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/2304.04576v1
Learnings from Data Integration for Augmented Language Models
Alon Halevy, Jane Dwivedi-Yu
cs.CL
One of the limitations of large language models is that they do not have access to up-to-date, proprietary or personal data. As a result, there are multiple efforts to extend language models with techniques for accessing external data. In that sense, LLMs share the vision of data integration systems whose goal is to pr...
http://arxiv.org/pdf/2304.04576v1
2023-04-10T13:28:35
2023-04-10T13:28:35
{ "title": [ -0.01626587, -0.00015556812, -0.049194336, -0.08074951, 0.0063476562, 0.0004813671, -0.064575195, 0.014549255, 0.002960205, 0.031921387, -0.027526855, -0.0020141602, -0.005126953, -0.014389038, -0.043060303, -0.0007967949, -0.018035889, ...
http://arxiv.org/abs/2304.04539v1
UATTA-EB: Uncertainty-Aware Test-Time Augmented Ensemble of BERTs for Classifying Common Mental Illnesses on Social Media Posts
Pratinav Seth, Mihir Agarwal
cs.CL, cs.AI, cs.LG
Given the current state of the world, because of existing situations around the world, millions of people suffering from mental illnesses feel isolated and unable to receive help in person. Psychological studies have shown that our state of mind can manifest itself in the linguistic features we use to communicate. Peop...
http://arxiv.org/pdf/2304.04539v1
2023-04-10T12:18:53
2023-04-10T12:18:53
{ "title": [ 0.047668457, 0.0012521744, 0.0032958984, -0.06555176, 0.019378662, -0.007507324, -0.014709473, -0.00957489, 0.026443481, 0.027313232, -0.03173828, 0.027252197, 0.0340271, -0.026351929, -0.03161621, 0.00749588, -0.013114929, 0.0459289...
http://arxiv.org/abs/2304.04487v1
Inference with Reference: Lossless Acceleration of Large Language Models
Nan Yang, Tao Ge, Liang Wang, Binxing Jiao, Daxin Jiang, Linjun Yang, Rangan Majumder, Furu Wei
cs.CL, cs.AI
We propose LLMA, an LLM accelerator to losslessly speed up Large Language Model (LLM) inference with references. LLMA is motivated by the observation that there are abundant identical text spans between the decoding result by an LLM and the reference that is available in many real world scenarios (e.g., retrieved docum...
http://arxiv.org/pdf/2304.04487v1
2023-04-10T09:55:14
2023-04-10T09:55:14
{ "title": [ -0.035003662, -0.026412964, -0.05581665, -0.07873535, 0.011199951, 0.0020961761, -0.056762695, 0.025985718, 0.0066947937, 0.025512695, -0.01474762, -0.019515991, 0.024276733, -0.020614624, -0.041107178, 0.013336182, -0.0013036728, 0....
http://arxiv.org/abs/2304.04478v1
Oh, Jeez! or Uh-huh? A Listener-aware Backchannel Predictor on ASR Transcriptions
Daniel Ortega, Chia-Yu Li, Ngoc Thang Vu
cs.CL, cs.SD, eess.AS
This paper presents our latest investigation on modeling backchannel in conversations. Motivated by a proactive backchanneling theory, we aim at developing a system which acts as a proactive listener by inserting backchannels, such as continuers and assessment, to influence speakers. Our model takes into account not on...
http://arxiv.org/pdf/2304.04478v1
2023-04-10T09:33:29
2023-04-10T09:33:29
{ "title": [ -0.00081920624, -0.009780884, -0.011352539, -0.058258057, -0.0541687, 0.0029640198, -0.08526611, 0.012199402, 0.005302429, 0.04244995, 0.00579834, 0.0011634827, 0.02835083, -0.05532837, -0.0071868896, 0.009391785, 0.009765625, 0.0383...
http://arxiv.org/abs/2304.04472v1
Modeling Speaker-Listener Interaction for Backchannel Prediction
Daniel Ortega, Sarina Meyer, Antje Schweitzer, Ngoc Thang Vu
cs.CL
We present our latest findings on backchannel modeling novelly motivated by the canonical use of the minimal responses Yeah and Uh-huh in English and their correspondent tokens in German, and the effect of encoding the speaker-listener interaction. Backchanneling theories emphasize the active and continuous role of the...
http://arxiv.org/pdf/2304.04472v1
2023-04-10T09:22:06
2023-04-10T09:22:06
{ "title": [ 0.0016145706, 0.020553589, -0.020126343, -0.052581787, -0.011459351, 0.0045928955, -0.06945801, -0.034484863, -0.0017719269, 0.036987305, 0.017288208, -0.012580872, 0.0074310303, -0.042022705, -0.041625977, 0.006175995, -0.005683899, ...
http://arxiv.org/abs/2304.05351v2
The Wall Street Neophyte: A Zero-Shot Analysis of ChatGPT Over MultiModal Stock Movement Prediction Challenges
Qianqian Xie, Weiguang Han, Yanzhao Lai, Min Peng, Jimin Huang
cs.CL, cs.LG, q-fin.ST
Recently, large language models (LLMs) like ChatGPT have demonstrated remarkable performance across a variety of natural language processing tasks. However, their effectiveness in the financial domain, specifically in predicting stock market movements, remains to be explored. In this paper, we conduct an extensive zero...
http://arxiv.org/pdf/2304.05351v2
2023-04-28T12:06:43
2023-04-10T04:31:00
{ "title": [ 0.0028972626, 0.0014181137, -0.027267456, -0.0037155151, -0.016555786, -0.021377563, -0.0044288635, -0.014808655, 0.0008430481, 0.038635254, -0.0012226105, 0.010505676, 0.0074272156, -0.024353027, -0.031173706, -0.0033435822, -0.00581359...
http://arxiv.org/abs/2304.04370v6
OpenAGI: When LLM Meets Domain Experts
Yingqiang Ge, Wenyue Hua, Kai Mei, Jianchao Ji, Juntao Tan, Shuyuan Xu, Zelong Li, Yongfeng Zhang
cs.AI, cs.CL, cs.LG
Human Intelligence (HI) excels at combining basic skills to solve complex tasks. This capability is vital for Artificial Intelligence (AI) and should be embedded in comprehensive AI Agents, enabling them to harness expert models for complex task-solving towards Artificial General Intelligence (AGI). Large Language Mode...
http://arxiv.org/pdf/2304.04370v6
2023-11-03T15:24:18
2023-04-10T03:55:35
{ "title": [ -0.026687622, 0.008605957, -0.06378174, -0.02268982, -0.028823853, -0.0071868896, -0.062408447, -0.014228821, -0.038360596, 0.020690918, -0.016143799, -0.0026397705, 0.007987976, -0.052581787, 0.023086548, -0.029663086, -0.011428833, ...
http://arxiv.org/abs/2304.04358v1
WebBrain: Learning to Generate Factually Correct Articles for Queries by Grounding on Large Web Corpus
Hongjing Qian, Yutao Zhu, Zhicheng Dou, Haoqi Gu, Xinyu Zhang, Zheng Liu, Ruofei Lai, Zhao Cao, Jian-Yun Nie, Ji-Rong Wen
cs.CL, cs.AI
In this paper, we introduce a new NLP task -- generating short factual articles with references for queries by mining supporting evidence from the Web. In this task, called WebBrain, the ultimate goal is to generate a fluent, informative, and factually-correct short article (e.g., a Wikipedia article) for a factual que...
http://arxiv.org/pdf/2304.04358v1
2023-04-10T02:55:48
2023-04-10T02:55:48
{ "title": [ 0.00207901, 0.0021419525, -0.052764893, -0.045013428, 0.026260376, -0.026550293, -0.01210022, -0.011329651, 0.0051956177, 0.02947998, -0.041046143, 0.021118164, 0.0016174316, 0.007045746, -0.04107666, -0.008705139, -0.019119263, 0.05...
http://arxiv.org/abs/2304.04339v1
Is ChatGPT a Good Sentiment Analyzer? A Preliminary Study
Zengzhi Wang, Qiming Xie, Zixiang Ding, Yi Feng, Rui Xia
cs.CL, cs.AI
Recently, ChatGPT has drawn great attention from both the research community and the public. We are particularly curious about whether it can serve as a universal sentiment analyzer. To this end, in this work, we provide a preliminary evaluation of ChatGPT on the understanding of opinions, sentiments, and emotions cont...
http://arxiv.org/pdf/2304.04339v1
2023-04-10T00:55:59
2023-04-10T00:55:59
{ "title": [ -0.000061392784, -0.013641357, -0.01020813, -0.021316528, 0.00035309792, 0.00062704086, -0.003791809, -0.025466919, -0.006061554, 0.03643799, -0.0184021, 0.009155273, -0.0070228577, -0.0050849915, -0.05706787, 0.0061798096, -0.0234375, ...
http://arxiv.org/abs/2304.04321v2
ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenes
Ran Gong, Jiangyong Huang, Yizhou Zhao, Haoran Geng, Xiaofeng Gao, Qingyang Wu, Wensi Ai, Ziheng Zhou, Demetri Terzopoulos, Song-Chun Zhu, Baoxiong Jia, Siyuan Huang
cs.AI, cs.CL, cs.CV, cs.RO
Understanding the continuous states of objects is essential for task learning and planning in the real world. However, most existing task learning benchmarks assume discrete (e.g., binary) object goal states, which poses challenges for the learning of complex tasks and transferring learned policy from simulated environ...
http://arxiv.org/pdf/2304.04321v2
2023-09-11T11:27:53
2023-04-09T21:42:57
{ "title": [ -0.035217285, 0.008255005, -0.05392456, -0.026870728, -0.008972168, -0.0362854, -0.038146973, 0.003730774, 0.0024375916, 0.010437012, -0.013839722, 0.02279663, -0.020477295, -0.02420044, -0.02670288, -0.00630188, -0.00068616867, 0.03...
http://arxiv.org/abs/2304.04280v1
FrenchMedMCQA: A French Multiple-Choice Question Answering Dataset for Medical domain
Yanis Labrak, Adrien Bazoge, Richard Dufour, Mickael Rouvier, Emmanuel Morin, Béatrice Daille, Pierre-Antoine Gourraud
cs.CL, cs.AI
This paper introduces FrenchMedMCQA, the first publicly available Multiple-Choice Question Answering (MCQA) dataset in French for medical domain. It is composed of 3,105 questions taken from real exams of the French medical specialization diploma in pharmacy, mixing single and multiple answers. Each instance of the dat...
http://arxiv.org/pdf/2304.04280v1
2023-04-09T16:57:40
2023-04-09T16:57:40
{ "title": [ -0.04272461, -0.010986328, -0.051513672, -0.0463562, 0.009414673, 0.0058174133, -0.02796936, 0.0077590942, 0.013999939, 0.024230957, 0.0035076141, -0.01689148, -0.004562378, -0.029144287, -0.054138184, 0.0068092346, -0.0066070557, 0....
http://arxiv.org/abs/2304.05368v3
Are Large Language Models Ready for Healthcare? A Comparative Study on Clinical Language Understanding
Yuqing Wang, Yun Zhao, Linda Petzold
cs.CL, cs.AI
Large language models (LLMs) have made significant progress in various domains, including healthcare. However, the specialized nature of clinical language understanding tasks presents unique challenges and limitations that warrant further investigation. In this study, we conduct a comprehensive evaluation of state-of-t...
http://arxiv.org/pdf/2304.05368v3
2023-07-30T19:09:02
2023-04-09T16:31:47
{ "title": [ -0.03579712, -0.011230469, -0.03567505, -0.080322266, -0.0024433136, 0.011924744, -0.07495117, 0.0049743652, -0.013557434, 0.019515991, -0.011116028, -0.0079193115, -0.003080368, -0.025939941, -0.030639648, -0.004951477, -0.008918762, ...
http://arxiv.org/abs/2304.04256v1
A Preliminary Evaluation of ChatGPT for Zero-shot Dialogue Understanding
Wenbo Pan, Qiguang Chen, Xiao Xu, Wanxiang Che, Libo Qin
cs.CL
Zero-shot dialogue understanding aims to enable dialogue to track the user's needs without any training data, which has gained increasing attention. In this work, we investigate the understanding ability of ChatGPT for zero-shot dialogue understanding tasks including spoken language understanding (SLU) and dialogue sta...
http://arxiv.org/pdf/2304.04256v1
2023-04-09T15:28:36
2023-04-09T15:28:36
{ "title": [ -0.0058631897, -0.023529053, -0.016174316, -0.047546387, -0.01158905, -0.003540039, -0.029327393, 0.015197754, 0.003900528, 0.024246216, -0.0013742447, 0.00920105, -0.017822266, -0.026672363, -0.0435791, 0.0031490326, -0.026107788, 0...
http://arxiv.org/abs/2304.04250v3
Editable User Profiles for Controllable Text Recommendation
Sheshera Mysore, Mahmood Jasim, Andrew McCallum, Hamed Zamani
cs.IR, cs.CL, cs.HC, cs.LG
Methods for making high-quality recommendations often rely on learning latent representations from interaction data. These methods, while performant, do not provide ready mechanisms for users to control the recommendation they receive. Our work tackles this problem by proposing LACE, a novel concept value bottleneck mo...
http://arxiv.org/pdf/2304.04250v3
2023-10-16T21:47:20
2023-04-09T14:52:18
{ "title": [ 0.0098724365, 0.0056495667, -0.024765015, -0.03955078, -0.007385254, -0.00030636787, -0.003019333, -0.015335083, -0.038970947, 0.049713135, -0.014472961, 0.038635254, 0.0015649796, -0.009460449, -0.017349243, 0.0022659302, -0.017974854, ...
http://arxiv.org/abs/2304.04212v1
RISC: Generating Realistic Synthetic Bilingual Insurance Contract
David Beauchemin, Richard Khoury
cs.CL, cs.LG
This paper presents RISC, an open-source Python package data generator (https://github.com/GRAAL-Research/risc). RISC generates look-alike automobile insurance contracts based on the Quebec regulatory insurance form in French and English. Insurance contracts are 90 to 100 pages long and use complex legal and insurance-...
http://arxiv.org/pdf/2304.04212v1
2023-04-09T10:42:18
2023-04-09T10:42:18
{ "title": [ -0.0791626, -0.020477295, -0.035827637, -0.049468994, -0.026138306, 0.028839111, -0.06719971, 0.0096206665, -0.036956787, 0.018417358, -0.019943237, 0.010017395, -0.02519226, -0.024093628, -0.00983429, -0.02432251, -0.03036499, 0.036...
http://arxiv.org/abs/2304.04193v2
Extractive Summarization via ChatGPT for Faithful Summary Generation
Haopeng Zhang, Xiao Liu, Jiawei Zhang
cs.CL
Extractive summarization is a crucial task in natural language processing that aims to condense long documents into shorter versions by directly extracting sentences. The recent introduction of large language models has attracted significant interest in the NLP community due to its remarkable performance on a wide rang...
http://arxiv.org/pdf/2304.04193v2
2023-10-09T23:40:26
2023-04-09T08:26:04
{ "title": [ -0.018203735, -0.007019043, -0.035705566, -0.077697754, -0.025314331, 0.021743774, -0.07647705, -0.006401062, -0.0063476562, 0.05734253, -0.0059776306, 0.04168701, -0.010925293, 0.0047454834, -0.05706787, 0.021331787, -0.012107849, 0...
http://arxiv.org/abs/2304.04190v1
Team QUST at SemEval-2023 Task 3: A Comprehensive Study of Monolingual and Multilingual Approaches for Detecting Online News Genre, Framing and Persuasion Techniques
Ye Jiang
cs.CL, cs.AI
This paper describes the participation of team QUST in the SemEval2023 task 3. The monolingual models are first evaluated with the under-sampling of the majority classes in the early stage of the task. Then, the pre-trained multilingual model is fine-tuned with a combination of the class weights and the sample weights....
http://arxiv.org/pdf/2304.04190v1
2023-04-09T08:14:01
2023-04-09T08:14:01
{ "title": [ 0.02407837, 0.01763916, -0.03277588, -0.0057792664, 0.009719849, -0.02243042, -0.016998291, -0.028930664, -0.023849487, 0.046722412, 0.0049438477, 0.05340576, 0.0011453629, -0.021362305, 0.013847351, 0.01411438, -0.023895264, 0.02752...
http://arxiv.org/abs/2304.04187v3
Similarity-Aware Multimodal Prompt Learning for Fake News Detection
Ye Jiang, Xiaomin Yu, Yimin Wang, Xiaoman Xu, Xingyi Song, Diana Maynard
cs.CL
The standard paradigm for fake news detection mainly utilizes text information to model the truthfulness of news. However, the discourse of online fake news is typically subtle and it requires expert knowledge to use textual information to debunk fake news. Recently, studies focusing on multimodal fake news detection h...
http://arxiv.org/pdf/2304.04187v3
2023-06-16T12:05:57
2023-04-09T08:10:05
{ "title": [ -0.0021629333, 0.026824951, -0.04360962, -0.034362793, 0.018936157, -0.016403198, -0.017486572, -0.020599365, -0.027526855, 0.015213013, -0.032989502, 0.018371582, 0.006122589, -0.022842407, 0.007736206, -0.020095825, -0.035614014, 0...
http://arxiv.org/abs/2304.05372v1
Can ChatGPT and Bard Generate Aligned Assessment Items? A Reliability Analysis against Human Performance
Abdolvahab Khademi
cs.CL, cs.AI
ChatGPT and Bard are AI chatbots based on Large Language Models (LLM) that are slated to promise different applications in diverse areas. In education, these AI technologies have been tested for applications in assessment and teaching. In assessment, AI has long been used in automated essay scoring and automated item g...
http://arxiv.org/pdf/2304.05372v1
2023-04-09T04:53:15
2023-04-09T04:53:15
{ "title": [ 0.0063705444, -0.0035037994, -0.012184143, 0.0023097992, -0.015625, -0.015014648, 0.0129776, -0.025390625, -0.0075416565, 0.03616333, -0.0021400452, 0.0043754578, -0.020065308, -0.019699097, -0.06237793, -0.0051078796, -0.045166016, ...
http://arxiv.org/abs/2304.04157v2
An investigation of speaker independent phrase break models in End-to-End TTS systems
Anandaswarup Vadapalli
eess.AS, cs.CL, cs.SD
This paper presents our work on phrase break prediction in the context of end-to-end TTS systems, motivated by the following questions: (i) Is there any utility in incorporating an explicit phrasing model in an end-to-end TTS system?, and (ii) How do you evaluate the effectiveness of a phrasing model in an end-to-end T...
http://arxiv.org/pdf/2304.04157v2
2023-04-21T05:03:27
2023-04-09T04:26:58
{ "title": [ -0.0018663406, 0.015380859, -0.026809692, -0.055358887, 0.0051994324, -0.0041503906, -0.04336548, -0.005126953, -0.007331848, 0.049926758, -0.0056495667, 0.0027694702, 0.035949707, -0.020111084, -0.039245605, 0.034729004, -0.01096344, ...
http://arxiv.org/abs/2304.04152v1
Continual Graph Convolutional Network for Text Classification
Tiandeng Wu, Qijiong Liu, Yi Cao, Yao Huang, Xiao-Ming Wu, Jiandong Ding
cs.CL, I.7.0
Graph convolutional network (GCN) has been successfully applied to capture global non-consecutive and long-distance semantic information for text classification. However, while GCN-based methods have shown promising results in offline evaluations, they commonly follow a seen-token-seen-document paradigm by constructing...
http://arxiv.org/pdf/2304.04152v1
2023-04-09T03:59:48
2023-04-09T03:59:48
{ "title": [ -0.014839172, -0.0061950684, -0.045410156, -0.06274414, -0.0028190613, -0.0058555603, -0.043548584, 0.0030536652, 0.011993408, 0.024337769, -0.014450073, 0.021316528, -0.011817932, -0.012435913, -0.04034424, 0.007041931, -0.016296387, ...
http://arxiv.org/abs/2304.11093v1
Hi Sheldon! Creating Deep Personalized Characters from TV Shows
Meidai Xuanyuan, Yuwang Wang, Honglei Guo, Xiao Ma, Yuchen Guo, Tao Yu, Qionghai Dai
cs.CL, cs.AI, cs.CV, cs.LG
Imagine an interesting multimodal interactive scenario that you can see, hear, and chat with an AI-generated digital character, who is capable of behaving like Sheldon from The Big Bang Theory, as a DEEP copy from appearance to personality. Towards this fantastic multimodal chatting scenario, we propose a novel task, n...
http://arxiv.org/pdf/2304.11093v1
2023-04-09T00:39:43
2023-04-09T00:39:43
{ "title": [ 0.03552246, -0.013504028, -0.024765015, -0.018447876, -0.02758789, -0.008491516, 0.01399231, 0.023727417, -0.036224365, 0.03741455, 0.0056037903, 0.008728027, 0.039093018, -0.054870605, 0.025421143, -0.0059547424, -0.0061950684, 0.03...
http://arxiv.org/abs/2304.04118v1
Multi-class Categorization of Reasons behind Mental Disturbance in Long Texts
Muskan Garg
cs.CL, cs.CY
Motivated with recent advances in inferring users' mental state in social media posts, we identify and formulate the problem of finding causal indicators behind mental illness in self-reported text. In the past, we witness the presence of rule-based studies for causal explanation analysis on curated Facebook data. The ...
http://arxiv.org/pdf/2304.04118v1
2023-04-08T22:44:32
2023-04-08T22:44:32
{ "title": [ 0.03665161, -0.0006494522, -0.023513794, -0.09051514, -0.009803772, -0.026855469, -0.02520752, 0.009178162, -0.01939392, 0.0118637085, -0.011756897, 0.01914978, 0.01977539, -0.042266846, 0.019897461, -0.015449524, -0.008255005, 0.058...
http://arxiv.org/abs/2304.04099v3
Unsupervised Story Discovery from Continuous News Streams via Scalable Thematic Embedding
Susik Yoon, Dongha Lee, Yunyi Zhang, Jiawei Han
cs.IR, cs.CL, cs.DB, cs.LG
Unsupervised discovery of stories with correlated news articles in real-time helps people digest massive news streams without expensive human annotations. A common approach of the existing studies for unsupervised online story discovery is to represent news articles with symbolic- or graph-based embedding and increment...
http://arxiv.org/pdf/2304.04099v3
2023-05-04T04:36:23
2023-04-08T20:41:15
{ "title": [ -0.007434845, 0.0050354004, -0.036254883, -0.04840088, 0.0113220215, -0.0079193115, -0.020324707, -0.0127334595, 0.01574707, 0.02810669, -0.053771973, 0.018997192, -0.017929077, 0.0018396378, -0.041046143, -0.016357422, -0.0056419373, ...
http://arxiv.org/abs/2304.04087v1
Interpretable Multi Labeled Bengali Toxic Comments Classification using Deep Learning
Tanveer Ahmed Belal, G. M. Shahariar, Md. Hasanul Kabir
cs.CL, cs.AI, cs.LG
This paper presents a deep learning-based pipeline for categorizing Bengali toxic comments, in which at first a binary classification model is used to determine whether a comment is toxic or not, and then a multi-label classifier is employed to determine which toxicity type the comment belongs to. For this purpose, we ...
http://arxiv.org/pdf/2304.04087v1
2023-04-08T19:28:26
2023-04-08T19:28:26
{ "title": [ 0.0068855286, 0.00046539307, -0.016647339, -0.081726074, 0.011360168, 0.0018692017, -0.031463623, 0.0030593872, 0.0022907257, 0.01171875, -0.03262329, 0.002281189, -0.03149414, -0.012268066, -0.043151855, -0.04095459, -0.013938904, 0...
http://arxiv.org/abs/2304.04054v1
tmn at SemEval-2023 Task 9: Multilingual Tweet Intimacy Detection using XLM-T, Google Translate, and Ensemble Learning
Anna Glazkova
cs.CL, cs.AI, cs.IR, cs.LG, 68T50, I.2.7; I.7.m; H.3.3
The paper describes a transformer-based system designed for SemEval-2023 Task 9: Multilingual Tweet Intimacy Analysis. The purpose of the task was to predict the intimacy of tweets in a range from 1 (not intimate at all) to 5 (very intimate). The official training set for the competition consisted of tweets in six lang...
http://arxiv.org/pdf/2304.04054v1
2023-04-08T15:50:16
2023-04-08T15:50:16
{ "title": [ 0.004146576, -0.03390503, 0.010437012, -0.0044898987, -0.03945923, 0.007575989, -0.035095215, -0.020523071, 0.033325195, 0.059906006, -0.016860962, 0.04269409, 0.027526855, -0.011077881, -0.0055274963, 0.036468506, -0.01058197, 0.019...
http://arxiv.org/abs/2304.04052v1
Decoder-Only or Encoder-Decoder? Interpreting Language Model as a Regularized Encoder-Decoder
Zihao Fu, Wai Lam, Qian Yu, Anthony Man-Cho So, Shengding Hu, Zhiyuan Liu, Nigel Collier
cs.CL, cs.AI, cs.LG
The sequence-to-sequence (seq2seq) task aims at generating the target sequence based on the given input source sequence. Traditionally, most of the seq2seq task is resolved by the Encoder-Decoder framework which requires an encoder to encode the source sequence and a decoder to generate the target text. Recently, a bun...
http://arxiv.org/pdf/2304.04052v1
2023-04-08T15:44:29
2023-04-08T15:44:29
{ "title": [ -0.042877197, 0.00067329407, -0.05154419, -0.033935547, -0.03640747, -0.016433716, -0.06866455, -0.0060806274, -0.009895325, 0.016220093, -0.013923645, -0.011116028, 0.013389587, -0.030059814, -0.029159546, 0.00819397, -0.0029907227, ...
http://arxiv.org/abs/2304.04029v2
Bipol: A Novel Multi-Axes Bias Evaluation Metric with Explainability for NLP
Lama Alkhaled, Tosin Adewumi, Sana Sabah Sabry
cs.CL
We introduce bipol, a new metric with explainability, for estimating social bias in text data. Harmful bias is prevalent in many online sources of data that are used for training machine learning (ML) models. In a step to address this challenge we create a novel metric that involves a two-step process: corpus-level eva...
http://arxiv.org/pdf/2304.04029v2
2023-09-16T15:47:24
2023-04-08T14:45:15
{ "title": [ -0.03060913, -0.00881958, -0.020553589, -0.04937744, -0.005302429, 0.016799927, -0.04650879, -0.016967773, 0.010101318, -0.0052604675, -0.03717041, -0.0034236908, 0.012588501, -0.019989014, -0.015808105, -0.01737976, -0.05239868, 0.0...
http://arxiv.org/abs/2304.04026v1
WikiGoldSK: Annotated Dataset, Baselines and Few-Shot Learning Experiments for Slovak Named Entity Recognition
Dávid Šuba, Marek Šuppa, Jozef Kubík, Endre Hamerlik, Martin Takáč
cs.CL, cs.AI
Named Entity Recognition (NER) is a fundamental NLP tasks with a wide range of practical applications. The performance of state-of-the-art NER methods depends on high quality manually anotated datasets which still do not exist for some languages. In this work we aim to remedy this situation in Slovak by introducing Wik...
http://arxiv.org/pdf/2304.04026v1
2023-04-08T14:37:52
2023-04-08T14:37:52
{ "title": [ -0.019241333, -0.013137817, -0.057281494, -0.04574585, 0.007881165, -0.0063171387, -0.031982422, 0.0020809174, 0.0018262863, 0.047302246, -0.032043457, 0.012969971, 0.033203125, -0.010375977, -0.031280518, 0.00082588196, 0.004917145, ...
http://arxiv.org/abs/2304.03952v1
MphayaNER: Named Entity Recognition for Tshivenda
Rendani Mbuvha, David I. Adelani, Tendani Mutavhatsindi, Tshimangadzo Rakhuhu, Aluwani Mauda, Tshifhiwa Joshua Maumela, Andisani Masindi, Seani Rananga, Vukosi Marivate, Tshilidzi Marwala
cs.CL, cs.AI
Named Entity Recognition (NER) plays a vital role in various Natural Language Processing tasks such as information retrieval, text classification, and question answering. However, NER can be challenging, especially in low-resource languages with limited annotated datasets and tools. This paper adds to the effort of add...
http://arxiv.org/pdf/2304.03952v1
2023-04-08T08:03:58
2023-04-08T08:03:58
{ "title": [ -0.016952515, -0.029846191, -0.05670166, -0.07434082, 0.0014505386, -0.0021438599, -0.07330322, -0.005657196, 0.0021743774, 0.019729614, -0.029434204, 0.008628845, 0.016159058, -0.023361206, -0.027648926, 0.011177063, 0.0018968582, 0...
http://arxiv.org/abs/2304.03938v1
Comparing Code Explanations Created by Students and Large Language Models
Juho Leinonen, Paul Denny, Stephen MacNeil, Sami Sarsa, Seth Bernstein, Joanne Kim, Andrew Tran, Arto Hellas
cs.CY, cs.AI, cs.CL, cs.HC, cs.SE
Reasoning about code and explaining its purpose are fundamental skills for computer scientists. There has been extensive research in the field of computing education on the relationship between a student's ability to explain code and other skills such as writing and tracing code. In particular, the ability to describe ...
http://arxiv.org/pdf/2304.03938v1
2023-04-08T06:52:54
2023-04-08T06:52:54
{ "title": [ -0.014625549, -0.03564453, -0.0019664764, -0.03475952, -0.046875, -0.0042495728, -0.020568848, 0.011245728, -0.012504578, 0.013946533, -0.022369385, -0.0033092499, -0.014808655, -0.027816772, -0.0010385513, -0.025726318, -0.016616821, ...
http://arxiv.org/abs/2304.03899v1
An Empirical Study and Improvement for Speech Emotion Recognition
Zhen Wu, Yizhe Lu, Xinyu Dai
cs.CL, cs.SD, eess.AS
Multimodal speech emotion recognition aims to detect speakers' emotions from audio and text. Prior works mainly focus on exploiting advanced networks to model and fuse different modality information to facilitate performance, while neglecting the effect of different fusion strategies on emotion recognition. In this wor...
http://arxiv.org/pdf/2304.03899v1
2023-04-08T03:24:06
2023-04-08T03:24:06
{ "title": [ -0.0051994324, 0.0027103424, -0.04043579, -0.05709839, 0.021362305, -0.013175964, -0.02468872, 0.007434845, 0.016036987, 0.042297363, -0.023773193, -0.0051994324, 0.021850586, -0.014526367, -0.05999756, 0.011703491, -0.005378723, 0.0...
http://arxiv.org/abs/2304.03898v1
The Short Text Matching Model Enhanced with Knowledge via Contrastive Learning
Qiqiang Zhong, Mengmeng Cui, Hanjie Mai, Qiang Zhang, Shaohua Xu, Xiangzheng Liu, Yanlong Du
cs.CL, cs.AI
In recent years, short Text Matching tasks have been widely applied in the fields ofadvertising search and recommendation. The difficulty lies in the lack of semantic information and word ambiguity caused by the short length of the text. Previous works have introduced complement sentences or knowledge bases to provide ...
http://arxiv.org/pdf/2304.03898v1
2023-04-08T03:24:05
2023-04-08T03:24:05
{ "title": [ -0.012207031, 0.01007843, -0.025146484, -0.035125732, 0.0055122375, -0.0104522705, -0.013755798, -0.032684326, 0.021438599, 0.03878784, -0.029251099, 0.009086609, 0.0013189316, -0.01777649, -0.052642822, 0.0026512146, -0.0030269623, ...
http://arxiv.org/abs/2304.03897v2
Factify 2: A Multimodal Fake News and Satire News Dataset
S Suryavardan, Shreyash Mishra, Parth Patwa, Megha Chakraborty, Anku Rani, Aishwarya Reganti, Aman Chadha, Amitava Das, Amit Sheth, Manoj Chinnakotla, Asif Ekbal, Srijan Kumar
cs.CL, cs.CV
The internet gives the world an open platform to express their views and share their stories. While this is very valuable, it makes fake news one of our society's most pressing problems. Manual fact checking process is time consuming, which makes it challenging to disprove misleading assertions before they cause signif...
http://arxiv.org/pdf/2304.03897v2
2023-10-02T14:48:45
2023-04-08T03:14:19
{ "title": [ -0.011703491, 0.007820129, -0.045837402, -0.054748535, 0.016189575, -0.006866455, -0.031311035, -0.018249512, -0.025558472, 0.027359009, -0.012916565, 0.019927979, -0.0023956299, -0.012741089, -0.023101807, -0.007293701, -0.033996582, ...
http://arxiv.org/abs/2304.03843v3
Why think step by step? Reasoning emerges from the locality of experience
Ben Prystawski, Michael Y. Li, Noah D. Goodman
cs.AI, cs.CL, cs.LG
Humans have a powerful and mysterious capacity to reason. Working through a set of mental steps enables us to make inferences we would not be capable of making directly even though we get no additional data from the world. Similarly, when large language models generate intermediate steps (a chain of thought) before ans...
http://arxiv.org/pdf/2304.03843v3
2023-11-02T20:32:43
2023-04-07T21:04:03
{ "title": [ 0.020599365, -0.001042366, -0.044647217, -0.07098389, 0.007850647, -0.025802612, -0.030807495, 0.016799927, -0.005027771, 0.017044067, -0.016540527, -0.0008916855, -0.012306213, -0.039642334, -0.0023059845, -0.019851685, 0.024261475, ...
http://arxiv.org/abs/2304.04758v1
Expectations over Unspoken Alternatives Predict Pragmatic Inferences
Jennifer Hu, Roger Levy, Judith Degen, Sebastian Schuster
cs.CL
Scalar inferences (SI) are a signature example of how humans interpret language based on unspoken alternatives. While empirical studies have demonstrated that human SI rates are highly variable -- both within instances of a single scale, and across different scales -- there have been few proposals that quantitatively e...
http://arxiv.org/pdf/2304.04758v1
2023-04-07T18:12:22
2023-04-07T18:12:22
{ "title": [ -0.000415802, 0.0010251999, -0.027862549, -0.052093506, -0.010063171, -0.022247314, -0.026626587, -0.0037555695, 0.005279541, -0.0012779236, -0.038879395, -0.03338623, -0.012802124, -0.035217285, -0.025421143, -0.036956787, -0.0014677048...
http://arxiv.org/abs/2304.03797v1
Bridging Nations: Quantifying the Role of Multilinguals in Communication on Social Media
Julia Mendelsohn, Sayan Ghosh, David Jurgens, Ceren Budak
cs.SI, cs.CL
Social media enables the rapid spread of many kinds of information, from memes to social movements. However, little is known about how information crosses linguistic boundaries. We apply causal inference techniques on the European Twitter network to quantify multilingual users' structural role and communication influen...
http://arxiv.org/pdf/2304.03797v1
2023-04-07T18:01:25
2023-04-07T18:01:25
{ "title": [ -0.0069084167, 0.00843811, -0.04321289, -0.03463745, 0.0132369995, -0.027709961, -0.018371582, -0.012550354, -0.049987793, 0.045318604, -0.0062332153, 0.032470703, -0.005760193, -0.015220642, -0.02003479, -0.008399963, -0.01474762, 0...
http://arxiv.org/abs/2304.03754v1
Language Models are Causal Knowledge Extractors for Zero-shot Video Question Answering
Hung-Ting Su, Yulei Niu, Xudong Lin, Winston H. Hsu, Shih-Fu Chang
cs.CL, cs.CV
Causal Video Question Answering (CVidQA) queries not only association or temporal relations but also causal relations in a video. Existing question synthesis methods pre-trained question generation (QG) systems on reading comprehension datasets with text descriptions as inputs. However, QG models only learn to ask asso...
http://arxiv.org/pdf/2304.03754v1
2023-04-07T17:45:49
2023-04-07T17:45:49
{ "title": [ -0.010925293, -0.009841919, -0.04626465, -0.0725708, -0.030090332, -0.014030457, -0.051513672, 0.025970459, 0.020385742, 0.02520752, -0.019760132, -0.0059318542, -0.0031833649, -0.019958496, -0.04437256, -0.002571106, 0.018157959, 0....
http://arxiv.org/abs/2304.03738v3
Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language Models
Emilio Ferrara
cs.CY, cs.CL
As the capabilities of generative language models continue to advance, the implications of biases ingrained within these models have garnered increasing attention from researchers, practitioners, and the broader public. This article investigates the challenges and risks associated with biases in large-scale language mo...
http://arxiv.org/pdf/2304.03738v3
2023-11-13T17:50:22
2023-04-07T17:14:00
{ "title": [ -0.022415161, -0.015457153, -0.02168274, -0.022705078, -0.010528564, -0.0003669262, -0.025817871, -0.013786316, -0.01876831, 0.024475098, -0.021362305, 0.018554688, 0.008361816, -0.01058197, -0.015274048, 0.00724411, -0.03479004, 0.0...
http://arxiv.org/abs/2304.03730v1
Gated Mechanism Enhanced Multi-Task Learning for Dialog Routing
Ziming Huang, Zhuoxuan Jiang, Ke Wang, Juntao Li, Shanshan Feng, Xian-Ling Mao
cs.CL
Currently, human-bot symbiosis dialog systems, e.g., pre- and after-sales in E-commerce, are ubiquitous, and the dialog routing component is essential to improve the overall efficiency, reduce human resource cost, and enhance user experience. Although most existing methods can fulfil this requirement, they can only mod...
http://arxiv.org/pdf/2304.03730v1
2023-04-07T16:51:46
2023-04-07T16:51:46
{ "title": [ -0.011672974, -0.0011339188, -0.0637207, -0.040008545, -0.0096206665, -0.03552246, -0.039855957, 0.018585205, -0.005542755, 0.052215576, 0.013748169, 0.017227173, -0.013809204, -0.02027893, -0.04046631, 0.0016546249, -0.020126343, 0....
http://arxiv.org/abs/2304.03728v1
Interpretable Unified Language Checking
Tianhua Zhang, Hongyin Luo, Yung-Sung Chuang, Wei Fang, Luc Gaitskell, Thomas Hartvigsen, Xixin Wu, Danny Fox, Helen Meng, James Glass
cs.CL
Despite recent concerns about undesirable behaviors generated by large language models (LLMs), including non-factual, biased, and hateful language, we find LLMs are inherent multi-task language checkers based on their latent representations of natural and social knowledge. We present an interpretable, unified, language...
http://arxiv.org/pdf/2304.03728v1
2023-04-07T16:47:49
2023-04-07T16:47:49
{ "title": [ -0.022567749, 0.0068969727, -0.02848816, -0.107299805, 0.009140015, 0.026885986, -0.09277344, -0.004486084, -0.017303467, 0.041290283, -0.007980347, -0.0017690659, -0.008483887, -0.02519226, -0.02154541, 0.0022621155, -0.020111084, 0...
http://arxiv.org/abs/2304.03717v1
On the Importance of Contrastive Loss in Multimodal Learning
Yunwei Ren, Yuanzhi Li
cs.LG, cs.CL, cs.CV
Recently, contrastive learning approaches (e.g., CLIP (Radford et al., 2021)) have received huge success in multimodal learning, where the model tries to minimize the distance between the representations of different views (e.g., image and its caption) of the same data point while keeping the representations of differe...
http://arxiv.org/pdf/2304.03717v1
2023-04-07T16:25:18
2023-04-07T16:25:18
{ "title": [ -0.0040779114, 0.00223732, -0.02357483, -0.03540039, -0.001250267, -0.018432617, -0.0027618408, -0.023452759, 0.0015583038, 0.03186035, -0.011512756, 0.007926941, -0.0143966675, -0.018447876, -0.035980225, 0.0016288757, -0.017242432, ...
http://arxiv.org/abs/2304.03682v3
BenCoref: A Multi-Domain Dataset of Nominal Phrases and Pronominal Reference Annotations
Shadman Rohan, Mojammel Hossain, Mohammad Mamun Or Rashid, Nabeel Mohammed
cs.CL
Coreference Resolution is a well studied problem in NLP. While widely studied for English and other resource-rich languages, research on coreference resolution in Bengali largely remains unexplored due to the absence of relevant datasets. Bengali, being a low-resource language, exhibits greater morphological richness c...
http://arxiv.org/pdf/2304.03682v3
2023-07-03T18:33:23
2023-04-07T15:08:46
{ "title": [ -0.02168274, 0.0041389465, -0.0446167, -0.076293945, 0.022537231, -0.008850098, -0.041381836, -0.002828598, 0.020507812, 0.038909912, 0.0132369995, 0.012069702, 0.037109375, -0.008361816, -0.047576904, 0.032989502, -0.030853271, 0.07...
http://arxiv.org/abs/2304.03639v1
Theoretical Conditions and Empirical Failure of Bracket Counting on Long Sequences with Linear Recurrent Networks
Nadine El-Naggar, Pranava Madhyastha, Tillman Weyde
cs.LG, cs.CL, cs.FL, cs.NE
Previous work has established that RNNs with an unbounded activation function have the capacity to count exactly. However, it has also been shown that RNNs are challenging to train effectively and generally do not learn exact counting behaviour. In this paper, we focus on this problem by studying the simplest possible ...
http://arxiv.org/pdf/2304.03639v1
2023-04-07T13:42:17
2023-04-07T13:42:17
{ "title": [ 0.009429932, -0.000058293343, -0.0284729, -0.06964111, 0.018463135, -0.027862549, -0.0077056885, 0.018188477, -0.03164673, 0.059509277, 0.012138367, 0.005897522, 0.0058898926, -0.024353027, -0.056549072, 0.0049057007, -0.016326904, 0...
http://arxiv.org/abs/2304.03612v1
What does ChatGPT return about human values? Exploring value bias in ChatGPT using a descriptive value theory
Ronald Fischer, Markus Luczak-Roesch, Johannes A Karl
cs.CL, cs.CY, cs.HC, 68T50
There has been concern about ideological basis and possible discrimination in text generated by Large Language Models (LLMs). We test possible value biases in ChatGPT using a psychological value theory. We designed a simple experiment in which we used a number of different probes derived from the Schwartz basic value t...
http://arxiv.org/pdf/2304.03612v1
2023-04-07T12:20:13
2023-04-07T12:20:13
{ "title": [ -0.0061149597, 0.010620117, -0.02822876, 0.033721924, 0.00038599968, -0.043182373, 0.040008545, -0.024993896, -0.027954102, 0.037078857, -0.01977539, 0.0096588135, -0.006099701, -0.01675415, -0.012878418, -0.00730896, -0.028564453, 0...
http://arxiv.org/abs/2304.03609v2
Revisiting Automated Prompting: Are We Actually Doing Better?
Yulin Zhou, Yiren Zhao, Ilia Shumailov, Robert Mullins, Yarin Gal
cs.CL, cs.LG
Current literature demonstrates that Large Language Models (LLMs) are great few-shot learners, and prompting significantly increases their performance on a range of downstream tasks in a few-shot learning setting. An attempt to automate human-led prompting followed, with some progress achieved. In particular, subsequen...
http://arxiv.org/pdf/2304.03609v2
2023-06-22T20:17:00
2023-04-07T12:06:44
{ "title": [ 0.008865356, 0.0025539398, -0.006000519, -0.036590576, 0.0022563934, -0.015930176, -0.011253357, -0.030654907, -0.019439697, 0.021255493, -0.023529053, 0.011672974, -0.0050849915, -0.02368164, -0.024795532, -0.00038552284, 0.015487671, ...
http://arxiv.org/abs/2304.03585v1
ArmanTTS single-speaker Persian dataset
Mohammd Hasan Shamgholi, Vahid Saeedi, Javad Peymanfard, Leila Alhabib, Hossein Zeinali
cs.CL
TTS, or text-to-speech, is a complicated process that can be accomplished through appropriate modeling using deep learning methods. In order to implement deep learning models, a suitable dataset is required. Since there is a scarce amount of work done in this field for the Persian language, this paper will introduce th...
http://arxiv.org/pdf/2304.03585v1
2023-04-07T10:52:55
2023-04-07T10:52:55
{ "title": [ 0.011894226, 0.0013055801, -0.019119263, -0.041168213, -0.033081055, -0.013153076, -0.057373047, -0.003578186, 0.0015058517, 0.037231445, -0.029830933, -0.022094727, 0.039611816, -0.049316406, 0.023086548, 0.013267517, 0.0035552979, ...
http://arxiv.org/abs/2304.11088v1
Profiling the news spreading barriers using news headlines
Abdul Sittar, Dunja Mladenic, Marko Grobelnik
cs.CL, cs.AI, cs.LG, cs.SI
News headlines can be a good data source for detecting the news spreading barriers in news media, which may be useful in many real-world applications. In this paper, we utilize semantic knowledge through the inference-based model COMET and sentiments of news headlines for barrier classification. We consider five barrie...
http://arxiv.org/pdf/2304.11088v1
2023-04-07T10:16:15
2023-04-07T10:16:15
{ "title": [ 0.02178955, 0.040618896, -0.018951416, -0.02734375, 0.0010919571, -0.017333984, -0.009277344, -0.035949707, -0.041107178, 0.041381836, -0.034698486, -0.0013551712, 0.007610321, -0.027816772, -0.017593384, -0.035125732, -0.002248764, ...
http://arxiv.org/abs/2304.03548v2
GEMINI: Controlling the Sentence-level Writing Style for Abstractive Text Summarization
Guangsheng Bao, Zebin Ou, Yue Zhang
cs.CL
Human experts write summaries using different techniques, including extracting a sentence from the document and rewriting it, or fusing various information from the document to abstract it. These techniques are flexible and thus difficult to be imitated by any single method. To address this issue, we propose an adaptiv...
http://arxiv.org/pdf/2304.03548v2
2023-10-08T07:53:11
2023-04-07T08:57:01
{ "title": [ -0.022415161, -0.0041885376, -0.033172607, -0.036468506, -0.025909424, 0.008453369, -0.047973633, -0.02017212, 0.017440796, 0.046813965, -0.015731812, 0.032287598, 0.009262085, 0.0026474, -0.05319214, 0.02229309, -0.010414124, 0.0317...
http://arxiv.org/abs/2304.03544v1
InfoCTM: A Mutual Information Maximization Perspective of Cross-Lingual Topic Modeling
Xiaobao Wu, Xinshuai Dong, Thong Nguyen, Chaoqun Liu, Liangming Pan, Anh Tuan Luu
cs.CL
Cross-lingual topic models have been prevalent for cross-lingual text analysis by revealing aligned latent topics. However, most existing methods suffer from producing repetitive topics that hinder further analysis and performance decline caused by low-coverage dictionaries. In this paper, we propose the Cross-lingual ...
http://arxiv.org/pdf/2304.03544v1
2023-04-07T08:49:43
2023-04-07T08:49:43
{ "title": [ -0.032592773, 0.0044822693, -0.030914307, -0.027572632, -0.011962891, 0.016204834, -0.06311035, -0.03942871, 0.019363403, 0.039764404, -0.007217407, 0.035125732, 0.0137786865, -0.010131836, -0.023727417, 0.029541016, -0.011917114, 0....
http://arxiv.org/abs/2304.03531v3
From Retrieval to Generation: Efficient and Effective Entity Set Expansion
Shulin Huang, Shirong Ma, Yangning Li, Yinghui Li, Yong Jiang, Hai-Tao Zheng, Ying Shen
cs.CL, cs.IR
Entity Set Expansion (ESE) is a critical task aiming at expanding entities of the target semantic class described by seed entities. Most existing ESE methods are retrieval-based frameworks that need to extract contextual features of entities and calculate the similarity between seed entities and candidate entities. To ...
http://arxiv.org/pdf/2304.03531v3
2023-10-31T15:41:00
2023-04-07T08:09:50
{ "title": [ -0.0044403076, -0.006198883, -0.056518555, -0.07598877, 0.04812622, 0.0011034012, -0.015701294, -0.0026016235, 0.014259338, 0.026000977, -0.015960693, 0.01889038, -0.0004620552, -0.006629944, -0.04019165, 0.016845703, -0.010276794, 0...
http://arxiv.org/abs/2304.03518v2
SSS at SemEval-2023 Task 10: Explainable Detection of Online Sexism using Majority Voted Fine-Tuned Transformers
Sriya Rallabandi, Sanchit Singhal, Pratinav Seth
cs.CL, cs.AI, cs.CY, cs.LG
This paper describes our submission to Task 10 at SemEval 2023-Explainable Detection of Online Sexism (EDOS), divided into three subtasks. The recent rise in social media platforms has seen an increase in disproportionate levels of sexism experienced by women on social media platforms. This has made detecting and expla...
http://arxiv.org/pdf/2304.03518v2
2023-04-23T19:09:48
2023-04-07T07:24:32
{ "title": [ 0.028778076, -0.008262634, -0.006629944, -0.025466919, 0.010353088, -0.019500732, 0.026107788, -0.0064353943, 0.008743286, 0.038482666, -0.025360107, 0.009223938, 0.020812988, -0.0079422, -0.03668213, -0.013076782, -0.024337769, 0.05...
http://arxiv.org/abs/2304.03512v2
Hierarchical Catalogue Generation for Literature Review: A Benchmark
Kun Zhu, Xiaocheng Feng, Xiachong Feng, Yingsheng Wu, Bing Qin
cs.CL
Multi-document scientific summarization can extract and organize important information from an abundant collection of papers, arousing widespread attention recently. However, existing efforts focus on producing lengthy overviews lacking a clear and logical hierarchy. To alleviate this problem, we present an atomic and ...
http://arxiv.org/pdf/2304.03512v2
2023-04-10T04:43:38
2023-04-07T07:13:35
{ "title": [ -0.036865234, 0.0010566711, -0.05831909, -0.078430176, -0.005470276, -0.00072956085, -0.07165527, -0.012512207, -0.016403198, 0.04171753, 0.0016345978, 0.029724121, -0.02519226, -0.0040359497, -0.066589355, -0.00056505203, -0.00843811, ...
http://arxiv.org/abs/2304.03464v2
Linking Representations with Multimodal Contrastive Learning
Abhishek Arora, Xinmei Yang, Shao-Yu Jheng, Melissa Dell
cs.CV, cs.CL, econ.GN, q-fin.EC
Many applications require grouping instances contained in diverse document datasets into classes. Most widely used methods do not employ deep learning and do not exploit the inherently multimodal nature of documents. Notably, record linkage is typically conceptualized as a string-matching problem. This study develops C...
http://arxiv.org/pdf/2304.03464v2
2023-04-11T01:34:48
2023-04-07T03:39:08
{ "title": [ 0.015113831, 0.011634827, -0.022750854, -0.045043945, 0.010940552, -0.022262573, 0.0051956177, -0.035003662, 0.013015747, 0.027618408, -0.013954163, 0.0031871796, -0.02520752, -0.02456665, -0.030044556, -0.004119873, -0.015579224, 0....
http://arxiv.org/abs/2304.03439v3
Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4
Hanmeng Liu, Ruoxi Ning, Zhiyang Teng, Jian Liu, Qiji Zhou, Yue Zhang
cs.CL, cs.AI
Harnessing logical reasoning ability is a comprehensive natural language understanding endeavor. With the release of Generative Pretrained Transformer 4 (GPT-4), highlighted as "advanced" at reasoning tasks, we are eager to learn the GPT-4 performance on various logical reasoning tasks. This report analyses multiple lo...
http://arxiv.org/pdf/2304.03439v3
2023-05-05T07:24:48
2023-04-07T01:37:45
{ "title": [ -0.015686035, 0.000041484833, -0.034851074, -0.019973755, 0.0061302185, -0.009010315, -0.010368347, -0.021560669, -0.022537231, 0.010803223, -0.016296387, 0.007217407, -0.019821167, -0.033111572, -0.015159607, 0.0023002625, -0.024887085,...
http://arxiv.org/abs/2304.03427v1
Cleansing Jewel: A Neural Spelling Correction Model Built On Google OCR-ed Tibetan Manuscripts
Queenie Luo, Yung-Sung Chuang
cs.CL, cs.AI, cs.CY, cs.LG
Scholars in the humanities rely heavily on ancient manuscripts to study history, religion, and socio-political structures in the past. Many efforts have been devoted to digitizing these precious manuscripts using OCR technology, but most manuscripts were blemished over the centuries so that an Optical Character Recogni...
http://arxiv.org/pdf/2304.03427v1
2023-04-07T00:45:12
2023-04-07T00:45:12
{ "title": [ -0.014190674, 0.019699097, -0.013587952, 0.015068054, -0.020141602, -0.013885498, 0.0011930466, -0.045318604, 0.0013999939, 0.06677246, -0.032104492, -0.012542725, 0.084228516, -0.015853882, -0.02217102, 0.009567261, -0.012001038, 0....
http://arxiv.org/abs/2304.03414v1
Towards Corpus-Scale Discovery of Selection Biases in News Coverage: Comparing What Sources Say About Entities as a Start
Sihao Chen, William Bruno, Dan Roth
cs.CL
News sources undergo the process of selecting newsworthy information when covering a certain topic. The process inevitably exhibits selection biases, i.e. news sources' typical patterns of choosing what information to include in news coverage, due to their agenda differences. To understand the magnitude and implication...
http://arxiv.org/pdf/2304.03414v1
2023-04-06T23:36:45
2023-04-06T23:36:45
{ "title": [ -0.019989014, 0.002040863, -0.052520752, -0.0073776245, 0.011505127, -0.024871826, 0.0056877136, -0.032287598, -0.015594482, 0.023757935, -0.061798096, 0.037139893, -0.013473511, 0.012397766, -0.037353516, -0.02885437, -0.0418396, 0....
http://arxiv.org/abs/2304.03401v2
Noise-Robust Dense Retrieval via Contrastive Alignment Post Training
Daniel Campos, ChengXiang Zhai, Alessandro Magnani
cs.IR, cs.AI, cs.CL
The success of contextual word representations and advances in neural information retrieval have made dense vector-based retrieval a standard approach for passage and document ranking. While effective and efficient, dual-encoders are brittle to variations in query distributions and noisy queries. Data augmentation can ...
http://arxiv.org/pdf/2304.03401v2
2023-04-10T14:22:32
2023-04-06T22:16:53
{ "title": [ 0.021438599, 0.031707764, -0.026000977, -0.037139893, 0.03137207, -0.017532349, 0.0025100708, -0.040405273, 0.05923462, 0.049468994, -0.03753662, 0.022247314, 0.018692017, -0.0038909912, -0.064208984, 0.031463623, 0.010414124, 0.0074...
http://arxiv.org/abs/2304.03399v1
Using LSTM and GRU With a New Dataset for Named Entity Recognition in the Arabic Language
Alaa Shaker, Alaa Aldarf, Igor Bessmertny
cs.CL
Named entity recognition (NER) is a natural language processing task (NLP), which aims to identify named entities and classify them like person, location, organization, etc. In the Arabic language, we can find a considerable size of unstructured data, and it needs to different preprocessing tool than languages like (En...
http://arxiv.org/pdf/2304.03399v1
2023-04-06T22:14:02
2023-04-06T22:14:02
{ "title": [ -0.012542725, 0.001791954, -0.040100098, -0.059051514, 0.015823364, -0.012336731, -0.031143188, -0.00068950653, -0.01626587, 0.027252197, -0.012825012, 0.0001655817, 0.003917694, -0.032836914, -0.022109985, 0.003271103, 0.0061187744, ...
http://arxiv.org/abs/2304.03394v2
Deep Learning for Opinion Mining and Topic Classification of Course Reviews
Anna Koufakou
cs.CL, cs.LG
Student opinions for a course are important to educators and administrators, regardless of the type of the course or the institution. Reading and manually analyzing open-ended feedback becomes infeasible for massive volumes of comments at institution level or online forums. In this paper, we collected and pre-processed...
http://arxiv.org/pdf/2304.03394v2
2023-06-16T14:15:10
2023-04-06T21:48:29
{ "title": [ -0.022460938, -0.0049095154, -0.033477783, -0.015945435, 0.0013790131, -0.01701355, 0.004081726, -0.023117065, 0.016281128, 0.012329102, -0.06982422, 0.007083893, -0.025177002, 0.01071167, -0.05065918, -0.027755737, -0.022140503, 0.0...
http://arxiv.org/abs/2304.03347v4
Towards Interpretable Mental Health Analysis with Large Language Models
Kailai Yang, Shaoxiong Ji, Tianlin Zhang, Qianqian Xie, Ziyan Kuang, Sophia Ananiadou
cs.CL
The latest large language models (LLMs) such as ChatGPT, exhibit strong capabilities in automated mental health analysis. However, existing relevant studies bear several limitations, including inadequate evaluations, lack of prompting strategies, and ignorance of exploring LLMs for explainability. To bridge these gaps,...
http://arxiv.org/pdf/2304.03347v4
2023-10-11T08:13:28
2023-04-06T19:53:59
{ "title": [ -0.0059928894, -0.0012378693, -0.03543091, -0.09136963, 0.02571106, 0.010101318, -0.070129395, -0.004676819, 0.004306793, 0.021484375, -0.0024604797, -0.013496399, 0.000644207, -0.028411865, -0.039489746, 0.0037441254, 0.002199173, 0...
http://arxiv.org/abs/2304.03325v1
ChatGPT-Crawler: Find out if ChatGPT really knows what it's talking about
Aman Rangapur, Haoran Wang
cs.CL, cs.IR, cs.LG, I.7, I.m
Large language models have gained considerable interest for their impressive performance on various tasks. Among these models, ChatGPT developed by OpenAI has become extremely popular among early adopters who even regard it as a disruptive technology in many fields like customer service, education, healthcare, and fina...
http://arxiv.org/pdf/2304.03325v1
2023-04-06T18:42:47
2023-04-06T18:42:47
{ "title": [ -0.03692627, -0.010131836, -0.054534912, -0.011260986, -0.013534546, -0.015434265, -0.025466919, -0.022247314, -0.016143799, 0.0048294067, -0.021026611, 0.018112183, -0.013702393, -0.029510498, -0.001660347, -0.011039734, -0.0043792725, ...
http://arxiv.org/abs/2304.03279v4
Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark
Alexander Pan, Jun Shern Chan, Andy Zou, Nathaniel Li, Steven Basart, Thomas Woodside, Jonathan Ng, Hanlin Zhang, Scott Emmons, Dan Hendrycks
cs.LG, cs.AI, cs.CL, cs.CY
Artificial agents have traditionally been trained to maximize reward, which may incentivize power-seeking and deception, analogous to how next-token prediction in language models (LMs) may incentivize toxicity. So do agents naturally learn to be Machiavellian? And how do we measure these behaviors in general-purpose mo...
http://arxiv.org/pdf/2304.03279v4
2023-06-13T01:01:42
2023-04-06T17:59:03
{ "title": [ -0.0047950745, 0.028915405, -0.009506226, 0.018554688, 0.026626587, -0.039001465, 0.024414062, -0.070495605, -0.07543945, 0.030349731, 0.017791748, 0.046936035, -0.036865234, -0.02848816, -0.014144897, -0.013496399, -0.024169922, 0.0...
http://arxiv.org/abs/2304.03277v1
Instruction Tuning with GPT-4
Baolin Peng, Chunyuan Li, Pengcheng He, Michel Galley, Jianfeng Gao
cs.CL, cs.AI
Prior work has shown that finetuning large language models (LLMs) using machine-generated instruction-following data enables such models to achieve remarkable zero-shot capabilities on new tasks, and no human-written instructions are needed. In this paper, we present the first attempt to use GPT-4 to generate instructi...
http://arxiv.org/pdf/2304.03277v1
2023-04-06T17:58:09
2023-04-06T17:58:09
{ "title": [ -0.026687622, 0.0032787323, -0.020050049, -0.01689148, -0.040618896, -0.0069770813, -0.028244019, -0.008415222, -0.0074768066, 0.04397583, -0.030090332, -0.008522034, -0.0028705597, -0.043273926, -0.009414673, 0.001531601, 0.013870239, ...
http://arxiv.org/abs/2304.03245v3
Large language models effectively leverage document-level context for literary translation, but critical errors persist
Marzena Karpinska, Mohit Iyyer
cs.CL
Large language models (LLMs) are competitive with the state of the art on a wide range of sentence-level translation datasets. However, their ability to translate paragraphs and documents remains unexplored because evaluation in these settings is costly and difficult. We show through a rigorous human evaluation that as...
http://arxiv.org/pdf/2304.03245v3
2023-05-22T22:00:51
2023-04-06T17:27:45
{ "title": [ 0.017745972, -0.019699097, -0.007980347, -0.03262329, -0.039001465, -0.021408081, -0.021621704, -0.01663208, 0.04425049, 0.07067871, -0.022964478, -0.009552002, 0.068115234, -0.026000977, -0.028549194, 0.009857178, -0.011947632, 0.00...
http://arxiv.org/abs/2304.03216v3
On the Pareto Front of Multilingual Neural Machine Translation
Liang Chen, Shuming Ma, Dongdong Zhang, Furu Wei, Baobao Chang
cs.CL, cs.AI
In this work, we study how the performance of a given direction changes with its sampling ratio in Multilingual Neural Machine Translation (MNMT). By training over 200 multilingual models with various model sizes, data sizes, and language directions, we find it interesting that the performance of certain translation di...
http://arxiv.org/pdf/2304.03216v3
2023-10-31T15:58:53
2023-04-06T16:49:19
{ "title": [ -0.024337769, -0.005897522, 0.014579773, -0.045837402, -0.015991211, 0.022903442, -0.046813965, -0.008346558, -0.013801575, 0.06689453, 0.0070114136, -0.0000192523, 0.03237915, -0.013496399, -0.057495117, 0.022964478, -0.019866943, 0...
http://arxiv.org/abs/2304.03208v1
Cerebras-GPT: Open Compute-Optimal Language Models Trained on the Cerebras Wafer-Scale Cluster
Nolan Dey, Gurpreet Gosal, Zhiming, Chen, Hemant Khachane, William Marshall, Ribhu Pathria, Marvin Tom, Joel Hestness
cs.LG, cs.CL
We study recent research advances that improve large language models through efficient pre-training and scaling, and open datasets and tools. We combine these advances to introduce Cerebras-GPT, a family of open compute-optimal language models scaled from 111M to 13B parameters. We train Cerebras-GPT models on the Eleu...
http://arxiv.org/pdf/2304.03208v1
2023-04-06T16:43:16
2023-04-06T16:43:16
{ "title": [ -0.04348755, -0.018554688, -0.03552246, -0.038208008, -0.033233643, 0.008636475, -0.08428955, -0.007297516, -0.011726379, 0.019729614, -0.020812988, 0.0017004013, 0.007911682, -0.03010559, -0.016998291, 0.007904053, -0.011352539, 0.0...
http://arxiv.org/abs/2304.03159v1
Bridging the Language Gap: Knowledge Injected Multilingual Question Answering
Zhichao Duan, Xiuxing Li, Zhengyan Zhang, Zhenyu Li, Ning Liu, Jianyong Wang
cs.CL
Question Answering (QA) is the task of automatically answering questions posed by humans in natural languages. There are different settings to answer a question, such as abstractive, extractive, boolean, and multiple-choice QA. As a popular topic in natural language processing tasks, extractive question answering task ...
http://arxiv.org/pdf/2304.03159v1
2023-04-06T15:41:25
2023-04-06T15:41:25
{ "title": [ -0.016540527, -0.024719238, -0.021347046, -0.06719971, -0.010864258, 0.0058288574, -0.043121338, 0.0028305054, 0.004032135, 0.026123047, -0.024917603, 0.005645752, -0.009475708, -0.0023288727, -0.051849365, -0.0048713684, 0.005252838, ...
http://arxiv.org/abs/2304.03153v1
Zero-Shot Next-Item Recommendation using Large Pretrained Language Models
Lei Wang, Ee-Peng Lim
cs.IR, cs.CL
Large language models (LLMs) have achieved impressive zero-shot performance in various natural language processing (NLP) tasks, demonstrating their capabilities for inference without training examples. Despite their success, no research has yet explored the potential of LLMs to perform next-item recommendations in the ...
http://arxiv.org/pdf/2304.03153v1
2023-04-06T15:35:11
2023-04-06T15:35:11
{ "title": [ 0.014259338, -0.013534546, -0.02243042, -0.05834961, -0.00466156, -0.0076828003, -0.033477783, -0.008758545, 0.0005311966, 0.04144287, -0.019592285, 0.024139404, -0.0062789917, 0.0011482239, -0.040863037, 0.0069999695, 0.005218506, 0...
http://arxiv.org/abs/2304.03145v1
Evaluating the Robustness of Machine Reading Comprehension Models to Low Resource Entity Renaming
Clemencia Siro, Tunde Oluwaseyi Ajayi
cs.CL
Question answering (QA) models have shown compelling results in the task of Machine Reading Comprehension (MRC). Recently these systems have proved to perform better than humans on held-out test sets of datasets e.g. SQuAD, but their robustness is not guaranteed. The QA model's brittleness is exposed when evaluated on ...
http://arxiv.org/pdf/2304.03145v1
2023-04-06T15:29:57
2023-04-06T15:29:57
{ "title": [ -0.045654297, 0.010986328, -0.025482178, -0.016571045, -0.01826477, 0.0016355515, -0.06707764, -0.039001465, 0.009880066, 0.052215576, -0.016067505, 0.0036945343, 0.034851074, -0.0030593872, -0.05618286, 0.021026611, -0.018997192, 0....
http://arxiv.org/abs/2304.03098v1
Static Fuzzy Bag-of-Words: a lightweight sentence embedding algorithm
Matteo Muffo, Roberto Tedesco, Licia Sbattella, Vincenzo Scotti
cs.CL, stat.ML
The introduction of embedding techniques has pushed forward significantly the Natural Language Processing field. Many of the proposed solutions have been presented for word-level encoding; anyhow, in the last years, new mechanism to treat information at an higher level of aggregation, like at sentence- and document-lev...
http://arxiv.org/pdf/2304.03098v1
2023-04-06T14:25:46
2023-04-06T14:25:46
{ "title": [ -0.013252258, -0.010238647, -0.0068473816, -0.03274536, -0.012710571, 0.008468628, -0.025482178, -0.031188965, 0.016525269, 0.030715942, -0.015823364, 0.013000488, -0.0021190643, -0.008049011, -0.04598999, 0.00409317, -0.014144897, 0...
http://arxiv.org/abs/2304.03087v1
Investigating Chain-of-thought with ChatGPT for Stance Detection on Social Media
Bowen Zhang, Xianghua Fu, Daijun Ding, Hu Huang, Yangyang Li, Liwen Jing
cs.CL
Stance detection predicts attitudes towards targets in texts and has gained attention with the rise of social media. Traditional approaches include conventional machine learning, early deep neural networks, and pre-trained fine-tuning models. However, with the evolution of very large pre-trained language models (VLPLMs...
http://arxiv.org/pdf/2304.03087v1
2023-04-06T14:12:02
2023-04-06T14:12:02
{ "title": [ 0.004180908, 0.0029296875, -0.005722046, -0.01600647, -0.006664276, -0.008293152, -0.0073890686, -0.027389526, 0.0012340546, 0.020126343, -0.053009033, 0.02835083, -0.027542114, -0.0039978027, -0.04006958, -0.019561768, -0.02029419, ...
http://arxiv.org/abs/2304.03047v2
ETPNav: Evolving Topological Planning for Vision-Language Navigation in Continuous Environments
Dong An, Hanqing Wang, Wenguan Wang, Zun Wang, Yan Huang, Keji He, Liang Wang
cs.CV, cs.CL, cs.RO
Vision-language navigation is a task that requires an agent to follow instructions to navigate in environments. It becomes increasingly crucial in the field of embodied AI, with potential applications in autonomous navigation, search and rescue, and human-robot interaction. In this paper, we propose to address a more p...
http://arxiv.org/pdf/2304.03047v2
2023-04-07T04:15:25
2023-04-06T13:07:17
{ "title": [ -0.026153564, -0.027786255, -0.014007568, -0.016571045, -0.016418457, -0.011459351, -0.010757446, 0.000119030476, 0.001531601, 0.020324707, -0.046936035, 0.0064582825, -0.03237915, 0.002281189, -0.041015625, -0.03262329, 0.028213501, ...
http://arxiv.org/abs/2304.03030v1
Compression of enumerations and gain
George Barmpalias, Xiaoyan Zhang, Bohua Zhan
cs.CL, cs.IT, math.IT, math.LO
We study the compressibility of enumerations, and its role in the relative Kolmogorov complexity of computably enumerable sets, with respect to density. With respect to a strong and a weak form of compression, we examine the gain: the amount of auxiliary information embedded in the compressed enumeration. Strong compre...
http://arxiv.org/pdf/2304.03030v1
2023-04-06T12:29:53
2023-04-06T12:29:53
{ "title": [ -0.0138549805, 0.010902405, -0.051757812, -0.064208984, -0.013450623, -0.006706238, -0.033172607, 0.017318726, 0.018569946, 0.026397705, -0.03286743, -0.018234253, -0.014007568, -0.031799316, -0.013130188, 0.013626099, 0.03894043, -0...
http://arxiv.org/abs/2304.02993v1
Natural Language Robot Programming: NLP integrated with autonomous robotic grasping
Muhammad Arshad Khan, Max Kenney, Jack Painter, Disha Kamale, Riza Batista-Navarro, Amir Ghalamzan-E
cs.RO, cs.CL, cs.HC
In this paper, we present a grammar-based natural language framework for robot programming, specifically for pick-and-place tasks. Our approach uses a custom dictionary of action words, designed to store together words that share meaning, allowing for easy expansion of the vocabulary by adding more action words from a ...
http://arxiv.org/pdf/2304.02993v1
2023-04-06T11:06:30
2023-04-06T11:06:30
{ "title": [ -0.034088135, -0.003358841, -0.040985107, -0.046020508, -0.0033855438, -0.0096206665, -0.028015137, 0.00021624565, -0.0063056946, 0.024246216, -0.05239868, 0.014122009, 0.0072517395, -0.012840271, -0.014663696, -0.01058197, -0.009147644,...
http://arxiv.org/abs/2304.02983v1
Leveraging Social Interactions to Detect Misinformation on Social Media
Tommaso Fornaciari, Luca Luceri, Emilio Ferrara, Dirk Hovy
cs.CL, cs.SI
Detecting misinformation threads is crucial to guarantee a healthy environment on social media. We address the problem using the data set created during the COVID-19 pandemic. It contains cascades of tweets discussing information weakly labeled as reliable or unreliable, based on a previous evaluation of the informatio...
http://arxiv.org/pdf/2304.02983v1
2023-04-06T10:30:04
2023-04-06T10:30:04
{ "title": [ 0.022201538, 0.017501831, -0.03439331, -0.047607422, 0.03074646, -0.022644043, -0.000041484833, -0.021743774, -0.042877197, 0.026565552, -0.027908325, 0.03036499, -0.01374054, -0.008125305, -0.017227173, -0.01826477, -0.019119263, 0....
http://arxiv.org/abs/2304.02945v1
Multi-label classification of open-ended questions with BERT
Matthias Schonlau, Julia Weiß, Jan Marquardt
stat.AP, cs.CL
Open-ended questions in surveys are valuable because they do not constrain the respondent's answer, thereby avoiding biases. However, answers to open-ended questions are text data which are harder to analyze. Traditionally, answers were manually classified as specified in the coding manual. Most of the effort to automa...
http://arxiv.org/pdf/2304.02945v1
2023-04-06T09:09:44
2023-04-06T09:09:44
{ "title": [ -0.0017738342, -0.000813961, -0.007911682, -0.04144287, -0.016586304, -0.0018043518, -0.028656006, -0.016921997, 0.00491333, 0.005458832, -0.010223389, 0.019882202, -0.034576416, -0.018005371, -0.02293396, -0.001408577, -0.02281189, ...
http://arxiv.org/abs/2304.02901v1
SpanRE: Entities and Overlapping Relations Extraction Based on Spans and Entity Attention
Hao Zhang
cs.CL
Extracting entities and relations is an essential task of information extraction. Triplets extracted from a sentence might overlap with each other. Previous methods either did not address the overlapping issues or solved overlapping issues partially. To tackle triplet overlapping problems completely, firstly we extract...
http://arxiv.org/pdf/2304.02901v1
2023-04-06T07:19:39
2023-04-06T07:19:39
{ "title": [ 0.0074691772, 0.002117157, -0.0635376, -0.08312988, 0.04006958, -0.012641907, -0.022140503, -0.002357483, 0.012023926, 0.021072388, -0.029434204, 0.019241333, 0.018569946, -0.004688263, -0.02218628, -0.0028419495, -0.02507019, 0.0613...
http://arxiv.org/abs/2304.02894v2
Affect as a proxy for literary mood
Emily Öhman, Riikka Rossi
cs.CL
We propose to use affect as a proxy for mood in literary texts. In this study, we explore the differences in computationally detecting tone versus detecting mood. Methodologically we utilize affective word embeddings to look at the affective distribution in different text segments. We also present a simple yet efficien...
http://arxiv.org/pdf/2304.02894v2
2023-05-23T08:09:23
2023-04-06T06:53:23
{ "title": [ 0.004272461, -0.0016155243, -0.0046539307, -0.03869629, -0.02885437, -0.020004272, -0.0041275024, -0.012451172, 0.019302368, 0.025543213, -0.020111084, -0.02507019, 0.02128601, -0.040252686, -0.04272461, 0.0034599304, 0.005493164, -0...
http://arxiv.org/abs/2304.02886v1
Automatic ICD-10 Code Association: A Challenging Task on French Clinical Texts
Yakini Tchouka, Jean-François Couchot, David Laiymani, Philippe Selles, Azzedine Rahmani
cs.CL, cs.AI
Automatically associating ICD codes with electronic health data is a well-known NLP task in medical research. NLP has evolved significantly in recent years with the emergence of pre-trained language models based on Transformers architecture, mainly in the English language. This paper adapts these models to automaticall...
http://arxiv.org/pdf/2304.02886v1
2023-04-06T06:31:54
2023-04-06T06:31:54
{ "title": [ -0.028030396, 0.01386261, -0.040924072, -0.050354004, 0.014175415, 0.008171082, -0.031829834, -0.0054893494, 0.00023806095, 0.03894043, -0.0032958984, 0.010696411, 0.008651733, -0.029510498, -0.016433716, 0.016525269, -0.015625, 0.02...
http://arxiv.org/abs/2304.05371v1
Those Aren't Your Memories, They're Somebody Else's: Seeding Misinformation in Chat Bot Memories
Conor Atkins, Benjamin Zi Hao Zhao, Hassan Jameel Asghar, Ian Wood, Mohamed Ali Kaafar
cs.CL, cs.AI, cs.CR, cs.LG
One of the new developments in chit-chat bots is a long-term memory mechanism that remembers information from past conversations for increasing engagement and consistency of responses. The bot is designed to extract knowledge of personal nature from their conversation partner, e.g., stating preference for a particular ...
http://arxiv.org/pdf/2304.05371v1
2023-04-06T05:09:39
2023-04-06T05:09:39
{ "title": [ 0.035308838, -0.012313843, -0.00050497055, -0.048187256, 0.0011987686, -0.011123657, 0.0026512146, -0.00497818, -0.0132369995, 0.04559326, -0.02331543, 0.021026611, 0.017654419, -0.009887695, -0.042266846, -0.018341064, -0.012062073, ...
http://arxiv.org/abs/2304.02868v1
Can Large Language Models Play Text Games Well? Current State-of-the-Art and Open Questions
Chen Feng Tsai, Xiaochen Zhou, Sierra S. Liu, Jing Li, Mo Yu, Hongyuan Mei
cs.CL, cs.AI, cs.LG
Large language models (LLMs) such as ChatGPT and GPT-4 have recently demonstrated their remarkable abilities of communicating with human users. In this technical report, we take an initiative to investigate their capacities of playing text games, in which a player has to understand the environment and respond to situat...
http://arxiv.org/pdf/2304.02868v1
2023-04-06T05:01:28
2023-04-06T05:01:28
{ "title": [ -0.028045654, -0.0050239563, -0.014030457, -0.032806396, -0.009048462, 0.0018911362, -0.039855957, -0.011398315, -0.008743286, 0.026412964, -0.025497437, 0.004989624, 0.005378723, -0.011756897, -0.030059814, -0.0034446716, -0.016326904, ...
http://arxiv.org/abs/2304.02822v1
Approach Intelligent Writing Assistants Usability with Seven Stages of Action
Avinash Bhat, Disha Shrivastava, Jin L. C. Guo
cs.HC, cs.CL, cs.LG
Despite the potential of Large Language Models (LLMs) as writing assistants, they are plagued by issues like coherence and fluency of the model output, trustworthiness, ownership of the generated content, and predictability of model performance, thereby limiting their usability. In this position paper, we propose to ad...
http://arxiv.org/pdf/2304.02822v1
2023-04-06T02:11:55
2023-04-06T02:11:55
{ "title": [ 0.013381958, 0.017440796, -0.041656494, -0.0032634735, 0.00699234, -0.02330017, 0.03503418, -0.009857178, -0.026504517, 0.04864502, -0.02255249, 0.008354187, 0.013633728, -0.02456665, -0.006713867, -0.020584106, 0.0076408386, 0.00946...
http://arxiv.org/abs/2304.02819v3
GPT detectors are biased against non-native English writers
Weixin Liang, Mert Yuksekgonul, Yining Mao, Eric Wu, James Zou
cs.CL, cs.AI, cs.HC, cs.LG
The rapid adoption of generative language models has brought about substantial advancements in digital communication, while simultaneously raising concerns regarding the potential misuse of AI-generated content. Although numerous detection methods have been proposed to differentiate between AI and human-generated conte...
http://arxiv.org/pdf/2304.02819v3
2023-07-10T18:48:45
2023-04-06T01:51:15
{ "title": [ 0.007232666, 0.0038757324, 0.0005393028, 0.05038452, -0.06524658, -0.021377563, 0.017074585, -0.03488159, -0.021194458, 0.057617188, -0.025405884, -0.008003235, 0.022201538, -0.023117065, 0.008796692, 0.0071487427, -0.024765015, 0.04...
http://arxiv.org/abs/2304.02812v1
Pragmatically Appropriate Diversity for Dialogue Evaluation
Katherine Stasaski, Marti A. Hearst
cs.CL
Linguistic pragmatics state that a conversation's underlying speech acts can constrain the type of response which is appropriate at each turn in the conversation. When generating dialogue responses, neural dialogue agents struggle to produce diverse responses. Currently, dialogue diversity is assessed using automatic m...
http://arxiv.org/pdf/2304.02812v1
2023-04-06T01:24:18
2023-04-06T01:24:18
{ "title": [ -0.001080513, 0.012817383, -0.060760498, -0.06939697, 0.032226562, -0.025299072, -0.032836914, 0.00087213516, 0.013931274, 0.012306213, 0.00015234947, -0.008598328, -0.055419922, -0.031677246, -0.06738281, -0.013008118, -0.029541016, ...
http://arxiv.org/abs/2304.02787v2
Context-Aware Classification of Legal Document Pages
Pavlos Fragkogiannis, Martina Forster, Grace E. Lee, Dell Zhang
cs.CL, I.2
For many business applications that require the processing, indexing, and retrieval of professional documents such as legal briefs (in PDF format etc.), it is often essential to classify the pages of any given document into their corresponding types beforehand. Most existing studies in the field of document image class...
http://arxiv.org/pdf/2304.02787v2
2023-04-25T14:59:24
2023-04-05T23:14:58
{ "title": [ -0.021331787, 0.008628845, -0.058135986, -0.04574585, 0.027114868, -0.012931824, -0.013656616, 0.00566864, 0.024658203, 0.013771057, -0.047576904, 0.01210022, 0.002773285, -0.005886078, -0.02861023, -0.0028362274, -0.020004272, 0.064...
http://arxiv.org/abs/2304.02785v1
Performance of Data Augmentation Methods for Brazilian Portuguese Text Classification
Marcellus Amadeus, Paulo Branco
cs.CL
Improving machine learning performance while increasing model generalization has been a constantly pursued goal by AI researchers. Data augmentation techniques are often used towards achieving this target, and most of its evaluation is made using English corpora. In this work, we took advantage of different existing da...
http://arxiv.org/pdf/2304.02785v1
2023-04-05T23:13:37
2023-04-05T23:13:37
{ "title": [ -0.026245117, 0.012901306, -0.05227661, -0.021759033, 0.003967285, -0.00047898293, -0.024261475, -0.0076828003, 0.0072669983, 0.040008545, -0.04989624, 0.008872986, 0.017791748, -0.0031089783, -0.048828125, -0.012367249, -0.023910522, ...
http://arxiv.org/abs/2304.02768v1
Application of Transformers based methods in Electronic Medical Records: A Systematic Literature Review
Vitor Alcantara Batista, Alexandre Gonçalves Evsukoff
cs.CL, cs.AI, cs.LG
The combined growth of available data and their unstructured nature has received increased interest in natural language processing (NLP) techniques to make value of these data assets since this format is not suitable for statistical analysis. This work presents a systematic literature review of state-of-the-art advance...
http://arxiv.org/pdf/2304.02768v1
2023-04-05T22:19:42
2023-04-05T22:19:42
{ "title": [ 0.018203735, 0.013389587, -0.009963989, -0.074523926, 0.021240234, 0.010169983, -0.047424316, -0.034851074, -0.024337769, 0.049224854, -0.03475952, 0.024261475, 0.044647217, -0.023208618, -0.009025574, 0.011703491, 0.0069122314, 0.04...
http://arxiv.org/abs/2304.02757v1
The Saudi Privacy Policy Dataset
Hend Al-Khalifa, Malak Mashaabi, Ghadi Al-Yahya, Raghad Alnashwan
cs.CL, cs.AI
This paper introduces the Saudi Privacy Policy Dataset, a diverse compilation of Arabic privacy policies from various sectors in Saudi Arabia, annotated according to the 10 principles of the Personal Data Protection Law (PDPL); the PDPL was established to be compatible with General Data Protection Regulation (GDPR); on...
http://arxiv.org/pdf/2304.02757v1
2023-04-05T21:40:37
2023-04-05T21:40:37
{ "title": [ 0.042266846, -0.0011711121, -0.039276123, -0.070617676, -0.009559631, -0.029403687, -0.033721924, 0.018249512, -0.08520508, 0.020568848, -0.031219482, 0.025863647, -0.028625488, -0.047943115, 0.052947998, -0.037384033, 0.029083252, 0...
http://arxiv.org/abs/2304.02754v2
Conceptual structure coheres in human cognition but not in large language models
Siddharth Suresh, Kushin Mukherjee, Xizheng Yu, Wei-Chun Huang, Lisa Padua, Timothy T Rogers
cs.AI, cs.CL, cs.LG
Neural network models of language have long been used as a tool for developing hypotheses about conceptual representation in the mind and brain. For many years, such use involved extracting vector-space representations of words and using distances among these to predict or understand human behavior in various semantic ...
http://arxiv.org/pdf/2304.02754v2
2023-11-10T17:42:31
2023-04-05T21:27:01
{ "title": [ -0.03189087, -0.011741638, -0.034484863, -0.047821045, -0.017059326, -0.016113281, -0.043151855, -0.012420654, -0.0021781921, 0.030548096, -0.013198853, -0.018096924, 0.0036621094, -0.02720642, -0.039855957, 0.0013208389, -0.010154724, ...
http://arxiv.org/abs/2304.02739v1
Bengali Fake Review Detection using Semi-supervised Generative Adversarial Networks
Md. Tanvir Rouf Shawon, G. M. Shahariar, Faisal Muhammad Shah, Mohammad Shafiul Alam, Md. Shahriar Mahbub
cs.CL, cs.AI, cs.LG
This paper investigates the potential of semi-supervised Generative Adversarial Networks (GANs) to fine-tune pretrained language models in order to classify Bengali fake reviews from real reviews with a few annotated data. With the rise of social media and e-commerce, the ability to detect fake or deceptive reviews is ...
http://arxiv.org/pdf/2304.02739v1
2023-04-05T20:40:09
2023-04-05T20:40:09
{ "title": [ -0.0035686493, 0.00038409233, -0.001666069, -0.007461548, -0.0027446747, -0.0061073303, -0.0024089813, -0.026351929, -0.044647217, 0.03338623, -0.0051956177, 0.021316528, -0.010147095, -0.034851074, -0.019073486, -0.022277832, -0.0315856...
http://arxiv.org/abs/2304.02738v1
Core Challenges in Embodied Vision-Language Planning
Jonathan Francis, Nariaki Kitamura, Felix Labelle, Xiaopeng Lu, Ingrid Navarro, Jean Oh
cs.RO, cs.AI, cs.CL, cs.CV, cs.HC
Recent advances in the areas of Multimodal Machine Learning and Artificial Intelligence (AI) have led to the development of challenging tasks at the intersection of Computer Vision, Natural Language Processing, and Robotics. Whereas many approaches and previous survey pursuits have characterised one or two of these dim...
http://arxiv.org/pdf/2304.02738v1
2023-04-05T20:37:13
2023-04-05T20:37:13
{ "title": [ -0.016357422, -0.016326904, 0.0018873215, -0.021728516, -0.031066895, -0.0072021484, -0.009422302, -0.010429382, -0.008384705, 0.030426025, -0.04345703, 0.01914978, -0.0069236755, -0.0053901672, -0.01965332, -0.0035896301, 0.024337769, ...
http://arxiv.org/abs/2304.02721v3
To Asymmetry and Beyond: Structured Pruning of Sequence to Sequence Models for Improved Inference Efficiency
Daniel Campos, ChengXiang Zhai
cs.CL, cs.AI
Sequence-to-sequence language models can be used to produce abstractive summaries which are coherent, relevant, and concise. Still, model sizes can make deployment in latency-sensitive or web-scale implementations difficult. This paper studies the relationship between model size, structured pruning, inference efficienc...
http://arxiv.org/pdf/2304.02721v3
2023-06-12T21:13:14
2023-04-05T19:44:20
{ "title": [ -0.023468018, -0.017745972, -0.03338623, -0.101135254, 0.0078086853, 0.017745972, -0.070251465, 0.026641846, 0.021774292, 0.013374329, -0.026382446, 0.009635925, 0.0063323975, -0.020233154, -0.033325195, 0.011009216, 0.010414124, 0.0...
http://arxiv.org/abs/2304.02625v1
Towards Explainable AI Writing Assistants for Non-native English Speakers
Yewon Kim, Mina Lee, Donghwi Kim, Sung-Ju Lee
cs.CL
We highlight the challenges faced by non-native speakers when using AI writing assistants to paraphrase text. Through an interview study with 15 non-native English speakers (NNESs) with varying levels of English proficiency, we observe that they face difficulties in assessing paraphrased texts generated by AI writing a...
http://arxiv.org/pdf/2304.02625v1
2023-04-05T17:51:36
2023-04-05T17:51:36
{ "title": [ -0.04107666, -0.017044067, -0.010101318, -0.013580322, -0.033843994, 0.005142212, -0.019561768, -0.0035495758, -0.010932922, 0.03302002, -0.03062439, 0.015098572, 0.0082092285, 0.009811401, -0.041931152, -0.005039215, -0.020904541, 0...
http://arxiv.org/abs/2304.02623v1
Beyond Summarization: Designing AI Support for Real-World Expository Writing Tasks
Zejiang Shen, Tal August, Pao Siangliulue, Kyle Lo, Jonathan Bragg, Jeff Hammerbacher, Doug Downey, Joseph Chee Chang, David Sontag
cs.CL, cs.HC
Large language models have introduced exciting new opportunities and challenges in designing and developing new AI-assisted writing support tools. Recent work has shown that leveraging this new technology can transform writing in many scenarios such as ideation during creative writing, editing support, and summarizatio...
http://arxiv.org/pdf/2304.02623v1
2023-04-05T17:47:11
2023-04-05T17:47:11
{ "title": [ -0.037017822, -0.013793945, -0.027572632, -0.039154053, -0.028366089, -0.002319336, -0.027908325, -0.013130188, -0.0016002655, 0.038482666, -0.03378296, 0.024139404, 0.008468628, 0.009124756, -0.039367676, 0.0012397766, -0.00869751, ...
http://arxiv.org/abs/2304.02554v1
Human-like Summarization Evaluation with ChatGPT
Mingqi Gao, Jie Ruan, Renliang Sun, Xunjian Yin, Shiping Yang, Xiaojun Wan
cs.CL
Evaluating text summarization is a challenging problem, and existing evaluation metrics are far from satisfactory. In this study, we explored ChatGPT's ability to perform human-like summarization evaluation using four human evaluation methods on five datasets. We found that ChatGPT was able to complete annotations rela...
http://arxiv.org/pdf/2304.02554v1
2023-04-05T16:17:32
2023-04-05T16:17:32
{ "title": [ -0.0134887695, 0.0079193115, -0.018753052, -0.03970337, -0.015296936, 0.0039367676, -0.03729248, -0.018554688, -0.01914978, 0.05331421, -0.016738892, 0.034698486, -0.0101623535, 0.014770508, -0.05505371, 0.019378662, -0.022262573, 0....
http://arxiv.org/abs/2304.02541v1
PWESuite: Phonetic Word Embeddings and Tasks They Facilitate
Vilém Zouhar, Kalvin Chang, Chenxuan Cui, Nathaniel Carlson, Nathaniel Robinson, Mrinmaya Sachan, David Mortensen
cs.CL
Word embeddings that map words into a fixed-dimensional vector space are the backbone of modern NLP. Most word embedding methods encode semantic information. However, phonetic information, which is important for some tasks, is often overlooked. In this work, we develop several novel methods which leverage articulatory ...
http://arxiv.org/pdf/2304.02541v1
2023-04-05T16:03:42
2023-04-05T16:03:42
{ "title": [ -0.009346008, -0.013183594, -0.038909912, -0.039794922, -0.011009216, -0.015304565, -0.027069092, 0.008956909, -0.013435364, 0.04135132, -0.0098724365, 0.01159668, 0.04147339, -0.029937744, -0.005126953, 0.028518677, 0.0032672882, 0....
http://arxiv.org/abs/2304.02496v1
Evaluation of ChatGPT Family of Models for Biomedical Reasoning and Classification
Shan Chen, Yingya Li, Sheng Lu, Hoang Van, Hugo JWL Aerts, Guergana K. Savova, Danielle S. Bitterman
cs.CL, cs.AI
Recent advances in large language models (LLMs) have shown impressive ability in biomedical question-answering, but have not been adequately investigated for more specific biomedical applications. This study investigates the performance of LLMs such as the ChatGPT family of models (GPT-3.5s, GPT-4) in biomedical tasks ...
http://arxiv.org/pdf/2304.02496v1
2023-04-05T15:11:25
2023-04-05T15:11:25
{ "title": [ -0.006706238, -0.0056877136, -0.023864746, -0.040802002, 0.0036258698, 0.011741638, -0.04244995, -0.021194458, 0.0030059814, 0.0131073, -0.0019683838, 0.012420654, -0.0018339157, -0.042144775, -0.022216797, 0.015281677, -0.022979736, ...
http://arxiv.org/abs/2304.02492v1
Quantifying the Roles of Visual, Linguistic, and Visual-Linguistic Complexity in Verb Acquisition
Yuchen Zhou, Michael J. Tarr, Daniel Yurovsky
cs.CL, cs.AI, cs.LG
Children typically learn the meanings of nouns earlier than the meanings of verbs. However, it is unclear whether this asymmetry is a result of complexity in the visual structure of categories in the world to which language refers, the structure of language itself, or the interplay between the two sources of informatio...
http://arxiv.org/pdf/2304.02492v1
2023-04-05T15:08:21
2023-04-05T15:08:21
{ "title": [ -0.007659912, -0.008033752, -0.029144287, -0.023986816, -0.018310547, -0.024368286, -0.013671875, -0.0029296875, -0.02355957, 0.023468018, -0.022033691, 0.031677246, -0.026611328, -0.015808105, -0.044036865, -0.019638062, -0.017745972, ...
http://arxiv.org/abs/2304.02426v5
ParroT: Translating during Chat using Large Language Models tuned with Human Translation and Feedback
Wenxiang Jiao, Jen-tse Huang, Wenxuan Wang, Zhiwei He, Tian Liang, Xing Wang, Shuming Shi, Zhaopeng Tu
cs.CL
Large language models (LLMs) like ChatGPT have exhibited remarkable abilities on a wide range of natural language processing~(NLP) tasks, including various machine translation abilities accomplished during chat. However, these models are only accessible through restricted APIs, which creates barriers to new research an...
http://arxiv.org/pdf/2304.02426v5
2023-11-02T07:44:52
2023-04-05T13:12:00
{ "title": [ -0.04611206, -0.020874023, -0.018341064, -0.03012085, -0.037841797, 0.009246826, -0.061950684, -0.0038700104, 0.0020008087, 0.0496521, -0.014419556, 0.009384155, 0.060455322, -0.017333984, -0.04272461, 0.015434265, 0.011505127, 0.025...