id
stringlengths
33
34
title
stringlengths
6
218
authors
stringlengths
5
6.98k
categories
stringlengths
5
188
summary
stringlengths
34
3.26k
link_pdf
stringlengths
33
34
updated
timestamp[s]
published
timestamp[s]
embeddings
dict
http://arxiv.org/abs/2310.01444v2
Adapting LLM Agents Through Communication
Kuan Wang, Yadong Lu, Michael Santacroce, Yeyun Gong, Chao Zhang, Yelong Shen
cs.CL, cs.AI
Recent advancements in large language models (LLMs) have shown potential for human-like agents. To help these agents adapt to new tasks without extensive human supervision, we propose the Learning through Communication (LTC) paradigm, a novel training approach enabling LLM agents to improve continuously through interac...
http://arxiv.org/pdf/2310.01444v2
2023-10-10T05:38:05
2023-10-01T07:50:30
{ "title": [ -0.06817627, -0.01777649, -0.04058838, -0.018981934, -0.038726807, 0.007648468, -0.05239868, 0.010536194, -0.0027313232, 0.023971558, -0.028900146, -0.0054740906, -0.007621765, -0.023223877, -0.0128479, -0.0024147034, -0.0039024353, ...
http://arxiv.org/abs/2310.00603v1
Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactuals
Yair Gat, Nitay Calderon, Amir Feder, Alexander Chapanin, Amit Sharma, Roi Reichart
cs.CL, cs.AI
Causal explanations of the predictions of NLP systems are essential to ensure safety and establish trust. Yet, existing methods often fall short of explaining model predictions effectively or efficiently and are often model-specific. In this paper, we address model-agnostic explanations, proposing two approaches for co...
http://arxiv.org/pdf/2310.00603v1
2023-10-01T07:31:04
2023-10-01T07:31:04
{ "title": [ -0.02619934, -0.015380859, -0.018554688, -0.040130615, -0.039245605, 0.0025081635, -0.07208252, -0.027755737, 0.0019235611, 0.010154724, -0.041992188, 0.014183044, -0.016403198, -0.0076789856, -0.03677368, -0.010528564, -0.027557373, ...
http://arxiv.org/abs/2310.00602v2
Wavelet Scattering Transform for Improving Generalization in Low-Resourced Spoken Language Identification
Spandan Dey, Premjeet Singh, Goutam Saha
eess.AS, cs.CL
Commonly used features in spoken language identification (LID), such as mel-spectrogram or MFCC, lose high-frequency information due to windowing. The loss further increases for longer temporal contexts. To improve generalization of the low-resourced LID systems, we investigate an alternate feature representation, wave...
http://arxiv.org/pdf/2310.00602v2
2023-10-03T08:00:53
2023-10-01T07:31:00
{ "title": [ 0.0115356445, 0.004627228, -0.025024414, -0.067871094, 0.008918762, -0.0019989014, -0.04260254, -0.0050849915, -0.017791748, 0.050689697, -0.00036025047, 0.00818634, 0.022735596, -0.025299072, -0.031463623, -0.006286621, -0.00083732605, ...
http://arxiv.org/abs/2310.00598v1
A Novel Computational and Modeling Foundation for Automatic Coherence Assessment
Aviya Maimon, Reut Tsarfaty
cs.CL, cs.AI
Coherence is an essential property of well-written texts, that refers to the way textual units relate to one another. In the era of generative AI, coherence assessment is essential for many NLP tasks; summarization, generation, long-form question-answering, and more. However, in NLP {coherence} is an ill-defined notion...
http://arxiv.org/pdf/2310.00598v1
2023-10-01T07:06:17
2023-10-01T07:06:17
{ "title": [ -0.024291992, 0.013000488, -0.04537964, -0.059387207, 0.009025574, 0.0016880035, -0.05517578, -0.019439697, 0.013793945, 0.057281494, 0.026123047, -0.023712158, 0.04220581, -0.033325195, -0.05734253, 0.037872314, -0.015686035, 0.0179...
http://arxiv.org/abs/2310.00597v1
A Task-oriented Dialog Model with Task-progressive and Policy-aware Pre-training
Lucen Zhong, Hengtong Lu, Caixia Yuan, Xiaojie Wang, Jiashen Sun, Ke Zeng, Guanglu Wan
cs.CL
Pre-trained conversation models (PCMs) have achieved promising progress in recent years. However, existing PCMs for Task-oriented dialog (TOD) are insufficient for capturing the sequential nature of the TOD-related tasks, as well as for learning dialog policy information. To alleviate these problems, this paper propose...
http://arxiv.org/pdf/2310.00597v1
2023-10-01T07:06:02
2023-10-01T07:06:02
{ "title": [ 0.026489258, 0.00680542, -0.05621338, -0.105529785, 0.021697998, -0.023406982, -0.047332764, 0.016799927, -0.011009216, 0.036346436, 0.01725769, -0.008850098, -0.007171631, -0.029922485, -0.03527832, -0.007095337, -0.0025634766, 0.00...
http://arxiv.org/abs/2310.00578v1
Nine-year-old children outperformed ChatGPT in emotion: Evidence from Chinese writing
Siyi Cao, Tongquan Zhou, Siruo Zhou
cs.CL
ChatGPT has been demonstrated to possess significant capabilities in generating intricate, human-like text, and recent studies have established that its performance in theory of mind tasks is comparable to that of a nine-year-old child. However, it remains uncertain whether ChatGPT surpasses nine-year-old children in C...
http://arxiv.org/pdf/2310.00578v1
2023-10-01T05:37:55
2023-10-01T05:37:55
{ "title": [ -0.015853882, -0.020996094, -0.011695862, 0.014167786, -0.03677368, -0.01499939, 0.009284973, -0.0036010742, -0.009788513, 0.032836914, -0.028762817, 0.015434265, -0.001203537, -0.016525269, -0.040496826, -0.021530151, -0.035491943, ...
http://arxiv.org/abs/2310.00576v1
GrowLength: Accelerating LLMs Pretraining by Progressively Growing Training Length
Hongye Jin, Xiaotian Han, Jingfeng Yang, Zhimeng Jiang, Chia-Yuan Chang, Xia Hu
cs.CL, cs.LG
The evolving sophistication and intricacies of Large Language Models (LLMs) yield unprecedented advancements, yet they simultaneously demand considerable computational resources and incur significant costs. To alleviate these challenges, this paper introduces a novel, simple, and effective method named ``\growlength'' ...
http://arxiv.org/pdf/2310.00576v1
2023-10-01T05:25:24
2023-10-01T05:25:24
{ "title": [ -0.03564453, -0.0152282715, -0.05496216, -0.05227661, -0.040740967, -0.0003745556, -0.066223145, 0.0012893677, -0.04135132, 0.021713257, -0.014762878, 0.045898438, -0.00087070465, -0.032989502, 0.0231781, -0.0065574646, -0.00015699863, ...
http://arxiv.org/abs/2310.00572v1
Colloquial Persian POS (CPPOS) Corpus: A Novel Corpus for Colloquial Persian Part of Speech Tagging
Leyla Rabiei, Farzaneh Rahmani, Mohammad Khansari, Zeinab Rajabi, Moein Salimi
cs.CL
Introduction: Part-of-Speech (POS) Tagging, the process of classifying words into their respective parts of speech (e.g., verb or noun), is essential in various natural language processing applications. POS tagging is a crucial preprocessing task for applications like machine translation, question answering, sentiment ...
http://arxiv.org/pdf/2310.00572v1
2023-10-01T05:06:33
2023-10-01T05:06:33
{ "title": [ -0.006587982, 0.016174316, -0.03137207, -0.024642944, 0.0039711, -0.00699234, -0.019500732, -0.011817932, 0.0051879883, 0.03878784, -0.05947876, 0.0045433044, 0.035888672, -0.03414917, -0.002199173, -0.00009763241, -0.028686523, 0.05...
http://arxiv.org/abs/2310.00566v2
Empowering Many, Biasing a Few: Generalist Credit Scoring through Large Language Models
Duanyu Feng, Yongfu Dai, Jimin Huang, Yifang Zhang, Qianqian Xie, Weiguang Han, Alejandro Lopez-Lira, Hao Wang
cs.LG, cs.AI, cs.CL, cs.CY
In the financial industry, credit scoring is a fundamental element, shaping access to credit and determining the terms of loans for individuals and businesses alike. Traditional credit scoring methods, however, often grapple with challenges such as narrow knowledge scope and isolated evaluation of credit tasks. Our wor...
http://arxiv.org/pdf/2310.00566v2
2023-11-14T07:30:18
2023-10-01T03:50:34
{ "title": [ -0.011360168, -0.013893127, -0.031158447, -0.042755127, -0.01739502, -0.008926392, -0.032409668, -0.010353088, 0.021820068, 0.024246216, -0.041656494, -0.008102417, 0.0052452087, -0.016448975, -0.042175293, -0.02154541, -0.02609253, ...
http://arxiv.org/abs/2310.00552v1
Siamese Representation Learning for Unsupervised Relation Extraction
Guangxin Zhang, Shu Chen
cs.CL, cs.AI
Unsupervised relation extraction (URE) aims at discovering underlying relations between named entity pairs from open-domain plain text without prior information on relational distribution. Existing URE models utilizing contrastive learning, which attract positive samples and repulse negative samples to promote better s...
http://arxiv.org/pdf/2310.00552v1
2023-10-01T02:57:43
2023-10-01T02:57:43
{ "title": [ 0.0072288513, 0.007003784, -0.061309814, -0.08569336, 0.01651001, -0.0073928833, -0.04562378, 0.002128601, 0.014945984, 0.032196045, -0.00374794, 0.010147095, -0.015823364, -0.010559082, -0.055114746, 0.003232956, -0.009315491, 0.048...
http://arxiv.org/abs/2310.00535v2
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Yuandong Tian, Yiping Wang, Zhenyu Zhang, Beidi Chen, Simon Du
cs.LG, cs.AI, cs.CL
We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures. This is achieved by integrating out the self-attention layer in Transformers, producing a modified dynamics of MLP layers only. JoMA removes unrealistic assumption...
http://arxiv.org/pdf/2310.00535v2
2023-10-03T04:23:26
2023-10-01T01:21:35
{ "title": [ -0.0035705566, 0.01638794, -0.027069092, -0.047576904, 0.020507812, -0.004760742, 0.0019569397, -0.012634277, -0.014083862, 0.02746582, -0.012619019, 0.012168884, 0.0021419525, -0.026809692, -0.024490356, 0.011672974, -0.01940918, 0....
http://arxiv.org/abs/2310.00533v2
SELF: Language-Driven Self-Evolution for Large Language Model
Jianqiao Lu, Wanjun Zhong, Wenyong Huang, Yufei Wang, Fei Mi, Baojun Wang, Weichao Wang, Lifeng Shang, Qun Liu
cs.CL, cs.AI, cs.LG
Large Language Models (LLMs) have showcased remarkable versatility across diverse domains. However, the pathway toward autonomous model development, a cornerstone for achieving human-level learning and advancing autonomous AI, remains largely uncharted. We introduce an innovative approach, termed "SELF" (Self-Evolution...
http://arxiv.org/pdf/2310.00533v2
2023-10-07T09:57:58
2023-10-01T00:52:24
{ "title": [ -0.040924072, -0.010421753, -0.041931152, -0.056427002, 0.007835388, 0.009239197, -0.03768921, 0.02355957, -0.001868248, 0.04168701, 0.0075683594, 0.010383606, -0.0010404587, -0.018539429, -0.046691895, 0.017456055, -0.0064430237, 0....
http://arxiv.org/abs/2310.04438v1
A Brief History of Prompt: Leveraging Language Models
Golam Md Muktadir
cs.CL, cs.AI
This paper presents a comprehensive exploration of the evolution of prompt engineering and generation in the field of natural language processing (NLP). Starting from the early language models and information retrieval systems, we trace the key developments that have shaped prompt engineering over the years. The introd...
http://arxiv.org/pdf/2310.04438v1
2023-09-30T22:27:37
2023-09-30T22:27:37
{ "title": [ -0.02645874, -0.025390625, -0.034484863, -0.047088623, -0.046417236, -0.006286621, -0.07714844, -0.0027217865, 0.0032920837, 0.027801514, -0.038146973, -0.026275635, 0.032409668, -0.031311035, -0.031433105, 0.011581421, -0.0062561035, ...
http://arxiv.org/abs/2310.00492v1
From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning
Xuansheng Wu, Wenlin Yao, Jianshu Chen, Xiaoman Pan, Xiaoyang Wang, Ninghao Liu, Dong Yu
cs.CL, cs.AI, cs.LG
Large Language Models (LLMs) have achieved remarkable success, demonstrating powerful instruction-following capabilities across diverse tasks. Instruction fine-tuning is critical in enabling LLMs to align with user intentions and effectively follow instructions. In this work, we investigate how instruction fine-tuning ...
http://arxiv.org/pdf/2310.00492v1
2023-09-30T21:16:05
2023-09-30T21:16:05
{ "title": [ -0.06335449, -0.009185791, -0.01940918, -0.02809143, -0.030349731, 0.0012102127, -0.066467285, -0.0037403107, -0.025238037, 0.01737976, -0.021774292, 0.029144287, -0.02709961, -0.021484375, -0.019546509, -0.010307312, -0.013771057, 0...
http://arxiv.org/abs/2310.00486v1
It HAS to be Subjective: Human Annotator Simulation via Zero-shot Density Estimation
Wen Wu, Wenlin Chen, Chao Zhang, Philip C. Woodland
cs.CL, cs.HC, cs.LG
Human annotator simulation (HAS) serves as a cost-effective substitute for human evaluation such as data annotation and system assessment. Human perception and behaviour during human evaluation exhibit inherent variability due to diverse cognitive processes and subjective interpretations, which should be taken into acc...
http://arxiv.org/pdf/2310.00486v1
2023-09-30T20:54:59
2023-09-30T20:54:59
{ "title": [ 0.006259918, 0.009933472, -0.0015687943, -0.020904541, -0.029220581, -0.024108887, -0.017593384, -0.039489746, 0.0079956055, 0.03387451, -0.016479492, 0.00242424, 0.010803223, -0.027511597, -0.043182373, -0.0031852722, 0.010604858, 0...
http://arxiv.org/abs/2310.01441v1
UPAR: A Kantian-Inspired Prompting Framework for Enhancing Large Language Model Capabilities
Hejia Geng, Boxun Xu, Peng Li
cs.CL, cs.AI
Large Language Models (LLMs) have demonstrated impressive inferential capabilities, with numerous research endeavors devoted to enhancing this capacity through prompting. Despite these efforts, a unified epistemological foundation is still conspicuously absent. Drawing inspiration from Kant's a priori philosophy, we pr...
http://arxiv.org/pdf/2310.01441v1
2023-09-30T20:18:50
2023-09-30T20:18:50
{ "title": [ -0.019729614, -0.0024929047, -0.041748047, -0.07763672, 0.0014228821, 0.009635925, -0.07385254, 0.0041656494, 0.032714844, 0.024490356, -0.017425537, -0.008361816, 0.020828247, -0.024169922, -0.048858643, 0.018463135, -0.008094788, 0...
http://arxiv.org/abs/2310.00448v1
Question-Answering Model for Schizophrenia Symptoms and Their Impact on Daily Life using Mental Health Forums Data
Christian Internò, Eloisa Ambrosini
cs.LG, cs.CL
In recent years, there is strong emphasis on mining medical data using machine learning techniques. A common problem is to obtain a noiseless set of textual documents, with a relevant content for the research question, and developing a Question Answering (QA) model for a specific medical field. The purpose of this pape...
http://arxiv.org/pdf/2310.00448v1
2023-09-30T17:50:50
2023-09-30T17:50:50
{ "title": [ 0.025680542, 0.007865906, -0.04321289, -0.059631348, 0.029037476, -0.034454346, 0.0014572144, 0.00037431717, 0.014312744, 0.02911377, 0.011505127, -0.010269165, 0.0075149536, -0.047058105, -0.036224365, 0.013244629, 0.008476257, 0.00...
http://arxiv.org/abs/2310.01440v1
The Many Voices of Duying: Revisiting the Disputed Essays Between Lu Xun and Zhou Zuoren
Xin Xie, Jiangqiong Li, Haining Wang
cs.CL
Lu Xun and Zhou Zuoren stand as two of the most influential writers in modern Chinese literature. Beyond their familial ties as brothers, they were also intimate collaborators during the nascent stages of their writing careers. This research employs quantitative methods to revisit three disputed essays pseudonymously p...
http://arxiv.org/pdf/2310.01440v1
2023-09-30T17:41:39
2023-09-30T17:41:39
{ "title": [ -0.00048971176, 0.0029964447, -0.010406494, 0.021194458, -0.015197754, -0.014266968, 0.02609253, -0.045288086, -0.0670166, 0.021652222, 0.031433105, 0.02420044, -0.011871338, -0.05593872, 0.026992798, -0.02923584, -0.01876831, 0.0208...
http://arxiv.org/abs/2310.00436v1
Enhancing Representation Generalization in Authorship Identification
Haining Wang
cs.CL
Authorship identification ascertains the authorship of texts whose origins remain undisclosed. That authorship identification techniques work as reliably as they do has been attributed to the fact that authorial style is properly captured and represented. Although modern authorship identification methods have evolved s...
http://arxiv.org/pdf/2310.00436v1
2023-09-30T17:11:00
2023-09-30T17:11:00
{ "title": [ -0.015098572, 0.022445679, -0.034454346, -0.041259766, 0.0025100708, -0.0060691833, -0.024978638, -0.02432251, 0.00027680397, 0.047180176, -0.016967773, 0.030471802, 0.016937256, -0.0038547516, -0.039154053, 0.008155823, -0.017227173, ...
http://arxiv.org/abs/2310.00410v1
Open-Domain Dialogue Quality Evaluation: Deriving Nugget-level Scores from Turn-level Scores
Rikiya Takehi, Akihisa Watanabe, Tetsuya Sakai
cs.CL, cs.IR
Existing dialogue quality evaluation systems can return a score for a given system turn from a particular viewpoint, e.g., engagingness. However, to improve dialogue systems by locating exactly where in a system turn potential problems lie, a more fine-grained evaluation may be necessary. We therefore propose an evalua...
http://arxiv.org/pdf/2310.00410v1
2023-09-30T15:14:50
2023-09-30T15:14:50
{ "title": [ 0.015838623, 0.011947632, -0.030014038, -0.06121826, 0.020050049, -0.01184845, -0.019088745, -0.018493652, -0.0026397705, 0.01512146, -0.006286621, 0.0051879883, -0.025772095, -0.027252197, -0.0446167, -0.018692017, -0.021957397, 0.0...
http://arxiv.org/abs/2310.00385v1
Dynamic Demonstrations Controller for In-Context Learning
Fei Zhao, Taotian Pang, Zhen Wu, Zheng Ma, Shujian Huang, Xinyu Dai
cs.CL, cs.AI
In-Context Learning (ICL) is a new paradigm for natural language processing (NLP), where a large language model (LLM) observes a small number of demonstrations and a test instance as its input, and directly makes predictions without updating model parameters. Previous studies have revealed that ICL is sensitive to the ...
http://arxiv.org/pdf/2310.00385v1
2023-09-30T14:04:22
2023-09-30T14:04:22
{ "title": [ 0.03289795, 0.013069153, -0.038604736, -0.055877686, 0.037902832, -0.038330078, 0.026062012, -0.011413574, -0.01739502, 0.04498291, -0.03564453, 0.0017671585, -0.002023697, -0.0054855347, -0.019424438, -0.010894775, 0.00970459, 0.020...
http://arxiv.org/abs/2310.00378v3
Measuring Value Understanding in Language Models through Discriminator-Critique Gap
Zhaowei Zhang, Fengshuo Bai, Jun Gao, Yaodong Yang
cs.CL, cs.AI, cs.CY, I.2.m; K.4.m
Recent advancements in Large Language Models (LLMs) have heightened concerns about their potential misalignment with human values. However, evaluating their grasp of these values is complex due to their intricate and adaptable nature. We argue that truly understanding values in LLMs requires considering both "know what...
http://arxiv.org/pdf/2310.00378v3
2023-10-19T03:18:58
2023-09-30T13:47:55
{ "title": [ -0.041137695, 0.0051879883, -0.023635864, -0.01235199, -0.021697998, -0.005695343, -0.047180176, -0.039154053, 0.032470703, -0.016616821, -0.038116455, -0.036315918, -0.022064209, -0.039093018, -0.048034668, -0.037872314, -0.030944824, ...
http://arxiv.org/abs/2310.00367v1
AutomaTikZ: Text-Guided Synthesis of Scientific Vector Graphics with TikZ
Jonas Belouadi, Anne Lauscher, Steffen Eger
cs.CL, cs.CV
Generating bitmap graphics from text has gained considerable attention, yet for scientific figures, vector graphics are often preferred. Given that vector graphics are typically encoded using low-level graphics primitives, generating them directly is difficult. To address this, we propose the use of TikZ, a well-known ...
http://arxiv.org/pdf/2310.00367v1
2023-09-30T13:15:49
2023-09-30T13:15:49
{ "title": [ -0.051483154, -0.003993988, -0.027862549, -0.047912598, -0.013946533, 0.01966858, -0.0597229, -0.018692017, -0.020645142, 0.04623413, -0.011329651, 0.03894043, 0.016647339, -0.014213562, -0.023010254, 0.020431519, 0.021087646, -0.024...
http://arxiv.org/abs/2310.00355v1
Gaze-Driven Sentence Simplification for Language Learners: Enhancing Comprehension and Readability
Taichi Higasa, Keitaro Tanaka, Qi Feng, Shigeo Morishima
cs.CL, cs.HC
Language learners should regularly engage in reading challenging materials as part of their study routine. Nevertheless, constantly referring to dictionaries is time-consuming and distracting. This paper presents a novel gaze-driven sentence simplification system designed to enhance reading comprehension while maintain...
http://arxiv.org/pdf/2310.00355v1
2023-09-30T12:18:31
2023-09-30T12:18:31
{ "title": [ -0.031082153, -0.019454956, -0.012176514, -0.024246216, -0.040222168, 0.0064811707, -0.035949707, -0.01550293, 0.0023498535, 0.04421997, -0.014572144, 0.006778717, 0.004638672, -0.0076904297, -0.060180664, -0.0051078796, 0.0005478859, ...
http://arxiv.org/abs/2310.00347v2
Unlocking Bias Detection: Leveraging Transformer-Based Models for Content Analysis
Shaina Raza, Oluwanifemi Bamgbose, Veronica Chatrath, Shardul Ghuge, Yan Sidyakin, Abdullah Y Muaad
cs.CL, cs.AI
Bias detection in text is imperative due to its role in reinforcing negative stereotypes, disseminating misinformation, and influencing decisions. Current language models often fall short in generalizing beyond their training sets. In response, we introduce the Contextualized Bi-Directional Dual Transformer (CBDT) Clas...
http://arxiv.org/pdf/2310.00347v2
2023-10-14T12:22:28
2023-09-30T12:06:04
{ "title": [ 0.00014603138, 0.018203735, -0.045318604, -0.028640747, 0.028030396, -0.019363403, -0.031921387, -0.054718018, 0.001958847, 0.023147583, -0.041625977, 0.0055236816, 0.02017212, -0.014289856, -0.026733398, -0.0017271042, -0.022842407, ...
http://arxiv.org/abs/2310.00322v2
Red Teaming Game: A Game-Theoretic Framework for Red Teaming Language Models
Chengdong Ma, Ziran Yang, Minquan Gao, Hai Ci, Jun Gao, Xuehai Pan, Yaodong Yang
cs.CL, cs.GT
Deployable Large Language Models (LLMs) must conform to the criterion of helpfulness and harmlessness, thereby achieving consistency between LLMs outputs and human values. Red-teaming techniques constitute a critical way towards this criterion. Existing work rely solely on manual red team designs and heuristic adversar...
http://arxiv.org/pdf/2310.00322v2
2023-10-10T09:00:40
2023-09-30T09:35:50
{ "title": [ -0.039855957, -0.013336182, -0.019363403, -0.016677856, -0.024169922, -0.006462097, -0.026748657, -0.021484375, -0.036346436, 0.027572632, -0.005382538, 0.015167236, -0.031311035, -0.022125244, -0.01676941, -0.022994995, -0.017959595, ...
http://arxiv.org/abs/2310.00313v2
In-Context Learning in Large Language Models: A Neuroscience-inspired Analysis of Representations
Safoora Yousefi, Leo Betthauser, Hosein Hasanbeig, Akanksha Saran, Raphaël Millière, Ida Momennejad
cs.CL
Large language models (LLMs) exhibit remarkable performance improvement through in-context learning (ICL) by leveraging task-specific examples in the input. However, the mechanisms behind this improvement remain elusive. In this work, we investigate embeddings and attention representations in Llama-2 70B and Vicuna 13B...
http://arxiv.org/pdf/2310.00313v2
2023-10-18T08:53:46
2023-09-30T09:01:35
{ "title": [ 0.00025343895, 0.0016565323, -0.022445679, -0.049621582, 0.0010976791, -0.015686035, -0.014289856, 0.0027236938, -0.0037136078, 0.03768921, -0.016220093, 0.003124237, 0.012435913, -0.016616821, -0.039794922, 0.00076532364, -0.0044898987,...
http://arxiv.org/abs/2310.00305v1
Towards LLM-based Fact Verification on News Claims with a Hierarchical Step-by-Step Prompting Method
Xuan Zhang, Wei Gao
cs.CL
While large pre-trained language models (LLMs) have shown their impressive capabilities in various NLP tasks, they are still under-explored in the misinformation domain. In this paper, we examine LLMs with in-context learning (ICL) for news claim verification, and find that only with 4-shot demonstration examples, the ...
http://arxiv.org/pdf/2310.00305v1
2023-09-30T08:33:04
2023-09-30T08:33:04
{ "title": [ -0.0390625, -0.004722595, -0.05831909, -0.082092285, -0.010375977, 0.0118637085, -0.06439209, 0.018112183, -0.037750244, 0.023239136, -0.010887146, 0.048095703, -0.015655518, -0.01966858, -0.012245178, -0.007003784, -0.0362854, 0.026...
http://arxiv.org/abs/2310.00299v2
RelBERT: Embedding Relations with Language Models
Asahi Ushio, Jose Camacho-Collados, Steven Schockaert
cs.CL
Many applications need access to background knowledge about how different concepts and entities are related. Although Knowledge Graphs (KG) and Large Language Models (LLM) can address this need to some extent, KGs are inevitably incomplete and their relational schema is often too coarse-grained, while LLMs are ineffici...
http://arxiv.org/pdf/2310.00299v2
2023-10-08T16:22:26
2023-09-30T08:15:36
{ "title": [ -0.023284912, -0.022445679, -0.02444458, -0.080566406, 0.0023765564, 0.03643799, -0.06738281, 0.00034332275, 0.027786255, 0.010261536, -0.017440796, -0.00084352493, 0.0076026917, -0.015975952, -0.041381836, 0.008682251, -0.02848816, ...
http://arxiv.org/abs/2310.00297v2
Understanding In-Context Learning from Repetitions
Jianhao Yan, Jin Xu, Chiyu Song, Chenming Wu, Yafu Li, Yue Zhang
cs.CL
This paper explores the elusive mechanism underpinning in-context learning in Large Language Models (LLMs). Our work provides a novel perspective by examining in-context learning via the lens of surface repetitions. We quantitatively investigate the role of surface features in text generation, and empirically establish...
http://arxiv.org/pdf/2310.00297v2
2023-10-10T01:22:47
2023-09-30T08:13:49
{ "title": [ 0.025436401, -0.0059051514, -0.021728516, -0.073791504, 0.020736694, -0.017730713, -0.0033226013, -0.014480591, -0.009155273, 0.03503418, -0.028427124, 0.018325806, -0.014884949, -0.010093689, -0.044281006, 0.0013799667, -0.00894165, ...
http://arxiv.org/abs/2310.00280v1
Corex: Pushing the Boundaries of Complex Reasoning through Multi-Model Collaboration
Qiushi Sun, Zhangyue Yin, Xiang Li, Zhiyong Wu, Xipeng Qiu, Lingpeng Kong
cs.AI, cs.CL
Large Language Models (LLMs) are evolving at an unprecedented pace and have exhibited considerable capability in the realm of natural language processing (NLP) with world knowledge. Benefiting from ultra-large-scale training corpora, a single LLM can manage typical NLP tasks competently. However, its performance in exe...
http://arxiv.org/pdf/2310.00280v1
2023-09-30T07:11:39
2023-09-30T07:11:39
{ "title": [ -0.044708252, -0.020324707, -0.068725586, -0.06210327, -0.0015192032, -0.00058698654, -0.048339844, -0.0017108917, 0.017181396, 0.015052795, -0.018707275, 0.01826477, -0.0076026917, -0.02003479, -0.029022217, 0.019348145, -0.0053215027, ...
http://arxiv.org/abs/2310.00274v1
AfriSpeech-200: Pan-African Accented Speech Dataset for Clinical and General Domain ASR
Tobi Olatunji, Tejumade Afonja, Aditya Yadavalli, Chris Chinenye Emezue, Sahib Singh, Bonaventure F. P. Dossou, Joanne Osuchukwu, Salomey Osei, Atnafu Lambebo Tonja, Naome Etori, Clinton Mbataku
cs.CL
Africa has a very low doctor-to-patient ratio. At very busy clinics, doctors could see 30+ patients per day -- a heavy patient burden compared with developed countries -- but productivity tools such as clinical automatic speech recognition (ASR) are lacking for these overworked clinicians. However, clinical ASR is matu...
http://arxiv.org/pdf/2310.00274v1
2023-09-30T06:38:43
2023-09-30T06:38:43
{ "title": [ -0.016189575, -0.013130188, -0.03451538, -0.0715332, -0.009124756, 0.0017652512, -0.061828613, 0.021453857, -0.025817871, 0.042907715, 0.015342712, 0.0010967255, 0.040039062, -0.052642822, -0.014694214, 0.015235901, 0.00995636, 0.022...
http://arxiv.org/abs/2310.00272v1
Investigating the Efficacy of Large Language Models in Reflective Assessment Methods through Chain of Thoughts Prompting
Baphumelele Masikisiki, Vukosi Marivate, Yvette Hlope
cs.CL, cs.AI
Large Language Models, such as Generative Pre-trained Transformer 3 (aka. GPT-3), have been developed to understand language through the analysis of extensive text data, allowing them to identify patterns and connections between words. While LLMs have demonstrated impressive performance across various text-related task...
http://arxiv.org/pdf/2310.00272v1
2023-09-30T06:25:27
2023-09-30T06:25:27
{ "title": [ -0.017929077, -0.034179688, -0.006526947, -0.039916992, -0.041168213, 0.007396698, -0.028198242, 0.0020885468, 0.023376465, 0.010971069, -0.003862381, -0.03265381, 0.0077400208, -0.04840088, -0.046051025, -0.012870789, -0.014572144, ...
http://arxiv.org/abs/2310.00259v1
AutoHall: Automated Hallucination Dataset Generation for Large Language Models
Zouying Cao, Yifei Yang, Hai Zhao
cs.CL
While Large language models (LLMs) have garnered widespread applications across various domains due to their powerful language understanding and generation capabilities, the detection of non-factual or hallucinatory content generated by LLMs remains scarce. Currently, one significant challenge in hallucination detectio...
http://arxiv.org/pdf/2310.00259v1
2023-09-30T05:20:02
2023-09-30T05:20:02
{ "title": [ -0.017654419, -0.011741638, -0.03314209, -0.05999756, -0.021438599, 0.0044059753, -0.05645752, 0.0048942566, 0.012710571, 0.041870117, -0.019348145, 0.015853882, 0.024398804, -0.0022010803, -0.05230713, 0.02532959, -0.015472412, 0.05...
http://arxiv.org/abs/2310.00230v1
SLM: Bridge the thin gap between speech and text foundation models
Mingqiu Wang, Wei Han, Izhak Shafran, Zelin Wu, Chung-Cheng Chiu, Yuan Cao, Yongqiang Wang, Nanxin Chen, Yu Zhang, Hagen Soltau, Paul Rubenstein, Lukas Zilka, Dian Yu, Zhong Meng, Golan Pundak, Nikhil Siddhartha, Johan Schalkwyk, Yonghui Wu
cs.CL, cs.SD, eess.AS
We present a joint Speech and Language Model (SLM), a multitask, multilingual, and dual-modal model that takes advantage of pretrained foundational speech and language models. SLM freezes the pretrained foundation models to maximally preserves their capabilities, and only trains a simple adapter with just 1\% (156M) of...
http://arxiv.org/pdf/2310.00230v1
2023-09-30T02:27:45
2023-09-30T02:27:45
{ "title": [ -0.030471802, -0.019104004, -0.060791016, -0.048706055, -0.04473877, -0.0030899048, -0.08441162, 0.010902405, -0.020767212, 0.009315491, -0.01348114, 0.012542725, 0.03289795, -0.059570312, 0.04147339, -0.022949219, -0.033416748, 0.04...
http://arxiv.org/abs/2310.00212v3
Pairwise Proximal Policy Optimization: Harnessing Relative Feedback for LLM Alignment
Tianhao Wu, Banghua Zhu, Ruoyu Zhang, Zhaojin Wen, Kannan Ramchandran, Jiantao Jiao
cs.LG, cs.AI, cs.CL
Large Language Models (LLMs) can acquire extensive world knowledge through pre-training on large corpora. However, due to exposure to low-quality data, LLMs may exhibit harmful behavior without aligning with human values. The dominant approach for steering LLMs towards beneficial behavior involves Reinforcement Learnin...
http://arxiv.org/pdf/2310.00212v3
2023-10-10T02:32:08
2023-09-30T01:23:22
{ "title": [ -0.051574707, 0.009506226, -0.03274536, -0.024917603, -0.0413208, 0.0019626617, -0.07086182, -0.027755737, -0.036132812, 0.012451172, -0.00944519, 0.019683838, -0.026428223, -0.03805542, 0.015777588, -0.02053833, -0.0058555603, -0.02...
http://arxiv.org/abs/2310.00207v1
Detecting Unseen Multiword Expressions in American Sign Language
Lee Kezar, Aryan Shukla
cs.CL
Multiword expressions present unique challenges in many translation tasks. In an attempt to ultimately apply a multiword expression detection system to the translation of American Sign Language, we built and tested two systems that apply word embeddings from GloVe to determine whether or not the word embeddings of lexe...
http://arxiv.org/pdf/2310.00207v1
2023-09-30T00:54:59
2023-09-30T00:54:59
{ "title": [ -0.024093628, -0.0017442703, -0.0074653625, -0.00680542, -0.026367188, 0.0024871826, -0.03451538, -0.019241333, 0.0129470825, 0.012283325, -0.06149292, 0.0026359558, -0.010131836, -0.0126571655, -0.054016113, -0.024490356, -0.028793335, ...
http://arxiv.org/abs/2310.00204v1
Finding Pragmatic Differences Between Disciplines
Lee Kezar, Jay Pujara
cs.CL
Scholarly documents have a great degree of variation, both in terms of content (semantics) and structure (pragmatics). Prior work in scholarly document understanding emphasizes semantics through document summarization and corpus topic modeling but tends to omit pragmatics such as document organization and flow. Using a...
http://arxiv.org/pdf/2310.00204v1
2023-09-30T00:46:14
2023-09-30T00:46:14
{ "title": [ -0.017181396, 0.021743774, -0.05053711, -0.0025558472, -0.01966858, -0.038085938, -0.0061302185, -0.011383057, -0.023422241, 0.0082092285, -0.018234253, -0.013130188, -0.050994873, -0.046722412, 0.013961792, -0.034088135, 0.0016288757, ...
http://arxiv.org/abs/2310.00196v1
The Sem-Lex Benchmark: Modeling ASL Signs and Their Phonemes
Lee Kezar, Elana Pontecorvo, Adele Daniels, Connor Baer, Ruth Ferster, Lauren Berger, Jesse Thomason, Zed Sevcikova Sehyr, Naomi Caselli
cs.CL, cs.CV
Sign language recognition and translation technologies have the potential to increase access and inclusion of deaf signing communities, but research progress is bottlenecked by a lack of representative data. We introduce a new resource for American Sign Language (ASL) modeling, the Sem-Lex Benchmark. The Benchmark is t...
http://arxiv.org/pdf/2310.00196v1
2023-09-30T00:25:43
2023-09-30T00:25:43
{ "title": [ -0.061065674, -0.012321472, -0.022949219, -0.019607544, -0.048034668, 0.016586304, -0.07342529, -0.00032234192, -0.0099105835, 0.028915405, -0.04083252, 0.018112183, -0.0003633499, -0.022781372, -0.018737793, -0.0068473816, -0.0016040802...
http://arxiv.org/abs/2310.00195v1
Exploring Strategies for Modeling Sign Language Phonology
Lee Kezar, Riley Carlin, Tejas Srinivasan, Zed Sehyr, Naomi Caselli, Jesse Thomason
cs.CL, cs.CV
Like speech, signs are composed of discrete, recombinable features called phonemes. Prior work shows that models which can recognize phonemes are better at sign recognition, motivating deeper exploration into strategies for modeling sign language phonemes. In this work, we learn graph convolution networks to recognize ...
http://arxiv.org/pdf/2310.00195v1
2023-09-30T00:19:10
2023-09-30T00:19:10
{ "title": [ -0.043029785, -0.008453369, -0.015914917, -0.024841309, -0.036224365, 0.012641907, -0.04522705, 0.0040779114, -0.010848999, 0.031188965, -0.028152466, 0.0012235641, -0.00045466423, -0.028213501, -0.01713562, -0.009506226, -0.01890564, ...
http://arxiv.org/abs/2310.00178v1
Contextual Biasing with the Knuth-Morris-Pratt Matching Algorithm
Weiran Wang, Zelin Wu, Diamantino Caseiro, Tsendsuren Munkhdalai, Khe Chai Sim, Pat Rondon, Golan Pundak, Gan Song, Rohit Prabhavalkar, Zhong Meng, Ding Zhao, Tara Sainath, Pedro Moreno Mengibar
cs.CL, eess.AS
Contextual biasing refers to the problem of biasing the automatic speech recognition (ASR) systems towards rare entities that are relevant to the specific user or application scenarios. We propose algorithms for contextual biasing based on the Knuth-Morris-Pratt algorithm for pattern matching. During beam search, we bo...
http://arxiv.org/pdf/2310.00178v1
2023-09-29T22:50:10
2023-09-29T22:50:10
{ "title": [ -0.023376465, 0.020263672, -0.052368164, -0.021743774, 0.014221191, -0.006313324, -0.017410278, -0.05041504, 0.009094238, 0.04727173, -0.03564453, 0.007835388, 0.02293396, -0.025466919, -0.037109375, 0.0034694672, -0.031677246, 0.043...
http://arxiv.org/abs/2310.00160v1
Self-Specialization: Uncovering Latent Expertise within Large Language Models
Junmo Kang, Hongyin Luo, Yada Zhu, James Glass, David Cox, Alan Ritter, Rogerio Feris, Leonid Karlinsky
cs.CL, cs.AI
Recent works have demonstrated the effectiveness of self-alignment in which a large language model is, by itself, aligned to follow general instructions through the automatic generation of instructional data using a handful of human-written seeds. Instead of general alignment, in this work, we focus on self-alignment f...
http://arxiv.org/pdf/2310.00160v1
2023-09-29T21:53:46
2023-09-29T21:53:46
{ "title": [ -0.015403748, 0.0049438477, -0.051116943, -0.06854248, 0.0051116943, -0.022399902, -0.04034424, 0.01612854, -0.01890564, 0.045166016, -0.0004708767, 0.025299072, -0.00056552887, -0.006816864, -0.0473938, 0.011566162, -0.0051612854, 0...
http://arxiv.org/abs/2310.00152v1
Automatic Prompt Rewriting for Personalized Text Generation
Cheng Li, Mingyang Zhang, Qiaozhu Mei, Weize Kong, Michael Bendersky
cs.CL
Facilitated by large language models (LLMs), personalized text generation has become a rapidly growing research direction. Most existing studies focus on designing specialized models for a particular domain, or they require fine-tuning the LLMs to generate personalized text. We consider a typical scenario in which the ...
http://arxiv.org/pdf/2310.00152v1
2023-09-29T21:15:49
2023-09-29T21:15:49
{ "title": [ -0.03353882, -0.01914978, -0.027053833, -0.06878662, -0.018310547, 0.036102295, -0.0657959, -0.0013923645, -0.0055999756, 0.041656494, -0.014785767, 0.022445679, 0.005264282, -0.009094238, -0.04046631, 0.007320404, -0.015220642, 0.01...
http://arxiv.org/abs/2310.00141v1
The Gift of Feedback: Improving ASR Model Quality by Learning from User Corrections through Federated Learning
Lillian Zhou, Yuxin Ding, Mingqing Chen, Harry Zhang, Rohit Prabhavalkar, Dhruv Guliani, Giovanni Motta, Rajiv Mathews
cs.CL, eess.AS
Automatic speech recognition (ASR) models are typically trained on large datasets of transcribed speech. As language evolves and new terms come into use, these models can become outdated and stale. In the context of models trained on the server but deployed on edge devices, errors may result from the mismatch between s...
http://arxiv.org/pdf/2310.00141v1
2023-09-29T21:04:10
2023-09-29T21:04:10
{ "title": [ -0.00031638145, 0.007083893, -0.052368164, -0.08416748, 0.000018894672, -0.0063552856, -0.05444336, 0.0047683716, -0.027862549, 0.047027588, -0.02230835, 0.006416321, -0.0007214546, -0.012130737, -0.044708252, -0.017486572, -0.0053100586...
http://arxiv.org/abs/2310.00100v3
Multilingual Natural Language Processing Model for Radiology Reports -- The Summary is all you need!
Mariana Lindo, Ana Sofia Santos, André Ferreira, Jianning Li, Gijs Luijten, Gustavo Correia, Moon Kim, Jens Kleesiek, Jan Egger, Victor Alves
cs.CL, cs.AI
The impression section of a radiology report summarizes important radiology findings and plays a critical role in communicating these findings to physicians. However, the preparation of these summaries is time-consuming and error-prone for radiologists. Recently, numerous models for radiology report summarization have ...
http://arxiv.org/pdf/2310.00100v3
2023-10-26T15:18:01
2023-09-29T19:20:27
{ "title": [ -0.013916016, -0.020324707, -0.032104492, -0.07897949, -0.029891968, 0.023925781, -0.10211182, 0.031402588, -0.012741089, 0.041900635, -0.015556335, -0.0040397644, 0.016723633, -0.016021729, -0.009002686, -0.0067977905, 0.010314941, ...
http://arxiv.org/abs/2310.00092v1
Voice2Action: Language Models as Agent for Efficient Real-Time Interaction in Virtual Reality
Yang Su
cs.CL, cs.AI, cs.HC
Large Language Models (LLMs) are trained and aligned to follow natural language instructions with only a handful of examples, and they are prompted as task-driven autonomous agents to adapt to various sources of execution environments. However, deploying agent LLMs in virtual reality (VR) has been challenging due to th...
http://arxiv.org/pdf/2310.00092v1
2023-09-29T19:06:52
2023-09-29T19:06:52
{ "title": [ -0.031311035, -0.012809753, -0.04916382, -0.06109619, 0.012191772, -0.0070228577, -0.034118652, 0.00063180923, -0.010444641, 0.053253174, 0.021942139, -0.016571045, 0.020263672, -0.03656006, -0.06542969, 0.011566162, -0.014022827, 0....
http://arxiv.org/abs/2310.00074v1
SocREval: Large Language Models with the Socratic Method for Reference-Free Reasoning Evaluation
Hangfeng He, Hongming Zhang, Dan Roth
cs.CL, cs.AI
To comprehensively assess the capacity of current models for complex reasoning, it is crucial to assess their step-by-step reasoning in a scalable manner. Established reference-based evaluation metrics rely on human-annotated reasoning chains to assess the model-derived chains. However, such ``gold-standard'' human-wri...
http://arxiv.org/pdf/2310.00074v1
2023-09-29T18:25:46
2023-09-29T18:25:46
{ "title": [ -0.032440186, -0.026473999, -0.04824829, -0.07269287, 0.00333786, -0.0028629303, -0.041168213, 0.018859863, -0.0033302307, 0.013328552, -0.004928589, -0.012138367, -0.022094727, -0.030593872, -0.04373169, -0.008560181, -0.042999268, ...
http://arxiv.org/abs/2309.17453v1
Efficient Streaming Language Models with Attention Sinks
Guangxuan Xiao, Yuandong Tian, Beidi Chen, Song Han, Mike Lewis
cs.CL, cs.AI
Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challenges. Firstly, during the decoding stage, caching previous tokens' Key and Value states (KV) consumes extensive memory. Secondly, popular LLMs...
http://arxiv.org/pdf/2309.17453v1
2023-09-29T17:59:56
2023-09-29T17:59:56
{ "title": [ -0.0065231323, 0.00056409836, -0.021148682, -0.08905029, 0.007785797, 0.01499176, -0.058654785, 0.021728516, 0.023071289, 0.036102295, -0.038116455, -0.0028858185, 0.017501831, 0.013092041, -0.053985596, 0.011474609, -0.0019292831, 0...
http://arxiv.org/abs/2309.17452v2
ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving
Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Minlie Huang, Nan Duan, Weizhu Chen
cs.CL, cs.AI
Large language models have made significant progress in various language tasks, yet they still struggle with complex mathematics. In this paper, we propose ToRA a series of Tool-integrated Reasoning Agents designed to solve challenging mathematical problems by seamlessly integrating natural language reasoning with the ...
http://arxiv.org/pdf/2309.17452v2
2023-10-04T08:13:03
2023-09-29T17:59:38
{ "title": [ -0.018157959, -0.013725281, -0.02658081, -0.045562744, 0.0003964901, 0.0025978088, -0.007820129, -0.0022220612, 0.029937744, 0.022476196, -0.031921387, -0.004005432, 0.01007843, -0.011375427, -0.04345703, -0.00248909, -0.010818481, 0...
http://arxiv.org/abs/2309.17447v1
A Large Language Model Approach to Educational Survey Feedback Analysis
Michael J. Parker, Caitlin Anderson, Claire Stone, YeaRim Oh
cs.CL
This paper assesses the potential for the large language models (LLMs) GPT-4 and GPT-3.5 to aid in deriving insight from education feedback surveys. Exploration of LLM use cases in education has focused on teaching and learning, with less exploration of capabilities in education feedback analysis. Survey analysis in ed...
http://arxiv.org/pdf/2309.17447v1
2023-09-29T17:57:23
2023-09-29T17:57:23
{ "title": [ -0.031585693, 0.002204895, -0.011558533, -0.005748749, -0.0056037903, -0.002040863, -0.017089844, -0.0309906, 0.016586304, 0.0262146, -0.028213501, -0.012794495, -0.03729248, -0.015655518, -0.06793213, -0.023284912, -0.016693115, 0.0...
http://arxiv.org/abs/2309.17446v2
L2CEval: Evaluating Language-to-Code Generation Capabilities of Large Language Models
Ansong Ni, Pengcheng Yin, Yilun Zhao, Martin Riddell, Troy Feng, Rui Shen, Stephen Yin, Ye Liu, Semih Yavuz, Caiming Xiong, Shafiq Joty, Yingbo Zhou, Dragomir Radev, Arman Cohan
cs.CL, cs.LG, cs.PL, cs.SE
Recently, large language models (LLMs), especially those that are pretrained on code, have demonstrated strong capabilities in generating programs from natural language inputs in a few-shot or even zero-shot manner. Despite promising results, there is a notable lack of a comprehensive evaluation of these models languag...
http://arxiv.org/pdf/2309.17446v2
2023-10-02T09:54:50
2023-09-29T17:57:00
{ "title": [ -0.046905518, -0.03338623, -0.03741455, -0.06500244, -0.008560181, 0.018798828, -0.057556152, 0.028076172, -0.003189087, 0.029327393, -0.01071167, 0.014266968, 0.009819031, -0.02178955, -0.024490356, 0.015609741, -0.01828003, 0.03004...
http://arxiv.org/abs/2309.17444v2
LLM-grounded Video Diffusion Models
Long Lian, Baifeng Shi, Adam Yala, Trevor Darrell, Boyi Li
cs.CV, cs.AI, cs.CL
Text-conditioned diffusion models have emerged as a promising tool for neural video generation. However, current models still struggle with intricate spatiotemporal prompts and often generate restricted or incorrect motion (e.g., even lacking the ability to be prompted for objects moving from left to right). To address...
http://arxiv.org/pdf/2309.17444v2
2023-10-02T01:46:44
2023-09-29T17:54:46
{ "title": [ 0.0027389526, 0.034179688, -0.052490234, -0.017944336, -0.0014600754, -0.03515625, -0.01260376, -0.016921997, -0.015792847, 0.030792236, -0.037719727, 0.015235901, 0.01802063, -0.046447754, 0.007156372, -0.011016846, -0.011024475, 0....
http://arxiv.org/abs/2309.17428v1
CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets
Lifan Yuan, Yangyi Chen, Xingyao Wang, Yi R. Fung, Hao Peng, Heng Ji
cs.CL, cs.AI, cs.LG
Large language models (LLMs) are often augmented with tools to solve complex tasks. By generating code snippets and executing them through task-specific Application Programming Interfaces (APIs), they can offload certain functions to dedicated external modules, such as image encoding and performing calculations. Howeve...
http://arxiv.org/pdf/2309.17428v1
2023-09-29T17:40:26
2023-09-29T17:40:26
{ "title": [ -0.063964844, -0.00982666, -0.050598145, 0.005870819, -0.044799805, 0.006072998, -0.05606079, -0.010978699, -0.06939697, 0.04324341, 0.0023918152, 0.029556274, 0.009735107, -0.043182373, 0.02394104, -0.022750854, 0.0014209747, 0.0187...
http://arxiv.org/abs/2309.17421v2
The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)
Zhengyuan Yang, Linjie Li, Kevin Lin, Jianfeng Wang, Chung-Ching Lin, Zicheng Liu, Lijuan Wang
cs.CV, cs.CL
Large multimodal models (LMMs) extend large language models (LLMs) with multi-sensory skills, such as visual understanding, to achieve stronger generic intelligence. In this paper, we analyze the latest model, GPT-4V(ision), to deepen the understanding of LMMs. The analysis focuses on the intriguing tasks that GPT-4V c...
http://arxiv.org/pdf/2309.17421v2
2023-10-11T05:07:37
2023-09-29T17:34:51
{ "title": [ -0.0184021, -0.0058937073, -0.01448822, -0.030319214, -0.014625549, 0.0013780594, -0.050628662, -0.037139893, -0.029708862, 0.030731201, -0.0026226044, -0.0004851818, 0.029769897, -0.04510498, 0.008277893, 0.0077705383, -0.005027771, ...
http://arxiv.org/abs/2309.17415v2
Intuitive or Dependent? Investigating LLMs' Robustness to Conflicting Prompts
Jiahao Ying, Yixin Cao, Kai Xiong, Yidong He, Long Cui, Yongbin Liu
cs.CL
This paper explores the robustness of LLMs' preference to their internal memory or the given prompt, which may contain contrasting information in real-world applications due to noise or task settings. To this end, we establish a quantitative benchmarking framework and conduct the role playing intervention to control LL...
http://arxiv.org/pdf/2309.17415v2
2023-10-03T05:16:36
2023-09-29T17:26:03
{ "title": [ -0.032409668, 0.015159607, -0.03945923, -0.007472992, -0.014007568, -0.020477295, -0.045684814, -0.045166016, -0.007411957, 0.009666443, -0.055114746, 0.017547607, -0.02909851, -0.014770508, -0.03363037, -0.03326416, -0.014694214, 0....
http://arxiv.org/abs/2309.17410v1
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
Vaidehi Patil, Peter Hase, Mohit Bansal
cs.CL, cs.AI, cs.LG
Pretrained language models sometimes possess knowledge that we do not wish them to, including memorized personal information and knowledge that could be used to harm people. They can also output toxic or harmful text. To mitigate these safety and informational issues, we propose an attack-and-defense framework for stud...
http://arxiv.org/pdf/2309.17410v1
2023-09-29T17:12:43
2023-09-29T17:12:43
{ "title": [ -0.038879395, 0.03012085, -0.08709717, 0.024719238, 0.0016040802, -0.04977417, -0.0028629303, -0.011512756, -0.05114746, 0.015609741, 0.009506226, 0.0107421875, -0.014625549, -0.056518555, 0.016067505, -0.035064697, 0.0025310516, 0.0...
http://arxiv.org/abs/2310.01434v1
Revolutionizing Mobile Interaction: Enabling a 3 Billion Parameter GPT LLM on Mobile
Samuel Carreira, Tomás Marques, José Ribeiro, Carlos Grilo
cs.CL, cs.AI, cs.LG
The field of Artificial Intelligence has witnessed remarkable progress in recent years, especially with the emergence of powerful large language models (LLMs) based on the transformer architecture. Cloud-based LLMs, such as OpenAI's ChatGPT, offer impressive capabilities but come with concerns regarding latency and pri...
http://arxiv.org/pdf/2310.01434v1
2023-09-29T16:30:49
2023-09-29T16:30:49
{ "title": [ -0.03036499, -0.0064582825, -0.03665161, -0.05001831, -0.027862549, 0.0045433044, -0.074645996, -0.008644104, -0.021697998, 0.0064697266, -0.03479004, 0.02305603, -0.017196655, -0.033935547, 0.016052246, -0.0105896, -0.015563965, 0.0...
http://arxiv.org/abs/2309.17332v2
Overview of the BioLaySumm 2023 Shared Task on Lay Summarization of Biomedical Research Articles
Tomas Goldsack, Zheheng Luo, Qianqian Xie, Carolina Scarton, Matthew Shardlow, Sophia Ananiadou, Chenghua Lin
cs.CL
This paper presents the results of the shared task on Lay Summarisation of Biomedical Research Articles (BioLaySumm), hosted at the BioNLP Workshop at ACL 2023. The goal of this shared task is to develop abstractive summarisation models capable of generating "lay summaries" (i.e., summaries that are comprehensible to n...
http://arxiv.org/pdf/2309.17332v2
2023-10-25T08:16:43
2023-09-29T15:43:42
{ "title": [ -0.022888184, -0.0062217712, -0.022216797, -0.06738281, -0.027694702, 0.0121536255, -0.067993164, -0.01083374, -0.012931824, 0.04083252, 0.008056641, 0.023330688, 0.005279541, -0.016906738, -0.010551453, 0.023910522, 0.011505127, 0.0...
http://arxiv.org/abs/2309.17313v1
Few-Shot Domain Adaptation for Charge Prediction on Unprofessional Descriptions
Jie Zhao, Ziyu Guan, Wei Zhao, Yue Jiang, Xiaofei He
cs.CL
Recent works considering professional legal-linguistic style (PLLS) texts have shown promising results on the charge prediction task. However, unprofessional users also show an increasing demand on such a prediction service. There is a clear domain discrepancy between PLLS texts and non-PLLS texts expressed by those la...
http://arxiv.org/pdf/2309.17313v1
2023-09-29T15:14:08
2023-09-29T15:14:08
{ "title": [ 0.03086853, -0.006198883, -0.012870789, -0.06402588, -0.024307251, -0.009963989, -0.03387451, -0.014572144, 0.028015137, 0.02468872, 0.010375977, -0.02645874, 0.007835388, -0.04815674, -0.016220093, -0.0037174225, -0.0132751465, 0.02...
http://arxiv.org/abs/2310.01432v2
Split and Merge: Aligning Position Biases in Large Language Model based Evaluators
Zongjie Li, Chaozheng Wang, Pingchuan Ma, Daoyuan Wu, Shuai Wang, Cuiyun Gao, Yang Liu
cs.CL, cs.AI
Large language models (LLMs) have shown promise as automated evaluators for assessing the quality of answers generated by AI systems. However, these LLM-based evaluators exhibit position bias, or inconsistency, when used to evaluate candidate answers in pairwise comparisons, favoring either the first or second answer r...
http://arxiv.org/pdf/2310.01432v2
2023-10-09T07:19:10
2023-09-29T14:38:58
{ "title": [ -0.019973755, 0.0014152527, -0.027633667, -0.039642334, -0.005126953, -0.00093889236, -0.027709961, -0.021240234, 0.023757935, 0.017715454, -0.022018433, -0.014961243, 0.007896423, -0.02104187, -0.055755615, -0.010887146, -0.036346436, ...
http://arxiv.org/abs/2310.00034v2
PB-LLM: Partially Binarized Large Language Models
Yuzhang Shang, Zhihang Yuan, Qiang Wu, Zhen Dong
cs.LG, cs.AI, cs.CL
This paper explores network binarization, a radical form of quantization, compressing model weights to a single bit, specifically for Large Language Models (LLMs) compression. Due to previous binarization methods collapsing LLMs, we propose a novel approach, Partially-Binarized LLM (PB-LLM), which can achieve extreme l...
http://arxiv.org/pdf/2310.00034v2
2023-11-07T20:41:22
2023-09-29T14:35:27
{ "title": [ -0.04232788, -0.01991272, -0.045562744, -0.08660889, -0.0010519028, 0.017608643, -0.08514404, 0.025390625, -0.00089740753, 0.031219482, -0.0062561035, -0.0038909912, 0.014205933, -0.028121948, -0.03253174, 0.020843506, -0.015235901, ...
http://arxiv.org/abs/2309.17280v1
STRONG -- Structure Controllable Legal Opinion Summary Generation
Yang Zhong, Diane Litman
cs.CL, cs.AI
We propose an approach for the structure controllable summarization of long legal opinions that considers the argument structure of the document. Our approach involves using predicted argument role information to guide the model in generating coherent summaries that follow a provided structure pattern. We demonstrate t...
http://arxiv.org/pdf/2309.17280v1
2023-09-29T14:31:41
2023-09-29T14:31:41
{ "title": [ 0.002380371, -0.012123108, -0.038879395, -0.07574463, -0.02772522, 0.0031204224, -0.07055664, 0.01084137, -0.024414062, 0.05142212, -0.025421143, 0.004245758, 0.025466919, -0.027832031, -0.0007696152, -0.002368927, 0.00084877014, -0....
http://arxiv.org/abs/2309.17272v1
Enhancing Large Language Models in Coding Through Multi-Perspective Self-Consistency
Baizhou Huang, Shuai Lu, Weizhu Chen, Xiaojun Wan, Nan Duan
cs.CL, cs.AI, cs.SE
Large language models (LLMs) have exhibited remarkable ability in textual generation. However, in complex reasoning tasks such as code generation, generating the correct answer in a single attempt remains a formidable challenge for LLMs. Previous research has explored solutions by aggregating multiple outputs, leveragi...
http://arxiv.org/pdf/2309.17272v1
2023-09-29T14:23:26
2023-09-29T14:23:26
{ "title": [ -0.0446167, -0.012710571, -0.070495605, -0.078125, 0.0009860992, -0.0042648315, -0.072265625, 0.048065186, 0.005695343, 0.06085205, 0.022613525, -0.030151367, 0.032562256, -0.030044556, -0.059570312, 0.01335907, -0.0058288574, 0.0189...
http://arxiv.org/abs/2309.17267v1
Wiki-En-ASR-Adapt: Large-scale synthetic dataset for English ASR Customization
Alexandra Antonova
eess.AS, cs.CL, cs.SD
We present a first large-scale public synthetic dataset for contextual spellchecking customization of automatic speech recognition (ASR) with focus on diverse rare and out-of-vocabulary (OOV) phrases, such as proper names or terms. The proposed approach allows creating millions of realistic examples of corrupted ASR hy...
http://arxiv.org/pdf/2309.17267v1
2023-09-29T14:18:59
2023-09-29T14:18:59
{ "title": [ -0.0289917, -0.015670776, -0.050933838, -0.07611084, -0.012687683, 0.00042653084, -0.06866455, 0.030838013, -0.039611816, 0.03768921, -0.008331299, -0.018554688, 0.01876831, -0.037963867, -0.011146545, -0.013145447, 0.004634857, 0.05...
http://arxiv.org/abs/2309.17255v2
Knowledge Graphs for the Life Sciences: Recent Developments, Challenges and Opportunities
Jiaoyan Chen, Hang Dong, Janna Hastings, Ernesto Jiménez-Ruiz, Vanessa López, Pierre Monnin, Catia Pesquita, Petr Škoda, Valentina Tamma
cs.AI, cs.CL, I.2.4; J.3
The term life sciences refers to the disciplines that study living organisms and life processes, and include chemistry, biology, medicine, and a range of other related disciplines. Research efforts in life sciences are heavily data-driven, as they produce and consume vast amounts of scientific data, much of which is in...
http://arxiv.org/pdf/2309.17255v2
2023-10-06T15:02:55
2023-09-29T14:03:34
{ "title": [ -0.04876709, -0.027450562, -0.068115234, -0.07940674, -0.0032444, 0.014434814, -0.07128906, 0.02720642, 0.0014047623, 0.023422241, -0.0104599, -0.0036182404, -0.008377075, -0.02935791, -0.04449463, 0.0032100677, 0.022460938, 0.028091...
http://arxiv.org/abs/2309.17249v1
Batch Calibration: Rethinking Calibration for In-Context Learning and Prompt Engineering
Han Zhou, Xingchen Wan, Lev Proleev, Diana Mincu, Jilin Chen, Katherine Heller, Subhrajit Roy
cs.CL, cs.AI, cs.LG
Prompting and in-context learning (ICL) have become efficient learning paradigms for large language models (LLMs). However, LLMs suffer from prompt brittleness and various bias factors in the prompt, including but not limited to the formatting, the choice verbalizers, and the ICL examples. To address this problem that ...
http://arxiv.org/pdf/2309.17249v1
2023-09-29T13:55:45
2023-09-29T13:55:45
{ "title": [ -0.039093018, 0.005924225, -0.064331055, -0.023712158, -0.0018119812, -0.016677856, -0.032409668, -0.021240234, -0.018325806, 0.0047302246, -0.04824829, 0.008728027, -0.03302002, -0.029830933, -0.015464783, -0.035369873, 0.0035762787, ...
http://arxiv.org/abs/2309.17234v1
LLM-Deliberation: Evaluating LLMs with Interactive Multi-Agent Negotiation Games
Sahar Abdelnabi, Amr Gomaa, Sarath Sivaprasad, Lea Schönherr, Mario Fritz
cs.CL, cs.CY, cs.LG
There is a growing interest in using Large Language Models (LLMs) as agents to tackle real-world tasks that may require assessing complex situations. Yet, we have a limited understanding of LLMs' reasoning and decision-making capabilities, partly stemming from a lack of dedicated evaluation benchmarks. As negotiating a...
http://arxiv.org/pdf/2309.17234v1
2023-09-29T13:33:06
2023-09-29T13:33:06
{ "title": [ -0.041412354, -0.0023994446, -0.018478394, 0.0062065125, -0.02859497, -0.005355835, -0.00831604, -0.019515991, -0.025878906, 0.008041382, -0.0362854, 0.027572632, -0.05911255, -0.018569946, -0.026779175, -0.03793335, -0.037353516, 0....
http://arxiv.org/abs/2309.17224v1
Training and inference of large language models using 8-bit floating point
Sergio P. Perez, Yan Zhang, James Briggs, Charlie Blake, Josh Levy-Kramer, Paul Balanca, Carlo Luschi, Stephen Barlow, Andrew William Fitzgibbon
cs.LG, cs.AR, cs.CL, cs.ET, cs.PF, I.2.7; B.2.4
FP8 formats are gaining popularity to boost the computational efficiency for training and inference of large deep learning models. Their main challenge is that a careful choice of scaling is needed to prevent degradation due to the reduced dynamic range compared to higher-precision formats. Although there exists ample ...
http://arxiv.org/pdf/2309.17224v1
2023-09-29T13:24:33
2023-09-29T13:24:33
{ "title": [ -0.03314209, -0.007972717, -0.049316406, -0.055908203, -0.001991272, -0.003452301, -0.052703857, 0.015930176, -0.0015792847, 0.03439331, -0.03048706, 0.0031433105, 0.016433716, -0.0082473755, -0.04055786, 0.0059318542, -0.012870789, ...
http://arxiv.org/abs/2309.17179v1
Alphazero-like Tree-Search can Guide Large Language Model Decoding and Training
Xidong Feng, Ziyu Wan, Muning Wen, Ying Wen, Weinan Zhang, Jun Wang
cs.LG, cs.AI, cs.CL
Large language models (LLMs) typically employ sampling or beam search, accompanied by prompts such as Chain-of-Thought (CoT), to boost reasoning and decoding ability. Recent work like Tree-of-Thought (ToT) and Reasoning via Planning (RAP) aim to augment the reasoning capabilities of LLMs by utilizing tree-search algori...
http://arxiv.org/pdf/2309.17179v1
2023-09-29T12:20:19
2023-09-29T12:20:19
{ "title": [ -0.031951904, -0.016616821, -0.024047852, -0.0385437, -0.06793213, -0.0009994507, -0.086242676, -0.0004234314, -0.006034851, 0.066467285, -0.012359619, 0.0028743744, 0.032196045, -0.018585205, -0.025939941, 0.00028395653, -0.0018119812, ...
http://arxiv.org/abs/2309.17176v1
RLAdapter: Bridging Large Language Models to Reinforcement Learning in Open Worlds
Wanpeng Zhang, Zongqing Lu
cs.AI, cs.CL
While reinforcement learning (RL) shows remarkable success in decision-making problems, it often requires a lot of interactions with the environment, and in sparse-reward environments, it is challenging to learn meaningful policies. Large Language Models (LLMs) can potentially provide valuable guidance to agents in lea...
http://arxiv.org/pdf/2309.17176v1
2023-09-29T12:16:19
2023-09-29T12:16:19
{ "title": [ -0.033355713, -0.01134491, -0.020355225, -0.038879395, 0.0015878677, 0.0013484955, -0.030288696, -0.019134521, -0.037231445, 0.010391235, -0.0075912476, 0.014701843, -0.03552246, -0.024291992, -0.003545761, -0.019699097, -0.0121536255, ...
http://arxiv.org/abs/2309.17171v1
Comparative Analysis of Named Entity Recognition in the Dungeons and Dragons Domain
Gayashan Weerasundara, Nisansa de Silva
cs.CL, cs.LG
Many NLP tasks, although well-resolved for general English, face challenges in specific domains like fantasy literature. This is evident in Named Entity Recognition (NER), which detects and categorizes entities in text. We analyzed 10 NER models on 7 Dungeons and Dragons (D&D) adventure books to assess domain-specific ...
http://arxiv.org/pdf/2309.17171v1
2023-09-29T12:09:36
2023-09-29T12:09:36
{ "title": [ -0.007194519, 0.008003235, -0.060150146, -0.033813477, 0.024017334, -0.032592773, -0.0005226135, -0.00025820732, -0.020248413, 0.033996582, -0.025894165, 0.015144348, 0.04168701, -0.022323608, 0.0002092123, 0.00831604, -0.0012626648, ...
http://arxiv.org/abs/2309.17169v1
An evaluation of GPT models for phenotype concept recognition
Tudor Groza, Harry Caufield, Dylan Gration, Gareth Baynam, Melissa A Haendel, Peter N Robinson, Chris J Mungall, Justin T Reese
cs.CL, cs.AI
Objective: Clinical deep phenotyping plays a critical role in both the diagnosis of patients with rare disorders as well as in building care coordination plans. The process relies on modelling and curating patient profiles using ontology concepts, usually from the Human Phenotype Ontology. Machine learning methods have...
http://arxiv.org/pdf/2309.17169v1
2023-09-29T12:06:55
2023-09-29T12:06:55
{ "title": [ 0.0066871643, 0.000562191, -0.020935059, -0.04135132, -0.0057678223, -0.011627197, -0.046783447, -0.02848816, 0.01689148, 0.017593384, -0.021026611, -0.0112838745, -0.017028809, -0.036193848, -0.046661377, 0.0014505386, -0.023925781, ...
http://arxiv.org/abs/2309.17167v2
DyVal: Graph-informed Dynamic Evaluation of Large Language Models
Kaijie Zhu, Jiaao Chen, Jindong Wang, Neil Zhenqiang Gong, Diyi Yang, Xing Xie
cs.AI, cs.CL, cs.LG
Large language models (LLMs) have achieved remarkable performance in various evaluation benchmarks. However, concerns about their performance are raised on potential data contamination in their considerable volume of training corpus. Moreover, the static nature and fixed complexity of current benchmarks may inadequatel...
http://arxiv.org/pdf/2309.17167v2
2023-10-05T07:27:12
2023-09-29T12:04:14
{ "title": [ -0.026473999, -0.008079529, -0.047790527, -0.08117676, 0.006843567, 0.0009536743, -0.06713867, 0.008865356, -0.0023937225, 0.021728516, -0.01574707, 0.013389587, -0.009117126, -0.023162842, -0.028518677, 0.008415222, -0.01763916, 0.0...
http://arxiv.org/abs/2309.17157v3
LatticeGen: A Cooperative Framework which Hides Generated Text in a Lattice for Privacy-Aware Generation on Cloud
Mengke Zhang, Tianxing He, Tianle Wang, Lu Mi, Fatemehsadat Mireshghallah, Binyi Chen, Hao Wang, Yulia Tsvetkov
cs.CL
In the current user-server interaction paradigm of prompted generation with large language models (LLM) on cloud, the server fully controls the generation process, which leaves zero options for users who want to keep the generated text to themselves. We propose LatticeGen, a cooperative framework in which the server st...
http://arxiv.org/pdf/2309.17157v3
2023-11-14T17:40:37
2023-09-29T11:46:07
{ "title": [ -0.030151367, -0.014022827, -0.07836914, -0.05682373, -0.027053833, -0.01889038, -0.049743652, 0.039276123, -0.039794922, 0.059631348, 0.007698059, 0.008033752, 0.061157227, -0.039123535, -0.009353638, 0.0064964294, -0.0138549805, 0....
http://arxiv.org/abs/2309.17147v2
Using Large Language Models for Qualitative Analysis can Introduce Serious Bias
Julian Ashwin, Aditya Chhabra, Vijayendra Rao
cs.CL, cs.AI, econ.GN, q-fin.EC
Large Language Models (LLMs) are quickly becoming ubiquitous, but the implications for social science research are not yet well understood. This paper asks whether LLMs can help us analyse large-N qualitative data from open-ended interviews, with an application to transcripts of interviews with Rohingya refugees in Cox...
http://arxiv.org/pdf/2309.17147v2
2023-10-05T12:25:18
2023-09-29T11:19:15
{ "title": [ 0.018371582, -0.0030784607, -0.012321472, -0.018035889, -0.029129028, -0.015411377, -0.01637268, -0.01574707, -0.042633057, 0.049560547, -0.047729492, -0.03656006, 0.008987427, -0.026931763, -0.0026168823, -0.044708252, -0.019515991, ...
http://arxiv.org/abs/2309.17134v1
Promoting Generalized Cross-lingual Question Answering in Few-resource Scenarios via Self-knowledge Distillation
Casimiro Pio Carrino, Carlos Escolano, José A. R. Fonollosa
cs.CL
Despite substantial progress in multilingual extractive Question Answering (QA), models with high and uniformly distributed performance across languages remain challenging, especially for languages with limited resources. We study cross-lingual transfer mainly focusing on the Generalized Cross-Lingual Transfer (G-XLT) ...
http://arxiv.org/pdf/2309.17134v1
2023-09-29T10:54:59
2023-09-29T10:54:59
{ "title": [ -0.0010976791, -0.006034851, -0.047424316, -0.07086182, -0.0007376671, -0.0009455681, -0.055419922, 0.012931824, 0.022827148, 0.03704834, -0.01423645, -0.003753662, 0.00028800964, -0.011779785, -0.060058594, 0.0012378693, 0.0075683594, ...
http://arxiv.org/abs/2309.17133v2
Fine-grained Late-interaction Multi-modal Retrieval for Retrieval Augmented Visual Question Answering
Weizhe Lin, Jinghong Chen, Jingbiao Mei, Alexandru Coca, Bill Byrne
cs.CL, cs.CV
Knowledge-based Visual Question Answering (KB-VQA) requires VQA systems to utilize knowledge from external knowledge bases to answer visually-grounded questions. Retrieval-Augmented Visual Question Answering (RA-VQA), a strong framework to tackle KB-VQA, first retrieves related documents with Dense Passage Retrieval (D...
http://arxiv.org/pdf/2309.17133v2
2023-10-28T16:03:35
2023-09-29T10:54:10
{ "title": [ 0.020645142, 0.010375977, -0.02885437, -0.048828125, 0.015617371, -0.009788513, -0.006843567, -0.008399963, 0.017471313, 0.048187256, -0.019012451, 0.0029621124, 0.00034928322, -0.010932922, -0.08258057, -0.013298035, 0.0137786865, -...
http://arxiv.org/abs/2309.17122v1
Benchmarking the Abilities of Large Language Models for RDF Knowledge Graph Creation and Comprehension: How Well Do LLMs Speak Turtle?
Johannes Frey, Lars-Peter Meyer, Natanael Arndt, Felix Brei, Kirill Bulert
cs.AI, cs.CL, cs.DB
Large Language Models (LLMs) are advancing at a rapid pace, with significant improvements at natural language processing and coding tasks. Yet, their ability to work with formal languages representing data, specifically within the realm of knowledge graph engineering, remains under-investigated. To evaluate the profici...
http://arxiv.org/pdf/2309.17122v1
2023-09-29T10:36:04
2023-09-29T10:36:04
{ "title": [ -0.04660034, -0.0051994324, -0.048797607, -0.05117798, -0.032958984, 0.0066184998, -0.10345459, -0.009094238, 0.024993896, 0.022399902, -0.062164307, 0.00459671, -0.0067977905, 0.0041542053, -0.049591064, -0.0262146, -0.0053710938, 0...
http://arxiv.org/abs/2309.17061v1
SCALE: Synergized Collaboration of Asymmetric Language Translation Engines
Xin Cheng, Xun Wang, Tao Ge, Si-Qing Chen, Furu Wei, Dongyan Zhao, Rui Yan
cs.CL, cs.AI
In this paper, we introduce SCALE, a collaborative framework that connects compact Specialized Translation Models (STMs) and general-purpose Large Language Models (LLMs) as one unified translation engine. By introducing translation from STM into the triplet in-context demonstrations, SCALE unlocks refinement and pivoti...
http://arxiv.org/pdf/2309.17061v1
2023-09-29T08:46:38
2023-09-29T08:46:38
{ "title": [ -0.05529785, -0.010444641, -0.04321289, -0.041412354, -0.023284912, 0.004421234, -0.0501709, 0.00749588, -0.017044067, 0.045196533, -0.011184692, 0.00089359283, 0.02420044, -0.040771484, -0.030044556, 0.0032863617, 0.0030879974, 0.01...
http://arxiv.org/abs/2309.17050v1
Interpretable Long-Form Legal Question Answering with Retrieval-Augmented Large Language Models
Antoine Louis, Gijs van Dijck, Gerasimos Spanakis
cs.CL
Many individuals are likely to face a legal dispute at some point in their lives, but their lack of understanding of how to navigate these complex issues often renders them vulnerable. The advancement of natural language processing opens new avenues for bridging this legal literacy gap through the development of automa...
http://arxiv.org/pdf/2309.17050v1
2023-09-29T08:23:19
2023-09-29T08:23:19
{ "title": [ -0.026657104, -0.00856781, -0.030410767, -0.062683105, -0.008216858, 0.015388489, -0.060516357, 0.007091522, 0.024917603, 0.02029419, -0.03692627, -0.00969696, 0.011940002, 0.002105713, -0.060180664, -0.010551453, -0.013214111, 0.032...
http://arxiv.org/abs/2309.17035v1
Contextualising Levels of Language Resourcedness affecting Digital Processing of Text
C. Maria Keet, Langa Khumalo
cs.CL, I.2.7
Application domains such as digital humanities and tool like chatbots involve some form of processing natural language, from digitising hardcopies to speech generation. The language of the content is typically characterised as either a low resource language (LRL) or high resource language (HRL), also known as resource-...
http://arxiv.org/pdf/2309.17035v1
2023-09-29T07:48:24
2023-09-29T07:48:24
{ "title": [ 0.0077323914, 0.0059394836, -0.022521973, -0.024780273, -0.029449463, -0.029388428, -0.009811401, -0.012062073, 0.002779007, 0.055145264, -0.023727417, -0.011108398, 0.026992798, -0.020690918, -0.05355835, -0.0019407272, -0.0065231323, ...
http://arxiv.org/abs/2310.01430v1
Sarcasm in Sight and Sound: Benchmarking and Expansion to Improve Multimodal Sarcasm Detection
Swapnil Bhosale, Abhra Chaudhuri, Alex Lee Robert Williams, Divyank Tiwari, Anjan Dutta, Xiatian Zhu, Pushpak Bhattacharyya, Diptesh Kanojia
cs.CL, cs.AI
The introduction of the MUStARD dataset, and its emotion recognition extension MUStARD++, have identified sarcasm to be a multi-modal phenomenon -- expressed not only in natural language text, but also through manners of speech (like tonality and intonation) and visual cues (facial expression). With this work, we aim t...
http://arxiv.org/pdf/2310.01430v1
2023-09-29T07:00:41
2023-09-29T07:00:41
{ "title": [ -0.015609741, -0.00029945374, -0.033477783, -0.030441284, 0.00058460236, -0.016448975, -0.031585693, -0.024383545, 0.0113220215, 0.023742676, -0.0335083, -0.026016235, 0.015365601, -0.018325806, -0.051849365, -0.015914917, -0.019317627, ...
http://arxiv.org/abs/2309.17012v1
Benchmarking Cognitive Biases in Large Language Models as Evaluators
Ryan Koo, Minhwa Lee, Vipul Raheja, Jong Inn Park, Zae Myung Kim, Dongyeop Kang
cs.CL, cs.AI, cs.LG
Large Language Models (LLMs) have recently been shown to be effective as automatic evaluators with simple prompting and in-context learning. In this work, we assemble 15 LLMs of four different size ranges and evaluate their output responses by preference ranking from the other LLMs as evaluators, such as System Star is...
http://arxiv.org/pdf/2309.17012v1
2023-09-29T06:53:10
2023-09-29T06:53:10
{ "title": [ -0.026535034, -0.0069503784, -0.03640747, -0.03878784, -0.0012464523, -0.011474609, -0.033721924, -0.030075073, 0.011550903, 0.021942139, -0.023544312, -0.008491516, 0.0035419464, -0.015853882, -0.052459717, -0.01184845, -0.028945923, ...
http://arxiv.org/abs/2309.16938v1
I Wish to Have an Argument: Argumentative Reasoning in Large Language Models
Adrian de Wynter, Tommy Yuan
cs.CL
We evaluate the ability of contemporary large language models (LLMs) to perform argumentative reasoning. We frame our experiments in terms of the argument mining (AM) and argument pair extraction (APE) tasks, and evaluate their ability to perform reasoning at increasing levels of abstraction in the input and output rep...
http://arxiv.org/pdf/2309.16938v1
2023-09-29T02:41:38
2023-09-29T02:41:38
{ "title": [ -0.013473511, -0.020721436, -0.04324341, -0.083740234, 0.00016462803, -0.0066947937, -0.054779053, 0.009094238, -0.0018758774, 0.007335663, -0.01776123, -0.010772705, 0.0034236908, -0.030212402, -0.02722168, -0.015174866, -0.026000977, ...
http://arxiv.org/abs/2309.16937v1
SSHR: Leveraging Self-supervised Hierarchical Representations for Multilingual Automatic Speech Recognition
Hongfei Xue, Qijie Shao, Kaixun Huang, Peikun Chen, Lei Xie, Jie Liu
cs.CL, cs.SD, eess.AS
Multilingual automatic speech recognition (ASR) systems have garnered attention for their potential to extend language coverage globally. While self-supervised learning (SSL) has demonstrated its effectiveness in multilingual ASR, it is worth noting that the various layers' representations of SSL potentially contain di...
http://arxiv.org/pdf/2309.16937v1
2023-09-29T02:35:36
2023-09-29T02:35:36
{ "title": [ 0.0008735657, 0.0079956055, -0.047546387, -0.09814453, 0.01701355, 0.015266418, -0.07232666, 0.025314331, -0.028182983, 0.054473877, -0.010986328, 0.011497498, -0.003774643, -0.013168335, -0.04055786, -0.0014209747, -0.0069770813, 0....
http://arxiv.org/abs/2309.16905v1
Towards a Unified Framework for Adaptable Problematic Content Detection via Continual Learning
Ali Omrani, Alireza S. Ziabari, Preni Golazizian, Jeffrey Sorensen, Morteza Dehghani
cs.CL
Detecting problematic content, such as hate speech, is a multifaceted and ever-changing task, influenced by social dynamics, user populations, diversity of sources, and evolving language. There has been significant efforts, both in academia and in industry, to develop annotated resources that capture various aspects of...
http://arxiv.org/pdf/2309.16905v1
2023-09-29T00:14:38
2023-09-29T00:14:38
{ "title": [ 0.040008545, 0.023239136, -0.019119263, -0.04660034, 0.0385437, -0.030197144, -0.0042266846, -0.029205322, 0.027267456, 0.01727295, -0.029037476, 0.0055236816, -0.029724121, -0.010131836, -0.05718994, -0.009307861, -0.018692017, 0.04...
http://arxiv.org/abs/2309.16898v1
A Sign Language Recognition System with Pepper, Lightweight-Transformer, and LLM
JongYoon Lim, Inkyu Sa, Bruce MacDonald, Ho Seok Ahn
cs.RO, cs.CL, cs.CV, cs.HC
This research explores using lightweight deep neural network architectures to enable the humanoid robot Pepper to understand American Sign Language (ASL) and facilitate non-verbal human-robot interaction. First, we introduce a lightweight and efficient model for ASL understanding optimized for embedded systems, ensurin...
http://arxiv.org/pdf/2309.16898v1
2023-09-28T23:54:41
2023-09-28T23:54:41
{ "title": [ -0.034973145, -0.0025539398, -0.064697266, -0.031921387, -0.032073975, -0.013908386, -0.05709839, 0.0062332153, -0.0008416176, 0.0113220215, -0.058624268, 0.008460999, -0.014266968, -0.025985718, 0.0126953125, -0.03463745, -0.0053634644,...
http://arxiv.org/abs/2309.16844v2
DeBERTinha: A Multistep Approach to Adapt DebertaV3 XSmall for Brazilian Portuguese Natural Language Processing Task
Israel Campiotti, Matheus Rodrigues, Yuri Albuquerque, Rafael Azevedo, Alyson Andrade
cs.CL
This paper presents an approach for adapting the DebertaV3 XSmall model pre-trained in English for Brazilian Portuguese natural language processing (NLP) tasks. A key aspect of the methodology involves a multistep training process to ensure the model is effectively tuned for the Portuguese language. Initial datasets fr...
http://arxiv.org/pdf/2309.16844v2
2023-10-31T00:14:18
2023-09-28T20:53:25
{ "title": [ -0.0357666, 0.00844574, -0.039886475, -0.05960083, -0.012580872, 0.02482605, -0.09765625, -0.0026283264, 0.018600464, 0.04977417, -0.01953125, -0.03149414, 0.045196533, -0.021987915, -0.042053223, 0.010971069, -0.02609253, 0.06622314...
http://arxiv.org/abs/2309.16804v1
Curriculum-Driven Edubot: A Framework for Developing Language Learning Chatbots Through Synthesizing Conversational Data
Yu Li, Shang Qu, Jili Shen, Shangchao Min, Zhou Yu
cs.CL
Chatbots have become popular in educational settings, revolutionizing how students interact with material and how teachers teach. We present Curriculum-Driven EduBot, a framework for developing a chatbot that combines the interactive features of chatbots with the systematic material of English textbooks to assist stude...
http://arxiv.org/pdf/2309.16804v1
2023-09-28T19:14:18
2023-09-28T19:14:18
{ "title": [ 0.0061912537, -0.041137695, -0.0054512024, -0.029022217, -0.039031982, -0.006210327, -0.0095825195, -0.00422287, -0.024429321, 0.047027588, -0.01739502, 0.043151855, 0.0044555664, -0.00079870224, -0.0126953125, -0.0066719055, -0.02220153...
http://arxiv.org/abs/2309.16797v1
Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
Chrisantha Fernando, Dylan Banarse, Henryk Michalewski, Simon Osindero, Tim Rocktäschel
cs.CL, cs.AI, cs.LG, cs.NE
Popular prompt strategies like Chain-of-Thought Prompting can dramatically improve the reasoning abilities of Large Language Models (LLMs) in various domains. However, such hand-crafted prompt-strategies are often sub-optimal. In this paper, we present Promptbreeder, a general-purpose self-referential self-improvement ...
http://arxiv.org/pdf/2309.16797v1
2023-09-28T19:01:07
2023-09-28T19:01:07
{ "title": [ -0.0060691833, -0.004699707, -0.037322998, -0.03778076, 0.0056381226, -0.014862061, 0.0047569275, -0.005405426, -0.025375366, 0.024032593, 0.014762878, 0.016418457, -0.016311646, -0.04095459, -0.013656616, 0.018173218, 0.007873535, 0...
http://arxiv.org/abs/2309.16781v1
Hallucination Reduction in Long Input Text Summarization
Tohida Rehman, Ronit Mandal, Abhishek Agarwal, Debarshi Kumar Sanyal
cs.CL, cs.IR, cs.LG
Hallucination in text summarization refers to the phenomenon where the model generates information that is not supported by the input source document. Hallucination poses significant obstacles to the accuracy and reliability of the generated summaries. In this paper, we aim to reduce hallucinated outputs or hallucinati...
http://arxiv.org/pdf/2309.16781v1
2023-09-28T18:22:16
2023-09-28T18:22:16
{ "title": [ 0.01966858, 0.012382507, -0.007080078, -0.05230713, -0.007160187, -0.008201599, -0.031921387, -0.022598267, 0.013656616, 0.060821533, 0.00070381165, 0.012031555, 0.010932922, -0.0005931854, -0.07446289, 0.011444092, -0.015380859, 0.0...
http://arxiv.org/abs/2309.16777v1
How many words does ChatGPT know? The answer is ChatWords
Gonzalo Martínez, Javier Conde, Pedro Reviriego, Elena Merino-Gómez, José Alberto Hernández, Fabrizio Lombardi
cs.CL, cs.AI
The introduction of ChatGPT has put Artificial Intelligence (AI) Natural Language Processing (NLP) in the spotlight. ChatGPT adoption has been exponential with millions of users experimenting with it in a myriad of tasks and application domains with impressive results. However, ChatGPT has limitations and suffers hallu...
http://arxiv.org/pdf/2309.16777v1
2023-09-28T18:13:02
2023-09-28T18:13:02
{ "title": [ -0.06976318, -0.021820068, -0.028549194, 0.02267456, -0.062927246, -0.007282257, -0.023513794, -0.021270752, 0.0037631989, 0.010154724, -0.02444458, 0.02255249, -0.023345947, -0.012893677, -0.0020637512, -0.004169464, 0.009292603, -0...
http://arxiv.org/abs/2309.16770v1
Persona-Coded Poly-Encoder: Persona-Guided Multi-Stream Conversational Sentence Scoring
Junfeng Liu, Christopher Symons, Ranga Raju Vatsavai
cs.CL, cs.AI, cs.LG
Recent advances in machine learning and deep learning have led to the widespread use of Conversational AI in many practical applications. However, it is still very challenging to leverage auxiliary information that can provide conversational context or personalized tuning to improve the quality of conversations. For ex...
http://arxiv.org/pdf/2309.16770v1
2023-09-28T18:07:01
2023-09-28T18:07:01
{ "title": [ -0.008224487, 0.0036468506, -0.020477295, -0.055603027, -0.015098572, 0.005264282, -0.045318604, -0.00019085407, 0.021713257, 0.029174805, -0.019073486, 0.023101807, 0.016784668, -0.010826111, -0.040161133, 0.014160156, -0.03366089, ...
http://arxiv.org/abs/2309.16671v3
Demystifying CLIP Data
Hu Xu, Saining Xie, Xiaoqing Ellen Tan, Po-Yao Huang, Russell Howes, Vasu Sharma, Shang-Wen Li, Gargi Ghosh, Luke Zettlemoyer, Christoph Feichtenhofer
cs.CV, cs.CL
Contrastive Language-Image Pre-training (CLIP) is an approach that has advanced research and applications in computer vision, fueling modern recognition systems and generative models. We believe that the main ingredient to the success of CLIP is its data and not the model architecture or pre-training objective. However...
http://arxiv.org/pdf/2309.16671v3
2023-10-02T07:12:53
2023-09-28T17:59:56
{ "title": [ 0.018569946, -0.0041542053, -0.030670166, -0.066101074, -0.038848877, -0.014862061, -0.048583984, 0.019821167, -0.039215088, 0.009010315, -0.021957397, -0.021270752, -0.059661865, -0.020385742, 0.019897461, -0.044281006, 0.05355835, ...
http://arxiv.org/abs/2309.16639v1
MindShift: Leveraging Large Language Models for Mental-States-Based Problematic Smartphone Use Intervention
Ruolan Wu, Chun Yu, Xiaole Pan, Yujia Liu, Ningning Zhang, Yue Fu, Yuhan Wang, Zhi Zheng, Li Chen, Qiaolei Jiang, Xuhai Xu, Yuanchun Shi
cs.CL, cs.AI, cs.HC, 68U35, H.5.2; I.2.7
Problematic smartphone use negatively affects physical and mental health. Despite the wide range of prior research, existing persuasive techniques are not flexible enough to provide dynamic persuasion content based on users' physical contexts and mental states. We first conduct a Wizard-of-Oz study (N=12) and an interv...
http://arxiv.org/pdf/2309.16639v1
2023-09-28T17:49:03
2023-09-28T17:49:03
{ "title": [ 0.0184021, -0.011421204, -0.014038086, -0.042663574, 0.006126404, -0.007873535, -0.048187256, -0.011779785, -0.00020778179, 0.015213013, -0.026489258, 0.0096206665, -0.01751709, -0.036132812, -0.02407837, -0.013175964, -0.021881104, ...
http://arxiv.org/abs/2309.16621v1
Stress Testing Chain-of-Thought Prompting for Large Language Models
Aayush Mishra, Karan Thakkar
cs.CL, cs.AI
This report examines the effectiveness of Chain-of-Thought (CoT) prompting in improving the multi-step reasoning abilities of large language models (LLMs). Inspired by previous studies \cite{Min2022RethinkingWork}, we analyze the impact of three types of CoT prompt perturbations, namely CoT order, CoT values, and CoT o...
http://arxiv.org/pdf/2309.16621v1
2023-09-28T17:21:33
2023-09-28T17:21:33
{ "title": [ -0.012886047, -0.01676941, -0.021270752, -0.059020996, -0.031829834, 0.0012044907, -0.05947876, 0.000320673, 0.015350342, -0.0028018951, -0.038360596, -0.010406494, 0.010948181, -0.027755737, -0.022201538, -0.0037784576, 0.0018072128, ...
http://arxiv.org/abs/2309.16609v1
Qwen Technical Report
Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, Binyuan Hui, Luo Ji, Mei Li, Junyang Lin, Runji Lin, Dayiheng Liu, Gao Liu, Chengqiang Lu, Keming Lu, Jianxin Ma, Rui Men, Xingzhang Ren, Xuancheng Ren, Chuanqi Tan, Sinan Tan, Jianhong Tu, Peng Wang, Shijie Wan...
cs.CL
Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural language processing tasks that were previously thought to be exclusive to humans. In this work, we introduce Qwen, the first installment of our large language model series. Qwen is a comprehensive language model seri...
http://arxiv.org/pdf/2309.16609v1
2023-09-28T17:07:49
2023-09-28T17:07:49
{ "title": [ 0.006374359, -0.0054740906, -0.0020751953, 0.0095825195, -0.029800415, -0.01512146, -0.007850647, -0.034210205, -0.06359863, 0.077819824, 0.0028953552, -0.020492554, 0.033843994, -0.023223877, 0.014755249, -0.022705078, 0.026992798, ...
http://arxiv.org/abs/2309.16599v1
Unlikelihood Tuning on Negative Samples Amazingly Improves Zero-Shot Translation
Changtong Zan, Liang Ding, Li Shen, Yibin Lei, Yibing Zhan, Weifeng Liu, Dacheng Tao
cs.CL
Zero-shot translation (ZST), which is generally based on a multilingual neural machine translation model, aims to translate between unseen language pairs in training data. The common practice to guide the zero-shot language mapping during inference is to deliberately insert the source and target language IDs, e.g., <EN...
http://arxiv.org/pdf/2309.16599v1
2023-09-28T17:02:36
2023-09-28T17:02:36
{ "title": [ 0.04034424, -0.033691406, 0.0021419525, -0.037017822, -0.042297363, -0.02609253, -0.006008148, 0.0016889572, -0.0019893646, 0.049224854, -0.033233643, 0.036315918, 0.018859863, -0.03062439, -0.0057907104, -0.004722595, -0.017028809, ...
http://arxiv.org/abs/2309.16583v3
GPT-Fathom: Benchmarking Large Language Models to Decipher the Evolutionary Path towards GPT-4 and Beyond
Shen Zheng, Yuyu Zhang, Yijie Zhu, Chenguang Xi, Pengyang Gao, Xun Zhou, Kevin Chen-Chuan Chang
cs.CL
With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage che...
http://arxiv.org/pdf/2309.16583v3
2023-10-23T09:19:09
2023-09-28T16:43:35
{ "title": [ -0.041809082, -0.013908386, -0.023880005, -0.041992188, -0.004638672, 0.012176514, -0.04824829, -0.0078125, 0.0047569275, 0.012710571, 0.0041656494, 0.010025024, -0.010032654, -0.029006958, -0.035736084, 0.020614624, -0.013786316, 0....
http://arxiv.org/abs/2309.16575v1
A Benchmark for Learning to Translate a New Language from One Grammar Book
Garrett Tanzer, Mirac Suzgun, Eline Visser, Dan Jurafsky, Luke Melas-Kyriazi
cs.CL
Large language models (LLMs) can perform impressive feats with in-context learning or lightweight finetuning. It is natural to wonder how well these models adapt to genuinely new tasks, but how does one find tasks that are unseen in internet-scale training sets? We turn to a field that is explicitly motivated and bottl...
http://arxiv.org/pdf/2309.16575v1
2023-09-28T16:32:28
2023-09-28T16:32:28
{ "title": [ -0.018310547, -0.025924683, -0.024429321, 0.0011482239, -0.056610107, 0.0010375977, -0.018569946, 0.014625549, -0.045135498, 0.02998352, -0.006778717, 0.05303955, -0.021575928, -0.034057617, 0.032714844, -0.022140503, 0.019317627, 0....