GPT, Anthropic's
Claude) generally outperform their open-source counterparts, the progress on
the latter has been rapid with claims of achieving parity or even better on
certain tasks. This has crucial implications not only on research but also on
business. In this work, on the first anniversary of ChatGPT, we provide an
exhaustive overview of this success, surveying all tasks where an open-source
LLM has claimed to be on par or better than ChatGPT. | cs.CL | ChatGPT's One-year Anniversary: Are Open-Source Large Language Models Catching up? |
2023-11-28T17:25:34Z | http://arxiv.org/pdf/2311.16978v1 | http://arxiv.org/abs/2311.16978v1 | Christos-Nikolaos Zacharopoulos, Théo Desbordes, Mathias Sablé-Meyer | 2023-11-28T17:25:34Z | cs.CL | Subject-verb agreement in the presence of an attractor noun located between
the main noun and the verb elicits complex behavior: judgments of
grammaticality are modulated by the grammatical features of the attractor. For
example, in the sentence "The girl near the boys likes climbing", the attractor
(boys) disagrees in grammatical number with the verb (likes), creating a
locally implausible transition probability. Here, we parametrically modulate
the distance between the attractor and the verb while keeping the length of the
sentence equal. We evaluate the performance of both humans and two artificial
neural network models: both make more mistakes when the attractor is closer to
the verb, but neural networks get close to the chance level while humans are
mostly able to overcome the attractor interference. Additionally, we report a
linear effect of attractor distance on reaction times. We hypothesize that a
possible reason for the proximity effect is the calculation of transition
probabilities between adjacent words. Nevertheless, classical models of
attraction such as the cue-based model might suffice to explain this
phenomenon, thus paving the way for new research. Data and analyses available
at https://osf.io/d4g6k | cs.CL | Assessing the influence of attractor-verb distance on grammatical agreement in humans and language models |
2023-11-28T17:12:06Z | http://arxiv.org/pdf/2311.16965v1 | http://arxiv.org/abs/2311.16965v1 | Aman Yadav, Abhishek Vichare | 2023-11-28T17:12:06Z | cs.CL | Artificial intelligence and machine learning have significantly bolstered the
technological world. This paper explores the potential of transfer learning in
natural language processing focusing mainly on sentiment analysis. The models
trained on the big data can also be used where data are scarce. The claim is
that, compared to training models from scratch, transfer learning, using
pre-trained BERT models, can increase sentiment classification accuracy. The
study adopts a sophisticated experimental design that uses the IMDb dataset of
sentimentally labelled movie reviews. Pre-processing includes tokenization and
encoding of text data, making it suitable for NLP models. The dataset is used
on a BERT based model, measuring its performance using accuracy. The result
comes out to be 100 per cent accurate. Although the complete accuracy could
appear impressive, it might be the result of overfitting or a lack of
generalization. Further analysis is required to ensure the model's ability to
handle diverse and unseen data. The findings underscore the effectiveness of
transfer learning in NLP, showcasing its potential to excel in sentiment
analysis tasks. However, the research calls for a cautious interpretation of
perfect accuracy and emphasizes the need for additional measures to validate
the model's generalization. | cs.CL | Natural Language Processing Through Transfer Learning: A Case Study on Sentiment Analysis |
2023-11-28T16:46:14Z | http://arxiv.org/pdf/2311.16941v1 | http://arxiv.org/abs/2311.16941v1 | Vaidehi Patil, Adyasha Maharana, Mohit Bansal | 2023-11-28T16:46:14Z | cs.LG, cs.AI, cs.CL, cs.CV, stat.ME | Most existing debiasing methods for multimodal models, including causal
intervention and inference methods, utilize approximate heuristics to represent
the biases, such as shallow features from early stages of training or unimodal
features for multimodal tasks like VQA, etc., which may not be accurate. In
this paper, we study bias arising from confounders in a causal graph for
multimodal data and examine a novel approach that leverages causally-motivated
information minimization to learn the confounder representations. Robust
predictive features contain diverse information that helps a model generalize
to out-of-distribution data. Hence, minimizing the information content of
features obtained from a pretrained biased model helps learn the simplest
predictive features that capture the underlying data distribution. We treat
these features as confounder representations and use them via methods motivated
by causal theory to remove bias from models. We find that the learned
confounder representations indeed capture dataset biases, and the proposed
debiasing methods improve out-of-distribution (OOD) performance on multiple
multimodal datasets without sacrificing in-distribution performance.
Additionally, we introduce a novel metric to quantify the sufficiency of
spurious features in models' predictions that further demonstrates the
effectiveness of our proposed methods. Our code is available at:
https://github.com/Vaidehi99/CausalInfoMin | cs.LG | Debiasing Multimodal Models via Causal Information Minimization |
2023-11-28T16:26:35Z | http://arxiv.org/pdf/2311.16922v1 | http://arxiv.org/abs/2311.16922v1 | Sicong Leng, Hang Zhang, Guanzheng Chen, Xin Li, Shijian Lu, Chunyan Miao, Lidong Bing | 2023-11-28T16:26:35Z | cs.CV, cs.AI, cs.CL | Large Vision-Language Models (LVLMs) have advanced considerably, intertwining
visual recognition and language understanding to generate content that is not
only coherent but also contextually attuned. Despite their success, LVLMs still
suffer from the issue of object hallucinations, where models generate plausible
yet incorrect outputs that include objects that do not exist in the images. To
mitigate this issue, we introduce Visual Contrastive Decoding (VCD), a simple
and training-free method that contrasts output distributions derived from
original and distorted visual inputs. The proposed VCD effectively reduces the
over-reliance on statistical bias and unimodal priors, two essential causes of
object hallucinations. This adjustment ensures the generated content is closely
grounded to visual inputs, resulting in contextually accurate outputs. Our
experiments show that VCD, without either additional training or the usage of
external tools, significantly mitigates the object hallucination issue across
different LVLM families. Beyond mitigating object hallucinations, VCD also
excels in general LVLM benchmarks, highlighting its wide-ranging applicability. | cs.CV | Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding |
2023-11-28T16:03:22Z | http://arxiv.org/pdf/2312.03739v1 | http://arxiv.org/abs/2312.03739v1 | Ullman Galen, Frey Lee, Woods Ali | 2023-11-28T16:03:22Z | cs.CL | Aspect-based sentiment analysis (ABSA), a nuanced task in text analysis,
seeks to discern sentiment orientation linked to specific aspect terms in text.
Traditional approaches often overlook or inadequately model the explicit
syntactic structures of sentences, crucial for effective aspect term
identification and sentiment determination. Addressing this gap, we introduce
an innovative model: Syntactic Dependency Enhanced Multi-Task Interaction
Architecture (SDEMTIA) for comprehensive ABSA. Our approach innovatively
exploits syntactic knowledge (dependency relations and types) using a
specialized Syntactic Dependency Embedded Interactive Network (SDEIN). We also
incorporate a novel and efficient message-passing mechanism within a multi-task
learning framework to bolster learning efficacy. Our extensive experiments on
benchmark datasets showcase our model's superiority, significantly surpassing
existing methods. Additionally, incorporating BERT as an auxiliary feature
extractor further enhances our model's performance. | cs.CL | Syntax-Informed Interactive Model for Comprehensive Aspect-Based Sentiment Analysis |
2023-11-28T15:31:11Z | http://arxiv.org/pdf/2311.16882v1 | http://arxiv.org/abs/2311.16882v1 | Bowen Li, Yongxin Yang, Steven McDonagh, Shifeng Zhang, Petru-Daniel Tudosiu, Sarah Parisot | 2023-11-28T15:31:11Z | cs.CV, cs.CL, cs.LG | Image editing affords increased control over the aesthetics and content of
generated images. Pre-existing works focus predominantly on text-based
instructions to achieve desired image modifications, which limit edit precision
and accuracy. In this work, we propose an inference-time editing optimisation,
designed to extend beyond textual edits to accommodate multiple editing
instruction types (e.g. spatial layout-based; pose, scribbles, edge maps). We
propose to disentangle the editing task into two competing subtasks: successful
local image modifications and global content consistency preservation, where
subtasks are guided through two dedicated loss functions. By allowing to adjust
the influence of each loss function, we build a flexible editing solution that
can be adjusted to user preferences. We evaluate our method using text, pose
and scribble edit conditions, and highlight our ability to achieve complex
edits, through both qualitative and quantitative experiments. | cs.CV | Optimisation-Based Multi-Modal Semantic Image Editing |
2023-11-28T15:28:22Z | http://arxiv.org/pdf/2312.03738v1 | http://arxiv.org/abs/2312.03738v1 | Jane Sunny, Tom Padraig, Roggie Terry, Woods Ali | 2023-11-28T15:28:22Z | cs.CL | Recent progress in aspect-level sentiment classification has been propelled
by the incorporation of graph neural networks (GNNs) leveraging syntactic
structures, particularly dependency trees. Nevertheless, the performance of
these models is often hampered by the innate inaccuracies of parsing
algorithms. To mitigate this challenge, we introduce SynthFusion, an innovative
graph ensemble method that amalgamates predictions from multiple parsers. This
strategy blends diverse dependency relations prior to the application of GNNs,
enhancing robustness against parsing errors while avoiding extra computational
burdens. SynthFusion circumvents the pitfalls of overparameterization and
diminishes the risk of overfitting, prevalent in models with stacked GNN
layers, by optimizing graph connectivity. Our empirical evaluations on the
SemEval14 and Twitter14 datasets affirm that SynthFusion not only outshines
models reliant on single dependency trees but also eclipses alternative
ensemble techniques, achieving this without an escalation in model complexity. | cs.CL | Syntactic Fusion: Enhancing Aspect-Level Sentiment Analysis Through Multi-Tree Graph Integration |
2023-11-28T15:12:47Z | http://arxiv.org/pdf/2311.16867v2 | http://arxiv.org/abs/2311.16867v2 | Ebtesam Almazrouei, Hamza Alobeidli, Abdulaziz Alshamsi, Alessandro Cappelli, Ruxandra Cojocaru, Mérouane Debbah, Étienne Goffinet, Daniel Hesslow, Julien Launay, Quentin Malartic, Daniele Mazzotta, Badreddine Noune, Baptiste Pannier, Guilherme Penedo | 2023-11-29T19:45:10Z | cs.CL, cs.AI | We introduce the Falcon series: 7B, 40B, and 180B parameters causal
decoder-only models trained on a diverse high-quality corpora predominantly
assembled from web data. The largest model, Falcon-180B, has been trained on
over 3.5 trillion tokens of text--the largest openly documented pretraining
run. Falcon-180B significantly outperforms models such as PaLM or Chinchilla,
and improves upon concurrently developed models such as LLaMA 2 or
Inflection-1. It nears the performance of PaLM-2-Large at a reduced pretraining
and inference cost, making it, to our knowledge, one of the three best language
models in the world along with GPT-4 and PaLM-2-Large. We report detailed
evaluations, as well as a deep dive into the methods and custom tooling
employed to pretrain Falcon. Notably, we report on our custom distributed
training codebase, allowing us to efficiently pretrain these models on up to
4,096 A100s on cloud AWS infrastructure with limited interconnect. We release a
600B tokens extract of our web dataset, as well as the Falcon-7/40/180B models
under a permissive license to foster open-science and accelerate the
development of an open ecosystem of large language models. | cs.CL | The Falcon Series of Open Language Models |
2023-11-28T15:12:11Z | http://arxiv.org/pdf/2311.16865v1 | http://arxiv.org/abs/2311.16865v1 | Noëmi Aepli, Chantal Amrhein, Florian Schottmann, Rico Sennrich | 2023-11-28T15:12:11Z | cs.CL, I.2.7 | For sensible progress in natural language processing, it is important that we
are aware of the limitations of the evaluation metrics we use. In this work, we
evaluate how robust metrics are to non-standardized dialects, i.e. spelling
differences in language varieties that do not have a standard orthography. To
investigate this, we collect a dataset of human translations and human
judgments for automatic machine translations from English to two Swiss German
dialects. We further create a challenge set for dialect variation and benchmark
existing metrics' performances. Our results show that existing metrics cannot
reliably evaluate Swiss German text generation outputs, especially on segment
level. We propose initial design adaptations that increase robustness in the
face of non-standardized dialects, although there remains much room for further
improvement. The dataset, code, and models are available here:
https://github.com/textshuttle/dialect_eval | cs.CL | A Benchmark for Evaluating Machine Translation Metrics on Dialects Without Standard Orthography |
2023-11-28T14:55:52Z | http://arxiv.org/pdf/2311.16842v2 | http://arxiv.org/abs/2311.16842v2 | Furui Cheng, Vilém Zouhar, Simran Arora, Mrinmaya Sachan, Hendrik Strobelt, Mennatallah El-Assady | 2024-04-04T15:18:30Z | cs.HC, cs.CL | Large Language Models (LLMs) are notorious for blending fact with fiction and
generating non-factual content, known as hallucinations. To address this
challenge, we propose an interactive system that helps users gain insight into
the reliability of the generated text. Our approach is based on the idea that
the self-consistency of multiple samples generated by the same LLM relates to
its confidence in individual claims in the generated texts. Using this idea, we
design RELIC, an interactive system that enables users to investigate and
verify semantic-level variations in multiple long-form responses. This allows
users to recognize potentially inaccurate information in the generated text and
make necessary corrections. From a user study with ten participants, we
demonstrate that our approach helps users better verify the reliability of the
generated text. We further summarize the design implications and lessons
learned from this research for future studies of reliable human-LLM
interactions. | cs.HC | RELIC: Investigating Large Language Model Responses using Self-Consistency |
2023-11-28T14:55:22Z | http://arxiv.org/pdf/2311.16840v1 | http://arxiv.org/abs/2311.16840v1 | Julie Hunter, Jérôme Louradour, Virgile Rennard, Ismaïl Harrando, Guokan Shang, Jean-Pierre Lorré | 2023-11-28T14:55:22Z | cs.CL, cs.AI | We present the Claire French Dialogue Dataset (CFDD), a resource created by
members of LINAGORA Labs in the context of the OpenLLM France initiative. CFDD
is a corpus containing roughly 160 million words from transcripts and stage
plays in French that we have assembled and publicly released in an effort to
further the development of multilingual, open source language models. This
paper describes the 24 individual corpora of which CFDD is composed and
provides links and citations to their original sources. It also provides our
proposed breakdown of the full CFDD dataset into eight categories of subcorpora
and describes the process we followed to standardize the format of the final
dataset. We conclude with a discussion of similar work and future directions. | cs.CL | The Claire French Dialogue Dataset |
2023-11-28T14:54:37Z | http://arxiv.org/pdf/2311.16839v2 | http://arxiv.org/abs/2311.16839v2 | Zhiyuan Zhao, Bin Wang, Linke Ouyang, Xiaoyi Dong, Jiaqi Wang, Conghui He | 2024-02-06T16:43:31Z | cs.CV, cs.CL | Multimodal large language models have made significant advancements in recent
years, yet they still suffer from a common issue known as the "hallucination
problem", in which the models generate textual descriptions that inaccurately
depict or entirely fabricate content from associated images. This paper
introduces a novel solution, Hallucination-Aware Direct Preference Optimization
(HA-DPO), which reframes the hallucination problem as a preference selection
task. The model is trained to favor the non-hallucinating response when
presented with two responses of the same image (one accurate and one
hallucinatory). Furthermore, this paper proposes an efficient pipeline for
constructing positive~(non-hallucinatory) and negative~(hallucinatory) sample
pairs, ensuring a high-quality, style-consistent dataset for robust preference
learning. When applied to three mainstream multimodal models, HA-DPO
significantly reduced hallucination issues and amplified the models'
generalization capabilities. Notably, the MiniGPT-4 model, when enhanced with
HA-DPO, demonstrated a substantial improvement: POPE accuracy rose from 51.13%
to 86.13% (an absolute improvement of 35%), and the MME score surged from
932.00 to 1326.46 (a relative improvement of 42.32%). The codes, models, and
datasets are made accessible at https://opendatalab.github.io/HA-DPO. | cs.CV | Beyond Hallucinations: Enhancing LVLMs through Hallucination-Aware Direct Preference Optimization |
2023-11-28T14:51:13Z | http://arxiv.org/pdf/2311.17126v1 | http://arxiv.org/abs/2311.17126v1 | Xiaohui Chen, Yongfei Liu, Yingxiang Yang, Jianbo Yuan, Quanzeng You, Li-Ping Liu, Hongxia Yang | 2023-11-28T14:51:13Z | cs.CV, cs.CL | Recent advancements in text-to-image (T2I) generative models have shown
remarkable capabilities in producing diverse and imaginative visuals based on
text prompts. Despite the advancement, these diffusion models sometimes
struggle to translate the semantic content from the text into images entirely.
While conditioning on the layout has shown to be effective in improving the
compositional ability of T2I diffusion models, they typically require manual
layout input. In this work, we introduce a novel approach to improving T2I
diffusion models using Large Language Models (LLMs) as layout generators. Our
method leverages the Chain-of-Thought prompting of LLMs to interpret text and
generate spatially reasonable object layouts. The generated layout is then used
to enhance the generated images' composition and spatial accuracy. Moreover, we
propose an efficient adapter based on a cross-attention mechanism, which
explicitly integrates the layout information into the stable diffusion models.
Our experiments demonstrate significant improvements in image quality and
layout accuracy, showcasing the potential of LLMs in augmenting generative
image models. | cs.CV | Reason out Your Layout: Evoking the Layout Master from Large Language Models for Text-to-Image Synthesis |
2023-11-28T14:49:23Z | http://arxiv.org/pdf/2311.16832v1 | http://arxiv.org/abs/2311.16832v1 | Jinfeng Zhou, Zhuang Chen, Dazhen Wan, Bosi Wen, Yi Song, Jifan Yu, Yongkang Huang, Libiao Peng, Jiaming Yang, Xiyao Xiao, Sahand Sabour, Xiaohan Zhang, Wenjing Hou, Yijia Zhang, Yuxiao Dong, Jie Tang, Minlie Huang | 2023-11-28T14:49:23Z | cs.CL, cs.AI | In this paper, we present CharacterGLM, a series of models built upon
ChatGLM, with model sizes ranging from 6B to 66B parameters. Our CharacterGLM
is designed for generating Character-based Dialogues (CharacterDial), which
aims to equip a conversational AI system with character customization for
satisfying people's inherent social desires and emotional needs. On top of
CharacterGLM, we can customize various AI characters or social agents by
configuring their attributes (identities, interests, viewpoints, experiences,
achievements, social relationships, etc.) and behaviors (linguistic features,
emotional expressions, interaction patterns, etc.). Our model outperforms most
mainstream close-source large langauge models, including the GPT series,
especially in terms of consistency, human-likeness, and engagement according to
manual evaluations. We will release our 6B version of CharacterGLM and a subset
of training data to facilitate further research development in the direction of
character-based dialogue generation. | cs.CL | CharacterGLM: Customizing Chinese Conversational AI Characters with Large Language Models |
2023-11-28T14:36:43Z | http://arxiv.org/pdf/2311.16822v2 | http://arxiv.org/abs/2311.16822v2 | Martin Briesch, Dominik Sobania, Franz Rothlauf | 2024-06-17T07:07:30Z | cs.LG, cs.CL, cs.NE | Large Language Models (LLM) are already widely used to generate content for a
variety of online platforms. As we are not able to safely distinguish
LLM-generated content from human-produced content, LLM-generated content is
used to train the next generation of LLMs, giving rise to a self-consuming
training loop. From the image generation domain we know that such a
self-consuming training loop reduces both quality and diversity of images
finally ending in a model collapse. However, it is unclear whether this
alarming effect can also be observed for LLMs. Therefore, we present the first
study investigating the self-consuming training loop for LLMs. Further, we
propose a novel method based on logic expressions that allows us to
unambiguously verify the correctness of LLM-generated content, which is
difficult for natural language text. We find that the self-consuming training
loop produces correct outputs, however, the output declines in its diversity
depending on the proportion of the used generated data. Fresh data can slow
down this decline, but not stop it. Given these concerning results, we
encourage researchers to study methods to negate this process. | cs.LG | Large Language Models Suffer From Their Own Output: An Analysis of the Self-Consuming Training Loop |
2023-11-28T13:51:32Z | http://arxiv.org/pdf/2311.16789v1 | http://arxiv.org/abs/2311.16789v1 | Hongru Wang, Lingzhi Wang, Yiming Du, Liang Chen, Jingyan Zhou, Yufei Wang, Kam-Fai Wong | 2023-11-28T13:51:32Z | cs.CL, cs.AI | Dialogue systems, including task-oriented_dialogue_system (TOD) and
open-domain_dialogue_system (ODD), have undergone significant transformations,
with language_models (LM) playing a central role. This survey delves into the
historical trajectory of dialogue systems, elucidating their intricate
relationship with advancements in language models by categorizing this
evolution into four distinct stages, each marked by pivotal LM breakthroughs:
1) Early_Stage: characterized by statistical LMs, resulting in rule-based or
machine-learning-driven dialogue_systems; 2) Independent development of TOD and
ODD based on neural_language_models (NLM; e.g., LSTM and GRU), since NLMs lack
intrinsic knowledge in their parameters; 3) fusion between different types of
dialogue systems with the advert of pre-trained_language_models (PLMs),
starting from the fusion between four_sub-tasks_within_TOD, and then
TOD_with_ODD; and 4) current LLM-based_dialogue_system, wherein LLMs can be
used to conduct TOD and ODD seamlessly. Thus, our survey provides a
chronological perspective aligned with LM breakthroughs, offering a
comprehensive review of state-of-the-art research outcomes. What's more, we
focus on emerging topics and discuss open challenges, providing valuable
insights into future directions for LLM-based_dialogue_systems. Through this
exploration, we pave the way for a deeper_comprehension of the evolution,
guiding future developments in LM-based dialogue_systems. | cs.CL | A Survey of the Evolution of Language Model-Based Dialogue Systems |
2023-11-28T13:50:50Z | http://arxiv.org/pdf/2311.16787v2 | http://arxiv.org/abs/2311.16787v2 | Vilém Zouhar, Věra Kloudová, Martin Popel, Ondřej Bojar | 2024-03-08T12:21:12Z | cs.CL | The overall translation quality reached by current machine translation (MT)
systems for high-resourced language pairs is remarkably good. Standard methods
of evaluation are not suitable nor intended to uncover the many translation
errors and quality deficiencies that still persist. Furthermore, the quality of
standard reference translations is commonly questioned and comparable quality
levels have been reached by MT alone in several language pairs. Navigating
further research in these high-resource settings is thus difficult. In this
article, we propose a methodology for creating more reliable document-level
human reference translations, called "optimal reference translations," with the
simple aim to raise the bar of what should be deemed "human translation
quality." We evaluate the obtained document-level optimal reference
translations in comparison with "standard" ones, confirming a significant
quality increase and also documenting the relationship between evaluation and
translation editing. | cs.CL | Evaluating Optimal Reference Translations |
2023-11-28T13:27:21Z | http://arxiv.org/pdf/2312.03737v1 | http://arxiv.org/abs/2312.03737v1 | Rajaraman Kanagasabai, Anitha Veeramani | 2023-11-28T13:27:21Z | cs.CL, cs.AI | This paper describes our system for addressing SMM4H 2023 Shared Task 2 on
"Classification of sentiment associated with therapies (aspect-oriented)". In
our work, we adopt an approach based on Natural language inference (NLI) to
formulate this task as a sentence pair classification problem, and train
transformer models to predict sentiment associated with a therapy on a given
text. Our best model achieved 75.22\% F1-score which was 11\% (4\%) more than
the mean (median) score of all teams' submissions. | cs.CL | A Generic NLI approach for Classification of Sentiment Associated with Therapies |
2023-11-28T13:20:41Z | http://arxiv.org/pdf/2312.03736v1 | http://arxiv.org/abs/2312.03736v1 | Aleksandar Kovačević, Bojana Bašaragin, Nikola Milošević, Goran Nenadić | 2023-11-28T13:20:41Z | cs.CL, cs.AI, cs.CR, cs.DL, cs.LG | Background: Electronic health records (EHRs) are a valuable resource for
data-driven medical research. However, the presence of protected health
information (PHI) makes EHRs unsuitable to be shared for research purposes.
De-identification, i.e. the process of removing PHI is a critical step in
making EHR data accessible. Natural language processing has repeatedly
demonstrated its feasibility in automating the de-identification process.
Objectives: Our study aims to provide systematic evidence on how the
de-identification of clinical free text has evolved in the last thirteen years,
and to report on the performances and limitations of the current
state-of-the-art systems. In addition, we aim to identify challenges and
potential research opportunities in this field. Methods: A systematic search in
PubMed, Web of Science and the DBLP was conducted for studies published between
January 2010 and February 2023. Titles and abstracts were examined to identify
the relevant studies. Selected studies were then analysed in-depth, and
information was collected on de-identification methodologies, data sources, and
measured performance. Results: A total of 2125 publications were identified for
the title and abstract screening. 69 studies were found to be relevant. Machine
learning (37 studies) and hybrid (26 studies) approaches are predominant, while
six studies relied only on rules. Majority of the approaches were trained and
evaluated on public corpora. The 2014 i2b2/UTHealth corpus is the most
frequently used (36 studies), followed by the 2006 i2b2 (18 studies) and 2016
CEGS N-GRID (10 studies) corpora. | cs.CL | De-identification of clinical free text using natural language processing: A systematic review of current approaches |
2023-11-28T13:08:26Z | http://arxiv.org/pdf/2311.16764v1 | http://arxiv.org/abs/2311.16764v1 | Amos Calamida, Farhad Nooralahzadeh, Morteza Rohanian, Koji Fujimoto, Mizuho Nishio, Michael Krauthammer | 2023-11-28T13:08:26Z | cs.CL | We propose a new automated evaluation metric for machine-generated radiology
reports using the successful COMET architecture adapted for the radiology
domain. We train and publish four medically-oriented model checkpoints,
including one trained on RadGraph, a radiology knowledge graph. Our results
show that our metric correlates moderately to high with established metrics
such as BERTscore, BLEU, and CheXbert scores. Furthermore, we demonstrate that
one of our checkpoints exhibits a high correlation with human judgment, as
assessed using the publicly available annotations of six board-certified
radiologists, using a set of 200 reports. We also performed our own analysis
gathering annotations with two radiologists on a collection of 100 reports. The
results indicate the potential effectiveness of our method as a
radiology-specific evaluation metric. The code, data, and model checkpoints to
reproduce our findings will be publicly available. | cs.CL | Radiology-Aware Model-Based Evaluation Metric for Report Generation |
2023-11-28T12:30:43Z | http://arxiv.org/pdf/2312.03735v1 | http://arxiv.org/abs/2312.03735v1 | David Herel, Tomas Mikolov | 2023-11-28T12:30:43Z | cs.CL, cs.AI | Generalization is arguably the most important goal of statistical language
modeling research. Publicly available benchmarks and papers published with an
open-source code have been critical to advancing the field. However, it is
often very difficult, and sometimes even impossible, to reproduce the results
fully as reported in publications. In this paper, we propose a simple framework
that should help advance the state of the art in language modeling in terms of
generalization. We propose to publish not just the code, but also probabilities
on dev and test sets with future publications so that one can easily add the
new model into an ensemble. This has crucial advantages: it is much easier to
determine whether a newly proposed model is actually complementary to the
current baseline. Therefore, instead of inventing new names for the old tricks,
the scientific community can advance faster. Finally, this approach promotes
diversity of ideas: one does not need to create an individual model that is the
new state of the art to attract attention; it will be sufficient to develop a
new model that learns patterns which other models do not. Thus, even a
suboptimal model can be found to have value. Remarkably, our approach has
yielded new state-of-the-art results across various language modeling
benchmarks up to 10%. | cs.CL | Advancing State of the Art in Language Modeling |
2023-11-28T12:29:33Z | http://arxiv.org/pdf/2311.16733v4 | http://arxiv.org/abs/2311.16733v4 | Mohamed Nejjar, Luca Zacharias, Fabian Stiehle, Ingo Weber | 2024-04-23T08:12:46Z | cs.SE, cs.AI, cs.CL | Large language models (LLMs) have been touted to enable increased
productivity in many areas of today's work life. Scientific research as an area
of work is no exception: the potential of LLM-based tools to assist in the
daily work of scientists has become a highly discussed topic across
disciplines. However, we are only at the very onset of this subject of study.
It is still unclear how the potential of LLMs will materialise in research
practice. With this study, we give first empirical evidence on the use of LLMs
in the research process. We have investigated a set of use cases for LLM-based
tools in scientific research, and conducted a first study to assess to which
degree current tools are helpful. In this paper we report specifically on use
cases related to software engineering, such as generating application code and
developing scripts for data analytics. While we studied seemingly simple use
cases, results across tools differ significantly. Our results highlight the
promise of LLM-based tools in general, yet we also observe various issues,
particularly regarding the integrity of the output these tools provide. | cs.SE | LLMs for Science: Usage for Code Generation and Data Analysis |
2023-11-28T11:05:20Z | http://arxiv.org/pdf/2312.03734v1 | http://arxiv.org/abs/2312.03734v1 | Ruixiang Jiang, Lingbo Liu, Changwen Chen | 2023-11-28T11:05:20Z | cs.CL, cs.AI | We show that the representation of one modality can effectively guide the
prompting of another modality for parameter-efficient multimodal fusion.
Specifically, we first encode one modality and use its representation as a
prior to conditionally prompt all frozen layers of the other modality. This is
achieved by disentangling the vanilla prompt vectors into three types of
specialized prompts that adaptively capture global-level and instance-level
features. To better produce the instance-wise prompt, we introduce the mixture
of prompt experts (MoPE) to dynamically route each instance to the most
suitable prompt experts for encoding. We further study a regularization term to
avoid degenerated prompt expert routing. Thanks to our design, our method can
effectively transfer the pretrained knowledge in unimodal encoders for
downstream multimodal tasks. Compared with vanilla prompting, we show that our
MoPE-based conditional prompting is more expressive, thereby scales better with
training data and the total number of prompts. We also demonstrate that our
prompt tuning is architecture-agnostic, thereby offering high modularity.
Extensive experiments over three multimodal datasets demonstrate
state-of-the-art results, matching or surpassing the performance achieved
through fine-tuning, while only necessitating 0.7% of the trainable parameters.
Code will be released: https://github.com/songrise/ConditionalPrompt. | cs.CL | Conditional Prompt Tuning for Multimodal Fusion |
2023-11-28T10:50:00Z | http://arxiv.org/pdf/2311.16678v1 | http://arxiv.org/abs/2311.16678v1 | Dan Ma, Jun Xu, Zongyu Wang, Xuezhi Cao, Yunsen Xian | 2023-11-28T10:50:00Z | cs.CL | Product reviews often contain a large number of implicit aspects and
object-attribute co-existence cases. Unfortunately, many existing studies in
Aspect-Based Sentiment Analysis (ABSA) have overlooked this issue, which can
make it difficult to extract opinions comprehensively and fairly. In this
paper, we propose a new task called Entity-Aspect-Opinion-Sentiment Quadruple
Extraction (EASQE), which aims to hierarchically decompose aspect terms into
entities and aspects to avoid information loss, non-exclusive annotations, and
opinion misunderstandings in ABSA tasks. To facilitate research in this new
task, we have constructed four datasets (Res14-EASQE, Res15-EASQE, Res16-EASQE,
and Lap14-EASQE) based on the SemEval Restaurant and Laptop datasets. We have
also proposed a novel two-stage sequence-tagging based Trigger-Opinion
framework as the baseline for the EASQE task. Empirical evaluations show that
our Trigger-Opinion framework can generate satisfactory EASQE results and can
also be applied to other ABSA tasks, significantly outperforming
state-of-the-art methods. We have made the four datasets and source code of
Trigger-Opinion publicly available to facilitate further research in this area. | cs.CL | Entity-Aspect-Opinion-Sentiment Quadruple Extraction for Fine-grained Sentiment Analysis |
2023-11-28T10:42:35Z | http://arxiv.org/pdf/2311.16675v1 | http://arxiv.org/abs/2311.16675v1 | Gioele Cadamuro, Marco Gruppo | 2023-11-28T10:42:35Z | cs.CL | We hereby present a solution to a semantic textual similarity (STS) problem
in which it is necessary to match two sentences containing, as the only
distinguishing factor, highly specific information (such as names, addresses,
identification codes), and from which we need to derive a definition for when
they are similar and when they are not. The solution revolves around the use of
a neural network, based on the siamese architecture, to create the
distributions of the distances between similar and dissimilar pairs of
sentences. The goal of these distributions is to find a discriminating factor,
that we call "threshold", which represents a well-defined quantity that can be
used to distinguish vector distances of similar pairs from vector distances of
dissimilar pairs in new predictions and later analyses. In addition, we
developed a way to score the predictions by combining attributes from both the
distributions' features and the way the distance function works. Finally, we
generalize the results showing that they can be transferred to a wider range of
domains by applying the system discussed to a well-known and widely used
benchmark dataset for STS problems. | cs.CL | A Distribution-Based Threshold for Determining Sentence Similarity |
2023-11-28T10:26:57Z | http://arxiv.org/pdf/2311.17107v1 | http://arxiv.org/abs/2311.17107v1 | Romain Lacombe, Kerrie Wu, Eddie Dilworth | 2023-11-28T10:26:57Z | cs.LG, cs.AI, cs.CL, cs.CY, cs.IR | Evaluating the accuracy of outputs generated by Large Language Models (LLMs)
is especially important in the climate science and policy domain. We introduce
the Expert Confidence in Climate Statements (ClimateX) dataset, a novel,
curated, expert-labeled dataset consisting of 8094 climate statements collected
from the latest Intergovernmental Panel on Climate Change (IPCC) reports,
labeled with their associated confidence levels. Using this dataset, we show
that recent LLMs can classify human expert confidence in climate-related
statements, especially in a few-shot learning setting, but with limited (up to
47%) accuracy. Overall, models exhibit consistent and significant
over-confidence on low and medium confidence statements. We highlight
implications of our results for climate communication, LLMs evaluation
strategies, and the use of LLMs in information retrieval systems. | cs.LG | ClimateX: Do LLMs Accurately Assess Human Expert Confidence in Climate Statements? |
2023-11-28T10:02:08Z | http://arxiv.org/pdf/2311.16650v1 | http://arxiv.org/abs/2311.16650v1 | Jiahuan Yan, Haojun Gao, Zhang Kai, Weize Liu, Danny Chen, Jian Wu, Jintai Chen | 2023-11-28T10:02:08Z | cs.CL | Deep learning approaches exhibit promising performances on various text
tasks. However, they are still struggling on medical text classification since
samples are often extremely imbalanced and scarce. Different from existing
mainstream approaches that focus on supplementary semantics with external
medical information, this paper aims to rethink the data challenges in medical
texts and present a novel framework-agnostic algorithm called Text2Tree that
only utilizes internal label hierarchy in training deep learning models. We
embed the ICD code tree structure of labels into cascade attention modules for
learning hierarchy-aware label representations. Two new learning schemes,
Similarity Surrogate Learning (SSL) and Dissimilarity Mixup Learning (DML), are
devised to boost text classification by reusing and distinguishing samples of
other labels following the label representation hierarchy, respectively.
Experiments on authoritative public datasets and real-world medical records
show that our approach stably achieves superior performances over classical and
advanced imbalanced classification methods. | cs.CL | Text2Tree: Aligning Text Representation to the Label Tree Hierarchy for Imbalanced Medical Classification |
2023-11-28T09:45:02Z | http://arxiv.org/pdf/2311.16639v2 | http://arxiv.org/abs/2311.16639v2 | Gaël Le Mens, Aina Gallego | 2024-05-13T14:16:05Z | cs.CL | We use instruction-tuned Large Language Models (LLMs) such as GPT-4, MiXtral,
and Llama 3 to position political texts within policy and ideological spaces.
We directly ask the LLMs where a text document or its author stand on the focal
policy dimension. We illustrate and validate the approach by scaling British
party manifestos on the economic, social, and immigration policy dimensions;
speeches from a European Parliament debate in 10 languages on the anti- to
pro-subsidy dimension; Senators of the 117th US Congress based on their tweets
on the left-right ideological spectrum; and tweets published by US
Representatives and Senators after the training cutoff date of GPT-4. The
correlation between the position estimates obtained with the best LLMs and
benchmarks based on coding by experts, crowdworkers or roll call votes exceeds
.90. This training-free approach also outperforms supervised classifiers
trained on large amounts of data. Using instruction-tuned LLMs to scale texts
in policy and ideological spaces is fast, cost-efficient, reliable, and
reproducible (in the case of open LLMs) even if the texts are short and written
in different languages. We conclude with cautionary notes about the need for
empirical validation. | cs.CL | Scaling Political Texts with Large Language Models: Asking a Chatbot Might Be All You Need |
2023-11-28T09:33:41Z | http://arxiv.org/pdf/2312.09451v1 | http://arxiv.org/abs/2312.09451v1 | Sourabh Zanwar, Daniel Wiechmann, Yu Qiao, Elma Kerz | 2023-11-28T09:33:41Z | cs.CL | This paper presents our system employed for the Social Media Mining for
Health 2023 Shared Task 4: Binary classification of English Reddit posts
self-reporting a social anxiety disorder diagnosis. We systematically
investigate and contrast the efficacy of hybrid and ensemble models that
harness specialized medical domain-adapted transformers in conjunction with
BiLSTM neural networks. The evaluation results outline that our best performing
model obtained 89.31% F1 on the validation set and 83.76% F1 on the test set. | cs.CL | MANTIS at #SMM4H 2023: Leveraging Hybrid and Ensemble Models for Detection of Social Anxiety Disorder on Reddit |
2023-11-28T09:21:48Z | http://arxiv.org/pdf/2311.16620v1 | http://arxiv.org/abs/2311.16620v1 | Itamar Zimerman, Lior Wolf | 2023-11-28T09:21:48Z | cs.LG, cs.CL, F.2.2; I.2.7 | Despite their dominance in modern DL and, especially, NLP domains,
transformer architectures exhibit sub-optimal performance on long-range tasks
compared to recent layers that are specifically designed for this purpose. In
this work, drawing inspiration from key attributes of long-range layers, such
as state-space layers, linear RNN layers, and global convolution layers, we
demonstrate that minimal modifications to the transformer architecture can
significantly enhance performance on the Long Range Arena (LRA) benchmark, thus
narrowing the gap with these specialized layers. We identify that two key
principles for long-range tasks are (i) incorporating an inductive bias towards
smoothness, and (ii) locality. As we show, integrating these ideas into the
attention mechanism improves results with a negligible amount of additional
computation and without any additional trainable parameters. Our theory and
experiments also shed light on the reasons for the inferior performance of
transformers on long-range tasks and identify critical properties that are
essential for successfully capturing long-range dependencies. | cs.LG | On the Long Range Abilities of Transformers |
2023-11-28T08:13:29Z | http://arxiv.org/pdf/2311.16588v2 | http://arxiv.org/abs/2311.16588v2 | Rui Yang, Qingcheng Zeng, Keen You, Yujie Qiao, Lucas Huang, Chia-Chun Hsieh, Benjamin Rosand, Jeremy Goldwasser, Amisha D Dave, Tiarnan D. L. Keenan, Emily Y Chew, Dragomir Radev, Zhiyong Lu, Hua Xu, Qingyu Chen, Irene Li | 2023-12-09T09:14:24Z | cs.CL | This study introduces Ascle, a pioneering natural language processing (NLP)
toolkit designed for medical text generation. Ascle is tailored for biomedical
researchers and healthcare professionals with an easy-to-use, all-in-one
solution that requires minimal programming expertise. For the first time, Ascle
evaluates and provides interfaces for the latest pre-trained language models,
encompassing four advanced and challenging generative functions:
question-answering, text summarization, text simplification, and machine
translation. In addition, Ascle integrates 12 essential NLP functions, along
with query and search capabilities for clinical databases. The toolkit, its
models, and associated data are publicly available via
https://github.com/Yale-LILY/MedGen. | cs.CL | Ascle: A Python Natural Language Processing Toolkit for Medical Text Generation |
2023-11-28T07:47:25Z | http://arxiv.org/pdf/2311.16579v1 | http://arxiv.org/abs/2311.16579v1 | Xinhong Chen, Zongxi Li, Yaowei Wang, Haoran Xie, Jianping Wang, Qing Li | 2023-11-28T07:47:25Z | cs.CL | The study of causal relationships between emotions and causes in texts has
recently received much attention. Most works focus on extracting causally
related clauses from documents. However, none of these works has considered
that the causal relationships among the extracted emotion and cause clauses can
only be valid under some specific context clauses. To highlight the context in
such special causal relationships, we propose a new task to determine whether
or not an input pair of emotion and cause has a valid causal relationship under
different contexts and extract the specific context clauses that participate in
the causal relationship. Since the task is new for which no existing dataset is
available, we conduct manual annotation on a benchmark dataset to obtain the
labels for our tasks and the annotations of each context clause's type that can
also be used in some other applications. We adopt negative sampling to
construct the final dataset to balance the number of documents with and without
causal relationships. Based on the constructed dataset, we propose an
end-to-end multi-task framework, where we design two novel and general modules
to handle the two goals of our task. Specifically, we propose a context masking
module to extract the context clauses participating in the causal
relationships. We propose a prediction aggregation module to fine-tune the
prediction results according to whether the input emotion and causes depend on
specific context clauses. Results of extensive comparative experiments and
ablation studies demonstrate the effectiveness and generality of our proposed
framework. | cs.CL | Recognizing Conditional Causal Relationships about Emotions and Their Corresponding Conditions |
2023-11-28T05:44:06Z | http://arxiv.org/pdf/2312.03733v2 | http://arxiv.org/abs/2312.03733v2 | Maia Kotelanski, Robert Gallo, Ashwin Nayak, Thomas Savage | 2023-12-08T07:04:52Z | cs.CL, cs.AI | Large Language Models have difficulty communicating uncertainty, which is a
significant obstacle to applying LLMs to complex medical tasks. This study
evaluates methods to measure LLM confidence when suggesting a diagnosis for
challenging clinical vignettes. GPT4 was asked a series of challenging case
questions using Chain of Thought and Self Consistency prompting. Multiple
methods were investigated to assess model confidence and evaluated on their
ability to predict the models observed accuracy. The methods evaluated were
Intrinsic Confidence, SC Agreement Frequency and CoT Response Length. SC
Agreement Frequency correlated with observed accuracy, yielding a higher Area
under the Receiver Operating Characteristic Curve compared to Intrinsic
Confidence and CoT Length analysis. SC agreement is the most useful proxy for
model confidence, especially for medical diagnosis. Model Intrinsic Confidence
and CoT Response Length exhibit a weaker ability to differentiate between
correct and incorrect answers, preventing them from being reliable and
interpretable markers for model confidence. We conclude GPT4 has a limited
ability to assess its own diagnostic accuracy. SC Agreement Frequency is the
most useful method to measure GPT4 confidence. | cs.CL | Methods to Estimate Large Language Model Confidence |
2023-11-28T05:00:27Z | http://arxiv.org/pdf/2311.16522v4 | http://arxiv.org/abs/2311.16522v4 | Hao Pei, Si Lin, Chuanfu Li, Che Wang, Haoming Chen, Sizhe Li | 2024-01-27T07:28:42Z | cs.LG, cs.CL, eess.SP | To enhance the intelligence degree in operation and maintenance, a novel
method for fault detection in power grids is proposed. The proposed GNN-based
approach first identifies fault nodes through a specialized feature extraction
method coupled with a knowledge graph. By incorporating temporal data, the
method leverages the status of nodes from preceding and subsequent time periods
to help current fault detection. To validate the effectiveness of the node
features, a correlation analysis of the output features from each node was
conducted. The results from experiments show that this method can accurately
locate fault nodes in simulation scenarios with a remarkable accuracy.
Additionally, the graph neural network based feature modeling allows for a
qualitative examination of how faults spread across nodes, which provides
valuable insights for analyzing fault nodes. | cs.LG | Dynamic Fault Characteristics Evaluation in Power Grid |
2023-11-28T04:49:17Z | http://arxiv.org/pdf/2311.16509v2 | http://arxiv.org/abs/2311.16509v2 | Kazuki Yamauchi, Yusuke Ijima, Yuki Saito | 2023-12-27T14:41:23Z | cs.CL, cs.LG, cs.SD, eess.AS | We propose StyleCap, a method to generate natural language descriptions of
speaking styles appearing in speech. Although most of conventional techniques
for para-/non-linguistic information recognition focus on the category
classification or the intensity estimation of pre-defined labels, they cannot
provide the reasoning of the recognition result in an interpretable manner.
StyleCap is a first step towards an end-to-end method for generating
speaking-style prompts from speech, i.e., automatic speaking-style captioning.
StyleCap is trained with paired data of speech and natural language
descriptions. We train neural networks that convert a speech representation
vector into prefix vectors that are fed into a large language model (LLM)-based
text decoder. We explore an appropriate text decoder and speech feature
representation suitable for this new task. The experimental results demonstrate
that our StyleCap leveraging richer LLMs for the text decoder, speech
self-supervised learning (SSL) features, and sentence rephrasing augmentation
improves the accuracy and diversity of generated speaking-style captions.
Samples of speaking-style captions generated by our StyleCap are publicly
available. | cs.CL | StyleCap: Automatic Speaking-Style Captioning from Speech Based on Speech and Language Self-supervised Learning Models |
2023-11-28T04:07:34Z | http://arxiv.org/pdf/2311.16466v2 | http://arxiv.org/abs/2311.16466v2 | Minkyu Shin, Jin Kim | 2024-02-12T16:20:45Z | cs.HC, cs.AI, cs.CL | Although large language models (LLMs) are reshaping various aspects of human
life, our current understanding of their impacts remains somewhat constrained.
Here we investigate the impact of LLMs on human communication, using data on
consumer complaints in the financial industry. By employing an AI detection
tool on more than 820K complaints gathered by the Consumer Financial Protection
Bureau (CFPB), we find a sharp increase in the likely use of LLMs shortly after
the release of ChatGPT. Moreover, the likely LLM usage was positively
correlated with message persuasiveness (i.e., increased likelihood of obtaining
relief from financial firms). Computational linguistic analyses suggest that
the positive correlation may be explained by LLMs' enhancement of various
linguistic features. Based on the results of these observational studies, we
hypothesize that LLM usage may enhance a comprehensive set of linguistic
features, increasing message persuasiveness to receivers with heterogeneous
linguistic preferences (i.e., linguistic feature alignment). We test this
hypothesis in preregistered experiments and find support for it. As an instance
of early empirical demonstrations of LLM usage for enhancing persuasion, our
research highlights the transformative potential of LLMs in human
communication. | cs.HC | Large language models can enhance persuasion through linguistic feature alignment |
2023-11-28T03:23:20Z | http://arxiv.org/pdf/2312.03732v1 | http://arxiv.org/abs/2312.03732v1 | Damjan Kalajdzievski | 2023-11-28T03:23:20Z | cs.CL, cs.LG, I.2.7 | As large language models (LLMs) have become increasingly compute and memory
intensive, parameter-efficient fine-tuning (PEFT) methods are now a common
strategy to fine-tune LLMs. A popular PEFT method is Low-Rank Adapters (LoRA),
which adds trainable low-rank "adapters" to selected layers. Each adapter
consists of a low-rank matrix product, multiplicatively scaled by a
rank-dependent factor. This scaling factor, which divides adapters by a factor
of the rank, results in slowed learning and stunted performance for LoRA with
higher-rank adapters. Consequently, the use of LoRA in practice has generally
been limited to very low ranks. In this work, we study the impact of the
scaling factor on the learning process and prove that LoRA adapters should be
divided by a factor of the square root of the rank. Modifying LoRA with the
appropriate scaling factor, which we call the rank-stabilized LoRA (rsLoRA)
method, easily provides for a fine-tuning compute/performance trade-off, where
larger ranks can be used to trade off increased computational resources during
training for better fine-tuning performance, with no change in inference
computing cost. | cs.CL | A Rank Stabilization Scaling Factor for Fine-Tuning with LoRA |
2023-11-28T03:16:12Z | http://arxiv.org/pdf/2311.16452v1 | http://arxiv.org/abs/2311.16452v1 | Harsha Nori, Yin Tat Lee, Sheng Zhang, Dean Carignan, Richard Edgar, Nicolo Fusi, Nicholas King, Jonathan Larson, Yuanzhi Li, Weishung Liu, Renqian Luo, Scott Mayer McKinney, Robert Osazuwa Ness, Hoifung Poon, Tao Qin, Naoto Usuyama, Chris White, Eric Horvitz | 2023-11-28T03:16:12Z | cs.CL, I.2.7 | Generalist foundation models such as GPT-4 have displayed surprising
capabilities in a wide variety of domains and tasks. Yet, there is a prevalent
assumption that they cannot match specialist capabilities of fine-tuned models.
For example, most explorations to date on medical competency benchmarks have
leveraged domain-specific training, as exemplified by efforts on BioGPT and
Med-PaLM. We build on a prior study of GPT-4's capabilities on medical
challenge benchmarks in the absence of special training. Rather than using
simple prompting to highlight the model's out-of-the-box capabilities, we
perform a systematic exploration of prompt engineering. We find that prompting
innovation can unlock deeper specialist capabilities and show that GPT-4 easily
tops prior leading results for medical benchmarks. The prompting methods we
explore are general purpose, and make no specific use of domain expertise,
removing the need for expert-curated content. Our experimental design carefully
controls for overfitting during the prompt engineering process. We introduce
Medprompt, based on a composition of several prompting strategies. With
Medprompt, GPT-4 achieves state-of-the-art results on all nine of the benchmark
datasets in the MultiMedQA suite. The method outperforms leading specialist
models such as Med-PaLM 2 by a significant margin with an order of magnitude
fewer calls to the model. Steering GPT-4 with Medprompt achieves a 27%
reduction in error rate on the MedQA dataset over the best methods to date
achieved with specialist models and surpasses a score of 90% for the first
time. Beyond medical problems, we show the power of Medprompt to generalize to
other domains and provide evidence for the broad applicability of the approach
via studies of the strategy on exams in electrical engineering, machine
learning, philosophy, accounting, law, nursing, and clinical psychology. | cs.CL | Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine |
2023-11-28T02:51:13Z | http://arxiv.org/pdf/2311.16444v2 | http://arxiv.org/abs/2311.16444v2 | Takehiko Ohkawa, Takuma Yagi, Taichi Nishimura, Ryosuke Furuta, Atsushi Hashimoto, Yoshitaka Ushiku, Yoichi Sato | 2023-11-29T06:01:34Z | cs.CV, cs.CL | We propose a novel benchmark for cross-view knowledge transfer of dense video
captioning, adapting models from web instructional videos with exocentric views
to an egocentric view. While dense video captioning (predicting time segments
and their captions) is primarily studied with exocentric videos (e.g.,
YouCook2), benchmarks with egocentric videos are restricted due to data
scarcity. To overcome the limited video availability, transferring knowledge
from abundant exocentric web videos is demanded as a practical approach.
However, learning the correspondence between exocentric and egocentric views is
difficult due to their dynamic view changes. The web videos contain mixed views
focusing on either human body actions or close-up hand-object interactions,
while the egocentric view is constantly shifting as the camera wearer moves.
This necessitates the in-depth study of cross-view transfer under complex view
changes. In this work, we first create a real-life egocentric dataset (EgoYC2)
whose captions are shared with YouCook2, enabling transfer learning between
these datasets assuming their ground-truth is accessible. To bridge the view
gaps, we propose a view-invariant learning method using adversarial training in
both the pre-training and fine-tuning stages. While the pre-training is
designed to learn invariant features against the mixed views in the web videos,
the view-invariant fine-tuning further mitigates the view gaps between both
datasets. We validate our proposed method by studying how effectively it
overcomes the view change problem and efficiently transfers the knowledge to
the egocentric domain. Our benchmark pushes the study of the cross-view
transfer into a new task domain of dense video captioning and will envision
methodologies to describe egocentric videos in natural language. | cs.CV | Exo2EgoDVC: Dense Video Captioning of Egocentric Procedural Activities Using Web Instructional Videos |
2023-11-28T02:36:53Z | http://arxiv.org/pdf/2312.03731v6 | http://arxiv.org/abs/2312.03731v6 | Xingtong Yu, Chang Zhou, Yuan Fang, Xinming Zhang | 2024-02-22T06:35:04Z | cs.CL, cs.LG | Graphs can inherently model interconnected objects on the Web, thereby
facilitating a series of Web applications, such as web analyzing and content
recommendation. Recently, Graph Neural Networks (GNNs) have emerged as a
mainstream technique for graph representation learning. However, their efficacy
within an end-to-end supervised framework is significantly tied to the
availabilityof task-specific labels. To mitigate labeling costs and enhance
robustness in few-shot settings, pre-training on self-supervised tasks has
emerged as a promising method, while prompting has been proposed to further
narrow the objective gap between pretext and downstream tasks. Although there
has been some initial exploration of prompt-based learning on graphs, they
primarily leverage a single pretext task, resulting in a limited subset of
general knowledge that could be learned from the pre-training data. Hence, in
this paper, we propose MultiGPrompt, a novel multi-task pre-training and
prompting framework to exploit multiple pretext tasks for more comprehensive
pre-trained knowledge. First, in pre-training, we design a set of pretext
tokens to synergize multiple pretext tasks. Second, we propose a dual-prompt
mechanism consisting of composed and open prompts to leverage task-specific and
global pre-training knowledge, to guide downstream tasks in few-shot settings.
Finally, we conduct extensive experiments on six public datasets to evaluate
and analyze MultiGPrompt. | cs.CL | MultiGPrompt for Multi-Task Pre-Training and Prompting on Graphs |
2023-11-28T02:31:52Z | http://arxiv.org/pdf/2311.17086v1 | http://arxiv.org/abs/2311.17086v1 | Jian Ma, Chen Chen, Qingsong Xie, Haonan Lu | 2023-11-28T02:31:52Z | cs.CV, cs.CL | Text-to-image diffusion models are well-known for their ability to generate
realistic images based on textual prompts. However, the existing works have
predominantly focused on English, lacking support for non-English text-to-image
models. The most commonly used translation methods cannot solve the generation
problem related to language culture, while training from scratch on a specific
language dataset is prohibitively expensive. In this paper, we are inspired to
propose a simple plug-and-play language transfer method based on knowledge
distillation. All we need to do is train a lightweight MLP-like
parameter-efficient adapter (PEA) with only 6M parameters under teacher
knowledge distillation along with a small parallel data corpus. We are
surprised to find that freezing the parameters of UNet can still achieve
remarkable performance on the language-specific prompt evaluation set,
demonstrating that PEA can stimulate the potential generation ability of the
original UNet. Additionally, it closely approaches the performance of the
English text-to-image model on a general prompt evaluation set. Furthermore,
our adapter can be used as a plugin to achieve significant results in
downstream tasks in cross-lingual text-to-image generation. Code will be
available at: https://github.com/OPPO-Mente-Lab/PEA-Diffusion | cs.CV | PEA-Diffusion: Parameter-Efficient Adapter with Knowledge Distillation in non-English Text-to-Image Generation |
2023-11-28T02:01:25Z | http://arxiv.org/pdf/2311.16421v3 | http://arxiv.org/abs/2311.16421v3 | Yuhang Wang, Yanxu Zhu, Chao Kong, Shuyu Wei, Xiaoyuan Yi, Xing Xie, Jitao Sang | 2024-06-20T11:52:47Z | cs.CL, cs.CY | As the scaling of Large Language Models (LLMs) has dramatically enhanced
their capabilities, there has been a growing focus on the alignment problem to
ensure their responsible and ethical use. While existing alignment efforts
predominantly concentrate on universal values such as the HHH principle, the
aspect of culture, which is inherently pluralistic and diverse, has not
received adequate attention. This work introduces a new benchmark, CDEval,
aimed at evaluating the cultural dimensions of LLMs. CDEval is constructed by
incorporating both GPT-4's automated generation and human verification,
covering six cultural dimensions across seven domains. Our comprehensive
experiments provide intriguing insights into the culture of mainstream LLMs,
highlighting both consistencies and variations across different dimensions and
domains. The findings underscore the importance of integrating cultural
considerations in LLM development, particularly for applications in diverse
cultural settings. Through CDEval, we aim to broaden the horizon of LLM
alignment research by including cultural dimensions, thus providing a more
holistic framework for the future development and evaluation of LLMs. This
benchmark serves as a valuable resource for cultural studies in LLMs, paving
the way for more culturally aware and sensitive models. | cs.CL | CDEval: A Benchmark for Measuring the Cultural Dimensions of Large Language Models |
2023-11-27T23:03:01Z | http://arxiv.org/pdf/2311.16362v1 | http://arxiv.org/abs/2311.16362v1 | Ranjita Naik, Spencer Rarrick, Vishal Chowdhary | 2023-11-27T23:03:01Z | cs.CL | Recent advances in neural methods have led to substantial improvement in the
quality of Neural Machine Translation (NMT) systems. However, these systems
frequently produce translations with inaccurate gender (Stanovsky et al.,
2019), which can be traced to bias in training data. Saunders and Byrne (2020)
tackle this problem with a handcrafted dataset containing balanced gendered
profession words. By using this data to fine-tune an existing NMT model, they
show that gender bias can be significantly mitigated, albeit at the expense of
translation quality due to catastrophic forgetting. They recover some of the
lost quality with modified training objectives or additional models at
inference. We find, however, that simply supplementing the handcrafted dataset
with a random sample from the base model training corpus is enough to
significantly reduce the catastrophic forgetting. We also propose a novel
domain-adaptation technique that leverages in-domain data created with the
counterfactual data generation techniques proposed by Zmigrod et al. (2019) to
further improve accuracy on the WinoMT challenge test set without significant
loss in translation quality. We show its effectiveness in NMT systems from
English into three morphologically rich languages French, Spanish, and Italian.
The relevant dataset and code will be available at Github. | cs.CL | Reducing Gender Bias in Machine Translation through Counterfactual Data Generation |
2023-11-27T22:23:27Z | http://arxiv.org/pdf/2311.17076v3 | http://arxiv.org/abs/2311.17076v3 | Chancharik Mitra, Brandon Huang, Trevor Darrell, Roei Herzig | 2024-04-01T03:17:09Z | cs.CV, cs.AI, cs.CL, cs.LG | The combination of strong visual backbones and Large Language Model (LLM)
reasoning has led to Large Multimodal Models (LMMs) becoming the current
standard for a wide range of vision and language (VL) tasks. However, recent
research has shown that even the most advanced LMMs still struggle to capture
aspects of compositional visual reasoning, such as attributes and relationships
between objects. One solution is to utilize scene graphs (SGs)--a formalization
of objects and their relations and attributes that has been extensively used as
a bridge between the visual and textual domains. Yet, scene graph data requires
scene graph annotations, which are expensive to collect and thus not easily
scalable. Moreover, finetuning an LMM based on SG data can lead to catastrophic
forgetting of the pretraining objective. To overcome this, inspired by
chain-of-thought methods, we propose Compositional Chain-of-Thought (CCoT), a
novel zero-shot Chain-of-Thought prompting method that utilizes SG
representations in order to extract compositional knowledge from an LMM.
Specifically, we first generate an SG using the LMM, and then use that SG in
the prompt to produce a response. Through extensive experiments, we find that
the proposed CCoT approach not only improves LMM performance on several vision
and language VL compositional benchmarks but also improves the performance of
several popular LMMs on general multimodal benchmarks, without the need for
fine-tuning or annotated ground-truth SGs. Code:
https://github.com/chancharikmitra/CCoT | cs.CV | Compositional Chain-of-Thought Prompting for Large Multimodal Models |
2023-11-27T21:54:50Z | http://arxiv.org/pdf/2311.16338v1 | http://arxiv.org/abs/2311.16338v1 | Rob Grzywinski, Joshua D'Arcy, Rob Naidoff, Ashish Shukla, Alex Browne, Ren Gibbons, Brinnae Bent | 2023-11-27T21:54:50Z | cs.CL, cs.AI | Instruction-following language models demand robust methodologies for
information retrieval to augment instructions for question-answering
applications. A primary challenge is the resolution of coreferences in the
context of chunking strategies for long documents. The critical barrier to
experimentation of handling coreferences is a lack of open source datasets,
specifically in question-answering tasks that require coreference resolution.
In this work we present our Coreference Resolution in Question-Answering
(CRaQAn) dataset, an open-source dataset that caters to the nuanced information
retrieval requirements of coreference resolution in question-answering tasks by
providing over 250 question-answer pairs containing coreferences. To develop
this dataset, we developed a novel approach for creating high-quality datasets
using an instruction-following model (GPT-4) and a Recursive Criticism and
Improvement Loop. | cs.CL | Releasing the CRaQAn (Coreference Resolution in Question-Answering): An open-source dataset and dataset creation methodology using instruction-following models |
2023-11-27T21:01:21Z | http://arxiv.org/pdf/2312.03730v2 | http://arxiv.org/abs/2312.03730v2 | Tahniat Khan, Mizanur Rahman, Veronica Chatrath, Oluwanifemi Bamgbose, Shaina Raza | 2023-12-08T19:42:35Z | cs.CL, cs.AI | In today's technologically driven world, the spread of fake news,
particularly during crucial events such as elections, presents an increasing
challenge to the integrity of information. To address this challenge, we
introduce FakeWatch ElectionShield, an innovative framework carefully designed
to detect fake news. We have created a novel dataset of North American
election-related news articles through a blend of advanced language models
(LMs) and thorough human verification, for precision and relevance. We propose
a model hub of LMs for identifying fake news. Our goal is to provide the
research community with adaptable and accurate classification models in
recognizing the dynamic nature of misinformation. Extensive evaluation of fake
news classifiers on our dataset and a benchmark dataset shows our that while
state-of-the-art LMs slightly outperform the traditional ML models, classical
models are still competitive with their balance of accuracy, explainability,
and computational efficiency. This research sets the foundation for future
studies to address misinformation related to elections. | cs.CL | FakeWatch ElectionShield: A Benchmarking Framework to Detect Fake News for Credible US Elections |
2023-11-27T20:33:54Z | http://arxiv.org/pdf/2311.16302v1 | http://arxiv.org/abs/2311.16302v1 | Anusha Sabbineni, Nikhil Anand, Maria Minakova | 2023-11-27T20:33:54Z | cs.LG, cs.CL | While data selection methods have been studied extensively in active
learning, data pruning, and data augmentation settings, there is little
evidence for the efficacy of these methods in industry scale settings,
particularly in low-resource languages. Our work presents ways of assessing
prospective training examples in those settings for their "usefulness" or
"difficulty". We also demonstrate how these measures can be used in selecting
important examples for training supervised machine learning models. We
primarily experiment with entropy and Error L2-Norm (EL2N) scores. We use these
metrics to curate high quality datasets from a large pool of \textit{Weak
Signal Labeled} data, which assigns no-defect high confidence hypotheses during
inference as ground truth labels. We then conduct training data augmentation
experiments using these de-identified datasets and demonstrate that score-based
selection can result in a 2% decrease in semantic error rate and 4%-7% decrease
in domain classification error rate when compared to the baseline technique of
random selection. | cs.LG | Comprehensive Benchmarking of Entropy and Margin Based Scoring Metrics for Data Selection |
2023-11-27T20:19:22Z | http://arxiv.org/pdf/2311.16298v1 | http://arxiv.org/abs/2311.16298v1 | Nikhil Anand, Joshua Tan, Maria Minakova | 2023-11-27T20:19:22Z | cs.LG, cs.CL | Modern ML systems ingest data aggregated from diverse sources, such as
synthetic, human-annotated, and live customer traffic. Understanding
\textit{which} examples are important to the performance of a learning
algorithm is crucial for efficient model training. Recently, a growing body of
literature has given rise to various "influence scores," which use training
artifacts such as model confidence or checkpointed gradients to identify
important subsets of data. However, these methods have primarily been developed
in computer vision settings, and it remains unclear how well they generalize to
language-based tasks using pretrained models.
In this paper, we explore the applicability of influence scores in language
classification tasks. We evaluate a diverse subset of these scores on the SNLI
dataset by quantifying accuracy changes in response to pruning training data
through random and influence-score-based sampling. We then stress-test one of
the scores -- "variance of gradients" (VoG) from Agarwal et al. (2022) -- in an
NLU model stack that was exposed to dynamic user speech patterns in a voice
assistant type of setting. Our experiments demonstrate that in many cases,
encoder-based language models can be finetuned on roughly 50% of the original
data without degradation in performance metrics. Along the way, we summarize
lessons learned from applying out-of-the-box implementations of influence
scores, quantify the effects of noisy and class-imbalanced data, and offer
recommendations on score-based sampling for better accuracy and training
efficiency. | cs.LG | Influence Scores at Scale for Efficient Language Data Sampling |
2023-11-27T20:10:13Z | http://arxiv.org/pdf/2311.16292v1 | http://arxiv.org/abs/2311.16292v1 | Kevin Wang, Seth Akins, Abdallah Mohammed, Ramon Lawrence | 2023-11-27T20:10:13Z | cs.CY, cs.CL | Generative AI systems such as ChatGPT have a disruptive effect on learning
and assessment. Computer science requires practice to develop skills in problem
solving and programming that are traditionally developed using assignments.
Generative AI has the capability of completing these assignments for students
with high accuracy, which dramatically increases the potential for academic
integrity issues and students not achieving desired learning outcomes. This
work investigates the performance of ChatGPT by evaluating it across three
courses (CS1,CS2,databases). ChatGPT completes almost all introductory
assessments perfectly. Existing detection methods, such as MOSS and JPlag
(based on similarity metrics) and GPTzero (AI detection), have mixed success in
identifying AI solutions. Evaluating instructors and teaching assistants using
heuristics to distinguish between student and AI code shows that their
detection is not sufficiently accurate. These observations emphasize the need
for adapting assessments and improved detection methods. | cs.CY | Student Mastery or AI Deception? Analyzing ChatGPT's Assessment Proficiency and Evaluating Detection Strategies |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.