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/2304.00350v1
When Crowd Meets Persona: Creating a Large-Scale Open-Domain Persona Dialogue Corpus
Won Ik Cho, Yoon Kyung Lee, Seoyeon Bae, Jihwan Kim, Sangah Park, Moosung Kim, Sowon Hahn, Nam Soo Kim
cs.CL
Building a natural language dataset requires caution since word semantics is vulnerable to subtle text change or the definition of the annotated concept. Such a tendency can be seen in generative tasks like question-answering and dialogue generation and also in tasks that create a categorization-based corpus, like topi...
http://arxiv.org/pdf/2304.00350v1
2023-04-01T16:10:36
2023-04-01T16:10:36
{ "title": [ -0.0006070137, 0.017242432, -0.04815674, -0.06921387, 0.030960083, -0.0070228577, -0.038085938, -0.007160187, 0.0022125244, 0.02671814, -0.023406982, 0.008255005, 0.011413574, -0.007987976, -0.04168701, 0.0034713745, -0.019821167, 0....
http://arxiv.org/abs/2304.01938v1
Evaluating Large Language Models on a Highly-specialized Topic, Radiation Oncology Physics
Jason Holmes, Zhengliang Liu, Lian Zhang, Yuzhen Ding, Terence T. Sio, Lisa A. McGee, Jonathan B. Ashman, Xiang Li, Tianming Liu, Jiajian Shen, Wei Liu
physics.med-ph, cs.CL, physics.ed-ph
We present the first study to investigate Large Language Models (LLMs) in answering radiation oncology physics questions. Because popular exams like AP Physics, LSAT, and GRE have large test-taker populations and ample test preparation resources in circulation, they may not allow for accurately assessing the true poten...
http://arxiv.org/pdf/2304.01938v1
2023-04-01T06:04:58
2023-04-01T06:04:58
{ "title": [ -0.008712769, 0.0018253326, -0.03717041, -0.022964478, -0.02709961, -0.022277832, -0.035308838, -0.010612488, -0.041412354, 0.035461426, 0.018188477, -0.003709793, 0.019439697, -0.041900635, 0.019638062, -0.0021095276, -0.0039253235, ...
http://arxiv.org/abs/2304.00235v1
What Does the Indian Parliament Discuss? An Exploratory Analysis of the Question Hour in the Lok Sabha
Suman Adhya, Debarshi Kumar Sanyal
cs.CL
The TCPD-IPD dataset is a collection of questions and answers discussed in the Lower House of the Parliament of India during the Question Hour between 1999 and 2019. Although it is difficult to analyze such a huge collection manually, modern text analysis tools can provide a powerful means to navigate it. In this paper...
http://arxiv.org/pdf/2304.00235v1
2023-04-01T05:43:22
2023-04-01T05:43:22
{ "title": [ -0.012580872, 0.008041382, -0.07141113, -0.022628784, 0.020599365, -0.042175293, -0.0020198822, -0.014251709, -0.04550171, 0.019515991, -0.023376465, 0.04208374, -0.015258789, -0.01600647, -0.0038967133, -0.03161621, -0.014434814, 0....
http://arxiv.org/abs/2304.00228v1
Large language models can rate news outlet credibility
Kai-Cheng Yang, Filippo Menczer
cs.CL
Although large language models (LLMs) have shown exceptional performance in various natural language processing tasks, they are prone to hallucinations. State-of-the-art chatbots, such as the new Bing, attempt to mitigate this issue by gathering information directly from the internet to ground their answers. In this se...
http://arxiv.org/pdf/2304.00228v1
2023-04-01T05:04:06
2023-04-01T05:04:06
{ "title": [ -0.0043754578, -0.0066833496, -0.026535034, -0.0019893646, -0.05947876, -0.0287323, -0.00059890747, -0.0076942444, -0.028045654, 0.04550171, -0.032470703, -0.0231781, 0.016296387, -0.037078857, 0.0057907104, -0.011695862, -0.0049743652, ...
http://arxiv.org/abs/2304.00215v3
Inductive Relation Prediction from Relational Paths and Context with Hierarchical Transformers
Jiaang Li, Quan Wang, Zhendong Mao
cs.CL, cs.AI, cs.LG
Relation prediction on knowledge graphs (KGs) is a key research topic. Dominant embedding-based methods mainly focus on the transductive setting and lack the inductive ability to generalize to new entities for inference. Existing methods for inductive reasoning mostly mine the connections between entities, i.e., relati...
http://arxiv.org/pdf/2304.00215v3
2023-07-10T10:27:06
2023-04-01T03:49:47
{ "title": [ 0.0016584396, -0.0023078918, -0.039001465, -0.110839844, 0.0070152283, 0.018249512, -0.061584473, 0.00667572, -0.0059890747, 0.009849548, -0.041748047, 0.018585205, -0.0026454926, -0.014526367, -0.01586914, -0.013626099, -0.01309967, ...
http://arxiv.org/abs/2304.00180v1
FCC: Fusing Conversation History and Candidate Provenance for Contextual Response Ranking in Dialogue Systems
Zihao Wang, Eugene Agichtein, Jinho Choi
cs.CL, cs.AI
Response ranking in dialogues plays a crucial role in retrieval-based conversational systems. In a multi-turn dialogue, to capture the gist of a conversation, contextual information serves as essential knowledge to achieve this goal. In this paper, we present a flexible neural framework that can integrate contextual in...
http://arxiv.org/pdf/2304.00180v1
2023-03-31T23:58:28
2023-03-31T23:58:28
{ "title": [ 0.027801514, -0.0029716492, -0.021621704, -0.070373535, 0.015144348, -0.0072021484, -0.02809143, 0.005657196, 0.016296387, 0.037475586, -0.0140686035, 0.022949219, -0.0020275116, -0.017181396, -0.058746338, -0.0068130493, -0.019165039, ...
http://arxiv.org/abs/2304.00173v1
Lego-Features: Exporting modular encoder features for streaming and deliberation ASR
Rami Botros, Rohit Prabhavalkar, Johan Schalkwyk, Ciprian Chelba, Tara N. Sainath, Françoise Beaufays
cs.CL, cs.AI, cs.SD, eess.AS
In end-to-end (E2E) speech recognition models, a representational tight-coupling inevitably emerges between the encoder and the decoder. We build upon recent work that has begun to explore building encoders with modular encoded representations, such that encoders and decoders from different models can be stitched toget...
http://arxiv.org/pdf/2304.00173v1
2023-03-31T23:33:21
2023-03-31T23:33:21
{ "title": [ -0.013908386, -0.02255249, -0.058624268, -0.03262329, -0.029785156, -0.010505676, -0.040618896, 0.038848877, -0.010253906, 0.04159546, 0.027114868, -0.0060577393, 0.024093628, -0.0473938, -0.008239746, -0.0006814003, 0.013198853, 0.0...
http://arxiv.org/abs/2304.00171v1
Practical Conformer: Optimizing size, speed and flops of Conformer for on-Device and cloud ASR
Rami Botros, Anmol Gulati, Tara N. Sainath, Krzysztof Choromanski, Ruoming Pang, Trevor Strohman, Weiran Wang, Jiahui Yu
cs.CL, cs.SD, eess.AS
Conformer models maintain a large number of internal states, the vast majority of which are associated with self-attention layers. With limited memory bandwidth, reading these from memory at each inference step can slow down inference. In this paper, we design an optimized conformer that is small enough to meet on-devi...
http://arxiv.org/pdf/2304.00171v1
2023-03-31T23:30:48
2023-03-31T23:30:48
{ "title": [ -0.06933594, -0.023269653, -0.06982422, -0.04852295, -0.041992188, 0.014251709, -0.09509277, 0.033966064, -0.06689453, 0.054901123, 0.027572632, 0.0071258545, 0.030517578, -0.048339844, 0.006324768, 0.0027446747, 0.017318726, 0.05969...
http://arxiv.org/abs/2304.00121v1
Decoding the End-to-end Writing Trajectory in Scholarly Manuscripts
Ryan Koo, Anna Martin, Linghe Wang, Dongyeop Kang
cs.CL, cs.HC
Scholarly writing presents a complex space that generally follows a methodical procedure to plan and produce both rationally sound and creative compositions. Recent works involving large language models (LLM) demonstrate considerable success in text generation and revision tasks; however, LLMs still struggle to provide...
http://arxiv.org/pdf/2304.00121v1
2023-03-31T20:33:03
2023-03-31T20:33:03
{ "title": [ -0.034484863, 0.000092983246, -0.040985107, -0.05419922, -0.0076522827, 0.0019464493, -0.05847168, 0.0054244995, -0.003118515, 0.025314331, -0.009567261, 0.014030457, 0.021652222, -0.034118652, -0.015419006, 0.011482239, 0.016464233, ...
http://arxiv.org/abs/2304.00116v1
Enhancing Large Language Models with Climate Resources
Mathias Kraus, Julia Anna Bingler, Markus Leippold, Tobias Schimanski, Chiara Colesanti Senni, Dominik Stammbach, Saeid Ashraf Vaghefi, Nicolas Webersinke
cs.CL, cs.IR
Large language models (LLMs) have significantly transformed the landscape of artificial intelligence by demonstrating their ability in generating human-like text across diverse topics. However, despite their impressive capabilities, LLMs lack recent information and often employ imprecise language, which can be detrimen...
http://arxiv.org/pdf/2304.00116v1
2023-03-31T20:24:14
2023-03-31T20:24:14
{ "title": [ -0.026977539, -0.012260437, -0.047943115, -0.08520508, -0.0038223267, -0.0005912781, -0.06530762, 0.025177002, 0.008117676, 0.04220581, -0.01852417, -0.00096416473, 0.0051879883, -0.0035190582, -0.04763794, 0.025909424, -0.0030155182, ...
http://arxiv.org/abs/2304.00115v1
Extracting Thyroid Nodules Characteristics from Ultrasound Reports Using Transformer-based Natural Language Processing Methods
Aman Pathak, Zehao Yu, Daniel Paredes, Elio Paul Monsour, Andrea Ortiz Rocha, Juan P. Brito, Naykky Singh Ospina, Yonghui Wu
cs.CL
The ultrasound characteristics of thyroid nodules guide the evaluation of thyroid cancer in patients with thyroid nodules. However, the characteristics of thyroid nodules are often documented in clinical narratives such as ultrasound reports. Previous studies have examined natural language processing (NLP) methods in e...
http://arxiv.org/pdf/2304.00115v1
2023-03-31T20:23:58
2023-03-31T20:23:58
{ "title": [ 0.0149002075, -0.0047035217, -0.016326904, -0.05307007, -0.0104522705, 0.004436493, -0.02949524, -0.019927979, -0.0003566742, 0.040130615, -0.041870117, -0.011268616, 0.0102005005, -0.024627686, -0.030426025, -0.0058403015, 0.006263733, ...
http://arxiv.org/abs/2304.00114v1
Dense Sparse Retrieval: Using Sparse Language Models for Inference Efficient Dense Retrieval
Daniel Campos, ChengXiang Zhai
cs.IR, cs.AI, cs.CL
Vector-based retrieval systems have become a common staple for academic and industrial search applications because they provide a simple and scalable way of extending the search to leverage contextual representations for documents and queries. As these vector-based systems rely on contextual language models, their usag...
http://arxiv.org/pdf/2304.00114v1
2023-03-31T20:21:32
2023-03-31T20:21:32
{ "title": [ -0.03451538, -0.008926392, -0.04537964, -0.06402588, 0.008522034, 0.009353638, -0.05041504, -0.0005393028, 0.03894043, 0.043518066, -0.024398804, -0.012145996, 0.03265381, -0.016342163, -0.056488037, 0.02658081, 0.008262634, 0.004699...
http://arxiv.org/abs/2304.00111v1
Identifying Symptoms of Delirium from Clinical Narratives Using Natural Language Processing
Aokun Chen, Daniel Paredes, Zehao Yu, Xiwei Lou, Roberta Brunson, Jamie N. Thomas, Kimberly A. Martinez, Robert J. Lucero, Tanja Magoc, Laurence M. Solberg, Urszula A. Snigurska, Sarah E. Ser, Mattia Prosperi, Jiang Bian, Ragnhildur I. Bjarnadottir, Yonghui Wu
cs.CL
Delirium is an acute decline or fluctuation in attention, awareness, or other cognitive function that can lead to serious adverse outcomes. Despite the severe outcomes, delirium is frequently unrecognized and uncoded in patients' electronic health records (EHRs) due to its transient and diverse nature. Natural language...
http://arxiv.org/pdf/2304.00111v1
2023-03-31T20:16:44
2023-03-31T20:16:44
{ "title": [ -0.01272583, -0.0034637451, -0.01737976, -0.043304443, -0.005207062, 0.0018005371, -0.026504517, -0.030181885, -0.0021533966, 0.024124146, -0.037994385, -0.006801605, -0.001282692, -0.030273438, -0.036895752, -0.019760132, -0.021820068, ...
http://arxiv.org/abs/2304.12191v1
"Genlangs" and Zipf's Law: Do languages generated by ChatGPT statistically look human?
Justin Diamond
cs.CL
OpenAI's GPT-4 is a Large Language Model (LLM) that can generate coherent constructed languages, or "conlangs," which we propose be called "genlangs" when generated by Artificial Intelligence (AI). The genlangs created by ChatGPT for this research (Voxphera, Vivenzia, and Lumivoxa) each have unique features, appear fac...
http://arxiv.org/pdf/2304.12191v1
2023-03-31T20:10:59
2023-03-31T20:10:59
{ "title": [ -0.040985107, -0.012687683, -0.04989624, -0.008811951, -0.042388916, -0.011657715, -0.039154053, 0.0056877136, -0.027786255, 0.03048706, -0.013877869, 0.04208374, -0.003873825, -0.013114929, -0.0035743713, 0.0037937164, -0.031082153, ...
http://arxiv.org/abs/2303.18223v12
A Survey of Large Language Models
Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, Yifan Du, Chen Yang, Yushuo Chen, Zhipeng Chen, Jinhao Jiang, Ruiyang Ren, Yifan Li, Xinyu Tang, Zikang Liu, Peiyu Liu, Jian-Yun Nie, Ji-Rong Wen
cs.CL, cs.AI
Language is essentially a complex, intricate system of human expressions governed by grammatical rules. It poses a significant challenge to develop capable AI algorithms for comprehending and grasping a language. As a major approach, language modeling has been widely studied for language understanding and generation in...
http://arxiv.org/pdf/2303.18223v12
2023-09-11T15:13:59
2023-03-31T17:28:46
{ "title": [ -0.035186768, -0.01133728, -0.03930664, -0.08392334, 0.013206482, 0.016952515, -0.075805664, 0.024215698, 0.013694763, 0.03515625, -0.012779236, -0.01108551, 0.007965088, -0.011207581, -0.05026245, 0.018356323, -0.014427185, 0.039794...
http://arxiv.org/abs/2303.18190v1
Assessing Language Model Deployment with Risk Cards
Leon Derczynski, Hannah Rose Kirk, Vidhisha Balachandran, Sachin Kumar, Yulia Tsvetkov, M. R. Leiser, Saif Mohammad
cs.CL
This paper introduces RiskCards, a framework for structured assessment and documentation of risks associated with an application of language models. As with all language, text generated by language models can be harmful, or used to bring about harm. Automating language generation adds both an element of scale and also ...
http://arxiv.org/pdf/2303.18190v1
2023-03-31T16:45:42
2023-03-31T16:45:42
{ "title": [ -0.032043457, -0.0013227463, -0.035339355, -0.033294678, -0.032287598, 0.010826111, -0.046142578, -0.008529663, -0.0335083, 0.033233643, -0.026229858, 0.0011434555, -0.026245117, -0.026428223, -0.018005371, -0.029846191, -0.0065689087, ...
http://arxiv.org/abs/2304.00025v1
Demo Alleviate: Demonstrating Artificial Intelligence Enabled Virtual Assistance for Telehealth: The Mental Health Case
Kaushik Roy, Vedant Khandelwal, Raxit Goswami, Nathan Dolbir, Jinendra Malekar, Amit Sheth
cs.CL
After the pandemic, artificial intelligence (AI) powered support for mental health care has become increasingly important. The breadth and complexity of significant challenges required to provide adequate care involve: (a) Personalized patient understanding, (b) Safety-constrained and medically validated chatbot patien...
http://arxiv.org/pdf/2304.00025v1
2023-03-31T16:41:15
2023-03-31T16:41:15
{ "title": [ 0.050628662, -0.0009908676, -0.028045654, -0.07519531, 0.00894165, -0.02897644, -0.013626099, -0.0149383545, 0.016983032, 0.056854248, -0.03378296, -0.023361206, 0.09484863, -0.04135132, -0.019683838, -0.00021409988, 0.009414673, 0.0...
http://arxiv.org/abs/2304.01008v2
Self-Supervised Multimodal Learning: A Survey
Yongshuo Zong, Oisin Mac Aodha, Timothy Hospedales
cs.LG, cs.AI, cs.CL
Multimodal learning, which aims to understand and analyze information from multiple modalities, has achieved substantial progress in the supervised regime in recent years. However, the heavy dependence on data paired with expensive human annotations impedes scaling up models. Meanwhile, given the availability of large-...
http://arxiv.org/pdf/2304.01008v2
2023-08-04T21:17:28
2023-03-31T16:11:56
{ "title": [ -0.025924683, 0.029846191, -0.057739258, -0.046905518, 0.014976501, -0.014175415, -0.046783447, -0.012588501, -0.012519836, 0.027709961, -0.004470825, 0.011009216, -0.027862549, -0.028533936, -0.023635864, 0.0022144318, -0.0033683777, ...
http://arxiv.org/abs/2303.18162v1
A Multiple Choices Reading Comprehension Corpus for Vietnamese Language Education
Son T. Luu, Khoi Trong Hoang, Tuong Quang Pham, Kiet Van Nguyen, Ngan Luu-Thuy Nguyen
cs.CL
Machine reading comprehension has been an interesting and challenging task in recent years, with the purpose of extracting useful information from texts. To attain the computer ability to understand the reading text and answer relevant information, we introduce ViMMRC 2.0 - an extension of the previous ViMMRC for the t...
http://arxiv.org/pdf/2303.18162v1
2023-03-31T15:54:54
2023-03-31T15:54:54
{ "title": [ -0.025924683, -0.025909424, -0.00818634, -0.023544312, -0.034698486, 0.0065727234, -0.028244019, -0.0054397583, 0.0028076172, 0.018630981, -0.06124878, -0.0011644363, -0.015052795, -0.008895874, -0.044677734, -0.038116455, -0.029022217, ...
http://arxiv.org/abs/2304.01016v3
Quick Dense Retrievers Consume KALE: Post Training Kullback Leibler Alignment of Embeddings for Asymmetrical dual encoders
Daniel Campos, Alessandro Magnani, ChengXiang Zhai
cs.CL, cs.AI, cs.IR
In this paper, we consider the problem of improving the inference latency of language model-based dense retrieval systems by introducing structural compression and model size asymmetry between the context and query encoders. First, we investigate the impact of pre and post-training compression on the MSMARCO, Natural Q...
http://arxiv.org/pdf/2304.01016v3
2023-06-01T22:08:03
2023-03-31T15:44:13
{ "title": [ -0.008010864, -0.003534317, 0.02305603, 0.021499634, -0.043914795, -0.0010976791, 0.0030975342, -0.040771484, -0.00730896, 0.043273926, -0.027069092, -0.00063610077, 0.04348755, -0.03164673, 0.007774353, -0.0038547516, 0.019256592, 0...
http://arxiv.org/abs/2305.06147v1
CQSumDP: A ChatGPT-Annotated Resource for Query-Focused Abstractive Summarization Based on Debatepedia
Md Tahmid Rahman Laskar, Mizanur Rahman, Israt Jahan, Enamul Hoque, Jimmy Huang
cs.CL
Debatepedia is a publicly available dataset consisting of arguments and counter-arguments on controversial topics that has been widely used for the single-document query-focused abstractive summarization task in recent years. However, it has been recently found that this dataset is limited by noise and even most querie...
http://arxiv.org/pdf/2305.06147v1
2023-03-31T15:39:54
2023-03-31T15:39:54
{ "title": [ 0.044006348, -0.010986328, -0.0057792664, -0.08227539, -0.006603241, 0.0012340546, -0.04373169, -0.00020682812, -0.02279663, 0.052734375, -0.0029850006, 0.042388916, -0.011726379, -0.0034942627, -0.046539307, 0.003458023, -0.014213562, ...
http://arxiv.org/abs/2303.18149v2
Can AI Chatbots Pass the Fundamentals of Engineering (FE) and Principles and Practice of Engineering (PE) Structural Exams?
M. Z. Naser, Brandon Ross, Jennier Ogle, Venkatesh Kodur, Rami Hawileh, Jamal Abdalla, Huu-Tai Thai
cs.CL
The engineering community has recently witnessed the emergence of chatbot technology with the release of OpenAI ChatGPT-4 and Google Bard. While these chatbots have been reported to perform well and even pass various standardized tests, including medical and law exams, this forum paper explores whether these chatbots c...
http://arxiv.org/pdf/2303.18149v2
2023-04-03T02:50:06
2023-03-31T15:37:17
{ "title": [ -0.030410767, -0.03036499, 0.041381836, 0.05105591, -0.044281006, 0.013755798, 0.04119873, -0.03213501, -0.017807007, 0.05606079, -0.0234375, 0.0066871643, 0.05697632, -0.032348633, 0.00995636, -0.004802704, 0.0044517517, 0.018218994...
http://arxiv.org/abs/2303.18121v1
BERTino: an Italian DistilBERT model
Matteo Muffo, Enrico Bertino
cs.CL, stat.ML
The recent introduction of Transformers language representation models allowed great improvements in many natural language processing (NLP) tasks. However, if on one hand the performances achieved by this kind of architectures are surprising, on the other their usability is limited by the high number of parameters whic...
http://arxiv.org/pdf/2303.18121v1
2023-03-31T15:07:40
2023-03-31T15:07:40
{ "title": [ 0.016296387, -0.02571106, -0.021652222, 0.0020503998, -0.06414795, -0.008552551, -0.064697266, -0.032714844, -0.006881714, 0.03729248, 0.004497528, -0.041625977, 0.046173096, -0.052001953, -0.0151901245, -0.012542725, 0.000045239925, ...
http://arxiv.org/abs/2303.18120v2
UKP-SQuARE v3: A Platform for Multi-Agent QA Research
Haritz Puerto, Tim Baumgärtner, Rachneet Sachdeva, Haishuo Fang, Hao Zhang, Sewin Tariverdian, Kexin Wang, Iryna Gurevych
cs.CL
The continuous development of Question Answering (QA) datasets has drawn the research community's attention toward multi-domain models. A popular approach is to use multi-dataset models, which are models trained on multiple datasets to learn their regularities and prevent overfitting to a single dataset. However, with ...
http://arxiv.org/pdf/2303.18120v2
2023-05-17T13:08:47
2023-03-31T15:07:36
{ "title": [ -0.04321289, -0.015052795, -0.04824829, -0.02935791, 0.0015535355, -0.011558533, -0.015213013, -0.009300232, 0.012763977, 0.04611206, -0.01096344, 0.00058221817, 0.011962891, -0.013175964, -0.048217773, 0.010124207, -0.037902832, 0.0...
http://arxiv.org/abs/2303.18116v1
Pair Programming with Large Language Models for Sampling and Estimation of Copulas
Jan Górecki
cs.CL, stat.CO, 65C60, 68N19, 68T50
Without writing a single line of code by a human, an example Monte Carlo simulation based application for stochastic dependence modeling with copulas is developed using a state-of-the-art large language model (LLM) fine-tuned for conversations. This includes interaction with ChatGPT in natural language and using mathem...
http://arxiv.org/pdf/2303.18116v1
2023-03-31T15:02:48
2023-03-31T15:02:48
{ "title": [ -0.07208252, -0.00957489, -0.015068054, -0.034057617, -0.022232056, 0.030166626, -0.06121826, -0.023788452, 0.0066947937, 0.013885498, -0.022262573, -0.010536194, -0.013053894, -0.033691406, -0.023468018, -0.0119018555, -0.027954102, ...
http://arxiv.org/abs/2303.18110v1
The Edinburgh International Accents of English Corpus: Towards the Democratization of English ASR
Ramon Sanabria, Nikolay Bogoychev, Nina Markl, Andrea Carmantini, Ondrej Klejch, Peter Bell
cs.CL, cs.LG, cs.SD, eess.AS
English is the most widely spoken language in the world, used daily by millions of people as a first or second language in many different contexts. As a result, there are many varieties of English. Although the great many advances in English automatic speech recognition (ASR) over the past decades, results are usually ...
http://arxiv.org/pdf/2303.18110v1
2023-03-31T14:56:54
2023-03-31T14:56:54
{ "title": [ 0.0065574646, 0.010551453, -0.027069092, -0.04711914, -0.0010299683, -0.008010864, -0.015220642, -0.0074310303, -0.027160645, 0.04309082, 0.0027999878, 0.0015449524, 0.039642334, -0.039398193, -0.026550293, 0.018630981, -0.002714157, ...
http://arxiv.org/abs/2303.18103v1
Dataset and Baseline System for Multi-lingual Extraction and Normalization of Temporal and Numerical Expressions
Sanxing Chen, Yongqiang Chen, Börje F. Karlsson
cs.CL, cs.AI
Temporal and numerical expression understanding is of great importance in many downstream Natural Language Processing (NLP) and Information Retrieval (IR) tasks. However, much previous work covers only a few sub-types and focuses only on entity extraction, which severely limits the usability of identified mentions. In ...
http://arxiv.org/pdf/2303.18103v1
2023-03-31T14:49:05
2023-03-31T14:49:05
{ "title": [ 0.0043144226, 0.026687622, -0.08892822, -0.09967041, 0.040161133, -0.032287598, -0.049102783, 0.03414917, 0.006160736, 0.049102783, -0.023223877, -0.0031356812, 0.01928711, -0.012794495, -0.046936035, 0.0036125183, 0.008338928, 0.020...
http://arxiv.org/abs/2303.18049v1
No Place to Hide: Dual Deep Interaction Channel Network for Fake News Detection based on Data Augmentation
Biwei Cao, Lulu Hua, Jiuxin Cao, Jie Gui, Bo Liu, James Tin-Yau Kwok
cs.CL
Online Social Network (OSN) has become a hotbed of fake news due to the low cost of information dissemination. Although the existing methods have made many attempts in news content and propagation structure, the detection of fake news is still facing two challenges: one is how to mine the unique key features and evolut...
http://arxiv.org/pdf/2303.18049v1
2023-03-31T13:33:53
2023-03-31T13:33:53
{ "title": [ 0.022201538, 0.008918762, -0.022079468, -0.048065186, 0.0063323975, -0.0184021, -0.0011281967, -0.013137817, -0.036743164, 0.025909424, -0.022583008, 0.037872314, 0.0014734268, -0.01838684, -0.000072956085, -0.012054443, -0.051483154, ...
http://arxiv.org/abs/2303.18027v2
Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinations
Jungo Kasai, Yuhei Kasai, Keisuke Sakaguchi, Yutaro Yamada, Dragomir Radev
cs.CL
As large language models (LLMs) gain popularity among speakers of diverse languages, we believe that it is crucial to benchmark them to better understand model behaviors, failures, and limitations in languages beyond English. In this work, we evaluate LLM APIs (ChatGPT, GPT-3, and GPT-4) on the Japanese national medica...
http://arxiv.org/pdf/2303.18027v2
2023-04-05T07:53:37
2023-03-31T13:04:47
{ "title": [ 0.011543274, 0.004184723, -0.0025863647, -0.012672424, -0.0006375313, -0.009292603, -0.010696411, -0.038513184, -0.006011963, 0.017044067, -0.012680054, -0.008934021, -0.010070801, -0.037078857, -0.036834717, -0.013435364, -0.026748657, ...
http://arxiv.org/abs/2303.18011v1
Exploiting Multilingualism in Low-resource Neural Machine Translation via Adversarial Learning
Amit Kumar, Ajay Pratap, Anil Kumar Singh
cs.CL
Generative Adversarial Networks (GAN) offer a promising approach for Neural Machine Translation (NMT). However, feeding multiple morphologically languages into a single model during training reduces the NMT's performance. In GAN, similar to bilingual models, multilingual NMT only considers one reference translation for...
http://arxiv.org/pdf/2303.18011v1
2023-03-31T12:34:14
2023-03-31T12:34:14
{ "title": [ -0.026138306, -0.011253357, -0.014282227, -0.04510498, -0.0121536255, 0.01335144, -0.04147339, -0.0044784546, -0.024475098, 0.057739258, 0.01876831, 0.0052719116, 0.0040779114, -0.018203735, -0.04977417, 0.0058517456, -0.00491333, 0....
http://arxiv.org/abs/2304.00020v2
SemiMemes: A Semi-supervised Learning Approach for Multimodal Memes Analysis
Pham Thai Hoang Tung, Nguyen Tan Viet, Ngo Tien Anh, Phan Duy Hung
cs.LG, cs.AI, cs.CL, cs.CV
The prevalence of memes on social media has created the need to sentiment analyze their underlying meanings for censoring harmful content. Meme censoring systems by machine learning raise the need for a semi-supervised learning solution to take advantage of the large number of unlabeled memes available on the internet ...
http://arxiv.org/pdf/2304.00020v2
2023-05-16T07:19:23
2023-03-31T11:22:03
{ "title": [ 0.012550354, 0.011428833, -0.040740967, -0.028335571, 0.005584717, -0.02519226, -0.027862549, -0.019317627, 0.024917603, 0.043670654, 0.005332947, 0.021850586, -0.0066604614, -0.026382446, -0.04586792, 0.018432617, -0.0137786865, 0.0...
http://arxiv.org/abs/2303.17972v2
$\varepsilon$ KÚ <MASK>: Integrating Yorùbá cultural greetings into machine translation
Idris Akinade, Jesujoba Alabi, David Adelani, Clement Odoje, Dietrich Klakow
cs.CL
This paper investigates the performance of massively multilingual neural machine translation (NMT) systems in translating Yor\`ub\'a greetings ($\varepsilon$ k\'u [MASK]), which are a big part of Yor\`ub\'a language and culture, into English. To evaluate these models, we present IkiniYor\`ub\'a, a Yor\`ub\'a-English tr...
http://arxiv.org/pdf/2303.17972v2
2023-04-24T16:29:38
2023-03-31T11:16:20
{ "title": [ -0.03555298, -0.018417358, -0.005718231, -0.013694763, -0.029342651, 0.002614975, -0.028366089, -0.005519867, -0.0030021667, 0.061279297, -0.0060157776, 0.010292053, 0.0690918, -0.031219482, -0.039154053, 0.023117065, 0.0067100525, 0...
http://arxiv.org/abs/2303.17932v1
Trimming Phonetic Alignments Improves the Inference of Sound Correspondence Patterns from Multilingual Wordlists
Frederic Blum, Johann-Mattis List
cs.CL
Sound correspondence patterns form the basis of cognate detection and phonological reconstruction in historical language comparison. Methods for the automatic inference of correspondence patterns from phonetically aligned cognate sets have been proposed, but their application to multilingual wordlists requires extremel...
http://arxiv.org/pdf/2303.17932v1
2023-03-31T09:55:48
2023-03-31T09:55:48
{ "title": [ -0.018432617, 0.00233078, -0.050231934, -0.044433594, -0.01991272, -0.015716553, -0.07989502, 0.0054855347, -0.018417358, 0.060638428, -0.0018291473, 0.00009459257, 0.022903442, -0.031921387, -0.024261475, -0.0010852814, -0.0006289482, ...
http://arxiv.org/abs/2303.17930v1
JobHam-place with smart recommend job options and candidate filtering options
Shiyao Wu
cs.CL
Due to the increasing number of graduates, many applicants experience the situation about finding a job, and employers experience difficulty filtering job applicants, which might negatively impact their effectiveness. However, most job-hunting websites lack job recommendation and CV filtering or ranking functionality, ...
http://arxiv.org/pdf/2303.17930v1
2023-03-31T09:54:47
2023-03-31T09:54:47
{ "title": [ 0.07635498, 0.007751465, -0.017044067, -0.052734375, 0.035095215, -0.030471802, 0.046173096, 0.011131287, -0.013282776, 0.037750244, -0.003932953, 0.013290405, 0.041259766, -0.05050659, 0.008522034, -0.005645752, -0.0016317368, 0.013...
http://arxiv.org/abs/2303.17927v1
Cross-Cultural Transfer Learning for Chinese Offensive Language Detection
Li Zhou, Laura Cabello, Yong Cao, Daniel Hershcovich
cs.CL
Detecting offensive language is a challenging task. Generalizing across different cultures and languages becomes even more challenging: besides lexical, syntactic and semantic differences, pragmatic aspects such as cultural norms and sensitivities, which are particularly relevant in this context, vary greatly. In this ...
http://arxiv.org/pdf/2303.17927v1
2023-03-31T09:50:07
2023-03-31T09:50:07
{ "title": [ 0.0073509216, 0.02357483, -0.036956787, -0.017242432, 0.03149414, -0.036315918, 0.01411438, -0.028625488, 0.008354187, 0.022659302, -0.027664185, 0.005104065, 0.0037193298, -0.01486969, -0.040618896, -0.010765076, -0.017684937, 0.060...
http://arxiv.org/abs/2303.17910v2
Selective Knowledge Distillation for Non-Autoregressive Neural Machine Translation
Min Liu, Yu Bao, Chengqi Zhao, Shujian Huang
cs.CL
Benefiting from the sequence-level knowledge distillation, the Non-Autoregressive Transformer (NAT) achieves great success in neural machine translation tasks. However, existing knowledge distillation has side effects, such as propagating errors from the teacher to NAT students, which may limit further improvements of ...
http://arxiv.org/pdf/2303.17910v2
2023-08-04T16:19:24
2023-03-31T09:16:13
{ "title": [ -0.02468872, 0.001540184, -0.03466797, -0.076538086, -0.0009274483, 0.0047798157, -0.07281494, 0.011352539, 0.014228821, 0.04257202, -0.020339966, -0.0096588135, 0.019958496, -0.006668091, -0.06085205, -0.000029623508, 0.0048675537, ...
http://arxiv.org/abs/2303.17876v2
WebQAmGaze: A Multilingual Webcam Eye-Tracking-While-Reading Dataset
Tiago Ribeiro, Stephanie Brandl, Anders Søgaard, Nora Hollenstein
cs.CL
We create WebQAmGaze, a multilingual low-cost eye-tracking-while-reading dataset, designed to support the development of fair and transparent NLP models. WebQAmGaze includes webcam eye-tracking data from 332 participants naturally reading English, Spanish, and German texts. Each participant performs two reading tasks c...
http://arxiv.org/pdf/2303.17876v2
2023-04-14T06:22:47
2023-03-31T08:18:30
{ "title": [ -0.03543091, -0.00945282, -0.026473999, -0.008773804, -0.024887085, -0.018356323, -0.01159668, -0.033447266, 0.026000977, 0.04699707, -0.016937256, -0.0030117035, 0.033294678, -0.009170532, -0.047332764, 0.013511658, 0.005958557, 0.0...
http://arxiv.org/abs/2303.17853v2
Can AI Put Gamma-Ray Astrophysicists Out of a Job?
Samuel T. Spencer, Vikas Joshi, Alison M. W. Mitchell
physics.pop-ph, astro-ph.HE, cs.CL
In what will likely be a litany of generative-model-themed arXiv submissions celebrating April the 1st, we evaluate the capacity of state-of-the-art transformer models to create a paper detailing the detection of a Pulsar Wind Nebula with a non-existent Imaging Atmospheric Cherenkov Telescope (IACT) Array. We do this t...
http://arxiv.org/pdf/2303.17853v2
2023-04-04T09:30:55
2023-03-31T07:29:47
{ "title": [ 0.017105103, 0.0046691895, -0.018936157, 0.0395813, -0.022918701, -0.038116455, 0.05987549, -0.041625977, -0.002275467, 0.034973145, -0.008758545, -0.020935059, 0.008796692, -0.044158936, -0.019439697, -0.025604248, 0.018051147, 0.02...
http://arxiv.org/abs/2303.17839v1
Learning Procedure-aware Video Representation from Instructional Videos and Their Narrations
Yiwu Zhong, Licheng Yu, Yang Bai, Shangwen Li, Xueting Yan, Yin Li
cs.CV, cs.AI, cs.CL, cs.LG, cs.MM
The abundance of instructional videos and their narrations over the Internet offers an exciting avenue for understanding procedural activities. In this work, we propose to learn video representation that encodes both action steps and their temporal ordering, based on a large-scale dataset of web instructional videos an...
http://arxiv.org/pdf/2303.17839v1
2023-03-31T07:02:26
2023-03-31T07:02:26
{ "title": [ 0.0031738281, 0.012031555, -0.04220581, -0.05230713, 0.009780884, -0.01914978, -0.0236969, -0.010528564, -0.006538391, 0.030151367, -0.039093018, 0.004447937, -0.0050239563, -0.010810852, -0.03012085, -0.020996094, -0.013092041, 0.03...
http://arxiv.org/abs/2304.08592v1
Improving Scene Text Recognition for Character-Level Long-Tailed Distribution
Sunghyun Park, Sunghyo Chung, Jungsoo Lee, Jaegul Choo
cs.CV, cs.CL
Despite the recent remarkable improvements in scene text recognition (STR), the majority of the studies focused mainly on the English language, which only includes few number of characters. However, STR models show a large performance degradation on languages with a numerous number of characters (e.g., Chinese and Kore...
http://arxiv.org/pdf/2304.08592v1
2023-03-31T06:11:33
2023-03-31T06:11:33
{ "title": [ -0.00466156, 0.01386261, -0.015281677, -0.02255249, -0.0015220642, -0.0072631836, 0.0006995201, -0.016067505, 0.013031006, 0.030181885, -0.048309326, 0.0017871857, -0.03353882, 0.0051651, -0.06719971, -0.033599854, -0.0011281967, 0.0...
http://arxiv.org/abs/2303.17811v2
Zero-shot Referring Image Segmentation with Global-Local Context Features
Seonghoon Yu, Paul Hongsuck Seo, Jeany Son
cs.CV, cs.AI, cs.CL
Referring image segmentation (RIS) aims to find a segmentation mask given a referring expression grounded to a region of the input image. Collecting labelled datasets for this task, however, is notoriously costly and labor-intensive. To overcome this issue, we propose a simple yet effective zero-shot referring image se...
http://arxiv.org/pdf/2303.17811v2
2023-04-03T08:58:36
2023-03-31T06:00:50
{ "title": [ -0.0031108856, 0.012908936, -0.03451538, -0.03164673, 0.013084412, -0.006576538, -0.0065574646, -0.011070251, 0.0128479, 0.030227661, -0.010749817, 0.0066375732, -0.0055122375, -0.013542175, -0.017745972, 0.012863159, 0.020217896, 0....
http://arxiv.org/abs/2303.17807v1
Exploring the Potential of Large Language models in Traditional Korean Medicine: A Foundation Model Approach to Culturally-Adapted Healthcare
Dongyeop Jang, Chang-Eop Kim
cs.CL, cs.LG, J.3
Introduction: Traditional Korean medicine (TKM) emphasizes individualized diagnosis and treatment, making AI modeling difficult due to limited data and implicit processes. GPT-3.5 and GPT-4, large language models, have shown impressive medical knowledge despite lacking medicine-specific training. This study aimed to as...
http://arxiv.org/pdf/2303.17807v1
2023-03-31T05:43:21
2023-03-31T05:43:21
{ "title": [ -0.04510498, -0.0010585785, -0.03756714, -0.0569458, -0.0039405823, 0.011520386, -0.076049805, -0.011459351, -0.039001465, 0.027542114, 0.009284973, -0.0073165894, -0.0027599335, -0.046020508, -0.011100769, -0.0004811287, 0.011146545, ...
http://arxiv.org/abs/2303.17799v1
Dialog act guided contextual adapter for personalized speech recognition
Feng-Ju Chang, Thejaswi Muniyappa, Kanthashree Mysore Sathyendra, Kai Wei, Grant P. Strimel, Ross McGowan
cs.CL, cs.SD, eess.AS
Personalization in multi-turn dialogs has been a long standing challenge for end-to-end automatic speech recognition (E2E ASR) models. Recent work on contextual adapters has tackled rare word recognition using user catalogs. This adaptation, however, does not incorporate an important cue, the dialog act, which is avail...
http://arxiv.org/pdf/2303.17799v1
2023-03-31T05:13:44
2023-03-31T05:13:44
{ "title": [ 0.016494751, 0.007133484, -0.074035645, -0.082458496, 0.015457153, -0.020019531, -0.04144287, 0.028808594, -0.022827148, 0.0703125, -0.024490356, 0.005958557, 0.02255249, -0.012626648, -0.0418396, 0.002035141, -0.010925293, 0.0475463...
http://arxiv.org/abs/2303.17786v1
Attention is Not Always What You Need: Towards Efficient Classification of Domain-Specific Text
Yasmen Wahba, Nazim Madhavji, John Steinbacher
cs.CL, cs.LG
For large-scale IT corpora with hundreds of classes organized in a hierarchy, the task of accurate classification of classes at the higher level in the hierarchies is crucial to avoid errors propagating to the lower levels. In the business world, an efficient and explainable ML model is preferred over an expensive blac...
http://arxiv.org/pdf/2303.17786v1
2023-03-31T03:17:23
2023-03-31T03:17:23
{ "title": [ -0.00920105, 0.007797241, -0.043945312, -0.05206299, 0.028533936, -0.016220093, -0.019851685, -0.020889282, 0.004878998, 0.02267456, -0.012870789, 0.021438599, -0.005630493, -0.008972168, -0.047790527, 0.0011520386, -0.01725769, 0.05...
http://arxiv.org/abs/2303.17760v2
CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
Guohao Li, Hasan Abed Al Kader Hammoud, Hani Itani, Dmitrii Khizbullin, Bernard Ghanem
cs.AI, cs.CL, cs.CY, cs.LG, cs.MA
The rapid advancement of chat-based language models has led to remarkable progress in complex task-solving. However, their success heavily relies on human input to guide the conversation, which can be challenging and time-consuming. This paper explores the potential of building scalable techniques to facilitate autonom...
http://arxiv.org/pdf/2303.17760v2
2023-11-02T17:34:57
2023-03-31T01:09:00
{ "title": [ -0.02909851, -0.020126343, -0.024734497, -0.03604126, -0.017181396, 0.005493164, -0.02822876, -0.0030574799, 0.0019130707, 0.04675293, -0.022598267, -0.005443573, 0.030197144, -0.007987976, -0.052825928, 0.0017366409, -0.030197144, 0...
http://arxiv.org/abs/2303.17750v1
Design by Contract Framework for Quantum Software
Masaomi Yamaguchi, Nobukazu Yoshioka
cs.CL
To realize reliable quantum software, techniques to automatically ensure the quantum software's correctness have recently been investigated. However, they primarily focus on fixed quantum circuits rather than the procedure of building quantum circuits. Despite being a common approach, the correctness of building circui...
http://arxiv.org/pdf/2303.17750v1
2023-03-31T00:21:28
2023-03-31T00:21:28
{ "title": [ -0.03781128, -0.0032863617, -0.035003662, -0.030288696, -0.005748749, -0.017410278, -0.0032806396, -0.01436615, -0.020965576, 0.03491211, 0.01625061, -0.008125305, -0.004627228, -0.034729004, -0.02760315, 0.0049362183, -0.009033203, ...
http://arxiv.org/abs/2304.08176v1
Towards Mitigating ChatGPT's Negative Impact on Education: Optimizing Question Design through Bloom's Taxonomy
Saber Elsayed
cs.CY, cs.CL
The popularity of generative text AI tools in answering questions has led to concerns regarding their potential negative impact on students' academic performance and the challenges that educators face in evaluating student learning. To address these concerns, this paper introduces an evolutionary approach that aims to ...
http://arxiv.org/pdf/2304.08176v1
2023-03-31T00:01:59
2023-03-31T00:01:59
{ "title": [ -0.0027980804, -0.014450073, -0.008728027, 0.011276245, -0.01914978, -0.009140015, 0.032104492, -0.020111084, -0.025634766, 0.01826477, -0.031799316, 0.03540039, -0.015945435, -0.0015602112, -0.0054359436, -0.009727478, -0.007865906, ...
http://arxiv.org/abs/2303.17728v1
Evaluation of GPT and BERT-based models on identifying protein-protein interactions in biomedical text
Hasin Rehana, Nur Bengisu Çam, Mert Basmaci, Yongqun He, Arzucan Özgür, Junguk Hur
cs.CL, cs.AI
Detecting protein-protein interactions (PPIs) is crucial for understanding genetic mechanisms, disease pathogenesis, and drug design. However, with the fast-paced growth of biomedical literature, there is a growing need for automated and accurate extraction of PPIs to facilitate scientific knowledge discovery. Pre-trai...
http://arxiv.org/pdf/2303.17728v1
2023-03-30T22:06:10
2023-03-30T22:06:10
{ "title": [ -0.0015048981, 0.00031256676, -0.0231781, -0.053131104, -0.005218506, -0.0031166077, -0.044433594, -0.0032749176, 0.018814087, 0.017410278, 0.005874634, 0.0028476715, -0.00010764599, -0.02746582, -0.041229248, 0.0115356445, -0.018539429,...
http://arxiv.org/abs/2303.17710v2
What Types of Questions Require Conversation to Answer? A Case Study of AskReddit Questions
Shih-Hong Huang, Chieh-Yang Huang, Ya-Fang Lin, Ting-Hao 'Kenneth' Huang
cs.HC, cs.CL
The proliferation of automated conversational systems such as chatbots, spoken-dialogue systems, and smart speakers, has significantly impacted modern digital life. However, these systems are primarily designed to provide answers to well-defined questions rather than to support users in exploring complex, ill-defined q...
http://arxiv.org/pdf/2303.17710v2
2023-04-03T21:14:54
2023-03-30T21:05:22
{ "title": [ 0.015312195, 0.010314941, -0.039611816, -0.024002075, 0.008148193, -0.046722412, 0.0031337738, -0.015083313, -0.037506104, -0.006717682, 0.008590698, 0.0030975342, -0.036254883, -0.045898438, 0.0019111633, -0.022247314, 0.005332947, ...
http://arxiv.org/abs/2304.12208v1
Can ChatGPT be used to generate scientific hypotheses?
Yang Jeong Park, Daniel Kaplan, Zhichu Ren, Chia-Wei Hsu, Changhao Li, Haowei Xu, Sipei Li, Ju Li
cs.CL
We investigate whether large language models can perform the creative hypothesis generation that human researchers regularly do. While the error rate is high, generative AI seems to be able to effectively structure vast amounts of scientific knowledge and provide interesting and testable hypotheses. The future scientif...
http://arxiv.org/pdf/2304.12208v1
2023-03-30T20:40:52
2023-03-30T20:40:52
{ "title": [ -0.015991211, -0.009216309, -0.004512787, 0.033416748, -0.060028076, 0.000893116, -0.0011329651, -0.026565552, -0.020080566, 0.03881836, 0.034576416, -0.006587982, -0.00008690357, -0.036102295, -0.00027275085, 0.0067100525, 0.0008234978,...
http://arxiv.org/abs/2303.17695v1
Task Oriented Conversational Modelling With Subjective Knowledge
Raja Kumar
cs.CL, cs.IR, cs.LG
Existing conversational models are handled by a database(DB) and API based systems. However, very often users' questions require information that cannot be handled by such systems. Nonetheless, answers to these questions are available in the form of customer reviews and FAQs. DSTC-11 proposes a three stage pipeline con...
http://arxiv.org/pdf/2303.17695v1
2023-03-30T20:23:49
2023-03-30T20:23:49
{ "title": [ -0.018249512, -0.0033168793, -0.055755615, -0.07183838, -0.008293152, -0.0041656494, -0.07299805, 0.010688782, 0.033416748, 0.04586792, -0.0036182404, -0.04156494, 0.011192322, -0.023132324, -0.0826416, 0.002527237, -0.007270813, 0.0...
http://arxiv.org/abs/2303.17683v1
Fine-Tuning BERT with Character-Level Noise for Zero-Shot Transfer to Dialects and Closely-Related Languages
Aarohi Srivastava, David Chiang
cs.CL, cs.AI
In this work, we induce character-level noise in various forms when fine-tuning BERT to enable zero-shot cross-lingual transfer to unseen dialects and languages. We fine-tune BERT on three sentence-level classification tasks and evaluate our approach on an assortment of unseen dialects and languages. We find that chara...
http://arxiv.org/pdf/2303.17683v1
2023-03-30T19:51:18
2023-03-30T19:51:18
{ "title": [ 0.026107788, -0.029220581, -0.010551453, -0.046173096, -0.031921387, -0.008132935, -0.030227661, 0.025497437, 0.02281189, 0.05001831, -0.023773193, 0.021087646, 0.06347656, -0.024536133, 0.0015697479, 0.029022217, -0.0037631989, 0.01...
http://arxiv.org/abs/2303.17651v2
Self-Refine: Iterative Refinement with Self-Feedback
Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, Shashank Gupta, Bodhisattwa Prasad Majumder, Katherine Hermann, Sean Welleck, Amir Yazdanbakhsh, Peter Clark
cs.CL, cs.AI, cs.LG
Like humans, large language models (LLMs) do not always generate the best output on their first try. Motivated by how humans refine their written text, we introduce Self-Refine, an approach for improving initial outputs from LLMs through iterative feedback and refinement. The main idea is to generate an initial output ...
http://arxiv.org/pdf/2303.17651v2
2023-05-25T19:13:47
2023-03-30T18:30:01
{ "title": [ -0.054382324, 0.0046653748, -0.03942871, -0.08129883, -0.007881165, 0.032562256, -0.08947754, -0.01109314, 0.005580902, 0.032562256, -0.01210022, 0.01184082, -0.007507324, -0.029205322, -0.028396606, 0.012458801, 0.017852783, 0.01936...
http://arxiv.org/abs/2303.17650v3
Comparing Abstractive Summaries Generated by ChatGPT to Real Summaries Through Blinded Reviewers and Text Classification Algorithms
Mayank Soni, Vincent Wade
cs.CL
Large Language Models (LLMs) have gathered significant attention due to their impressive performance on a variety of tasks. ChatGPT, developed by OpenAI, is a recent addition to the family of language models and is being called a disruptive technology by a few, owing to its human-like text-generation capabilities. Alth...
http://arxiv.org/pdf/2303.17650v3
2023-08-28T09:34:59
2023-03-30T18:28:33
{ "title": [ 0.0093307495, -0.008895874, -0.0039253235, -0.028305054, -0.051483154, 0.0019302368, -0.04208374, -0.032287598, -0.014167786, 0.043060303, -0.03640747, 0.060943604, -0.022140503, 0.008270264, -0.032440186, 0.009262085, -0.03439331, 0...
http://arxiv.org/abs/2303.17649v3
Aligning a medium-size GPT model in English to a small closed domain in Spanish
Oscar R. Navarrete-Parra, Victor Uc-Cetina, Jorge Reyes-Magana
cs.CL, cs.LG
In this paper, we propose a methodology to align a medium-sized GPT model, originally trained in English for an open domain, to a small closed domain in Spanish. The application for which the model is finely tuned is the question answering task. To achieve this we also needed to train and implement another neural netwo...
http://arxiv.org/pdf/2303.17649v3
2023-05-30T21:48:49
2023-03-30T18:27:15
{ "title": [ -0.03466797, -0.018829346, -0.030090332, 0.00819397, -0.06173706, -0.020019531, -0.056121826, -0.017166138, -0.020126343, 0.03829956, 0.004638672, -0.01007843, 0.028335571, -0.0574646, 0.0033893585, 0.0005636215, -0.00894165, 0.01567...
http://arxiv.org/abs/2303.17647v1
Detecting and Grounding Important Characters in Visual Stories
Danyang Liu, Frank Keller
cs.CL, cs.CV
Characters are essential to the plot of any story. Establishing the characters before writing a story can improve the clarity of the plot and the overall flow of the narrative. However, previous work on visual storytelling tends to focus on detecting objects in images and discovering relationships between them. In this...
http://arxiv.org/pdf/2303.17647v1
2023-03-30T18:24:06
2023-03-30T18:24:06
{ "title": [ -0.0024681091, -0.0055351257, -0.028152466, -0.042510986, -0.0061683655, -0.0076942444, -0.0044898987, -0.012878418, 0.0059013367, 0.030181885, -0.022644043, 0.004550934, 0.020126343, -0.015991211, -0.032409668, -0.0017690659, -0.0197753...
http://arxiv.org/abs/2304.00993v1
Unsupervised Word Segmentation Using Temporal Gradient Pseudo-Labels
Tzeviya Sylvia Fuchs, Yedid Hoshen
eess.AS, cs.CL, cs.LG, cs.SD
Unsupervised word segmentation in audio utterances is challenging as, in speech, there is typically no gap between words. In a preliminary experiment, we show that recent deep self-supervised features are very effective for word segmentation but require supervision for training the classification head. To extend their ...
http://arxiv.org/pdf/2304.00993v1
2023-03-30T17:59:46
2023-03-30T17:59:46
{ "title": [ -0.030303955, 0.0030975342, -0.038085938, -0.045898438, -0.0008177757, -0.0006814003, -0.034210205, -0.010299683, -0.0131073, 0.04598999, -0.02406311, 0.0137786865, 0.0073509216, -0.001042366, -0.039733887, -0.0066566467, 0.0038051605, ...
http://arxiv.org/abs/2303.17590v2
Going Beyond Nouns With Vision & Language Models Using Synthetic Data
Paola Cascante-Bonilla, Khaled Shehada, James Seale Smith, Sivan Doveh, Donghyun Kim, Rameswar Panda, Gül Varol, Aude Oliva, Vicente Ordonez, Rogerio Feris, Leonid Karlinsky
cs.CV, cs.CL
Large-scale pre-trained Vision & Language (VL) models have shown remarkable performance in many applications, enabling replacing a fixed set of supported classes with zero-shot open vocabulary reasoning over (almost arbitrary) natural language prompts. However, recent works have uncovered a fundamental weakness of thes...
http://arxiv.org/pdf/2303.17590v2
2023-08-30T17:46:17
2023-03-30T17:57:43
{ "title": [ -0.031051636, -0.005882263, -0.045776367, -0.05154419, -0.009483337, -0.0134887695, -0.042999268, 0.001619339, 0.0052108765, 0.018997192, -0.050994873, -0.017074585, 0.015716553, -0.012908936, -0.028793335, -0.014732361, -0.011833191, ...
http://arxiv.org/abs/2303.17580v3
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, Yueting Zhuang
cs.CL, cs.AI, cs.CV, cs.LG
Solving complicated AI tasks with different domains and modalities is a key step toward artificial general intelligence. While there are abundant AI models available for different domains and modalities, they cannot handle complicated AI tasks. Considering large language models (LLMs) have exhibited exceptional ability...
http://arxiv.org/pdf/2303.17580v3
2023-05-25T15:50:20
2023-03-30T17:48:28
{ "title": [ -0.033569336, -0.026794434, -0.052581787, -0.007949829, -0.02217102, -0.020004272, -0.0020542145, -0.008270264, -0.008605957, 0.0256958, -0.03375244, 0.028503418, 0.0019607544, -0.014205933, -0.008811951, -0.010269165, -0.0016088486, ...
http://arxiv.org/abs/2303.17574v1
Elastic Weight Removal for Faithful and Abstractive Dialogue Generation
Nico Daheim, Nouha Dziri, Mrinmaya Sachan, Iryna Gurevych, Edoardo M. Ponti
cs.CL, cs.AI, cs.LG
Ideally, dialogue systems should generate responses that are faithful to the knowledge contained in relevant documents. However, many models generate hallucinated responses instead that contradict it or contain unverifiable information. To mitigate such undesirable behaviour, it has been proposed to fine-tune a `negati...
http://arxiv.org/pdf/2303.17574v1
2023-03-30T17:40:30
2023-03-30T17:40:30
{ "title": [ -0.024368286, -0.017974854, -0.053588867, -0.070373535, -0.010482788, 0.004295349, -0.0625, 0.02041626, -0.0013370514, 0.039794922, 0.033447266, -0.016296387, -0.017166138, -0.03829956, -0.05307007, 0.004058838, -0.01285553, 0.016967...
http://arxiv.org/abs/2303.17564v2
BloombergGPT: A Large Language Model for Finance
Shijie Wu, Ozan Irsoy, Steven Lu, Vadim Dabravolski, Mark Dredze, Sebastian Gehrmann, Prabhanjan Kambadur, David Rosenberg, Gideon Mann
cs.LG, cs.AI, cs.CL, q-fin.GN
The use of NLP in the realm of financial technology is broad and complex, with applications ranging from sentiment analysis and named entity recognition to question answering. Large Language Models (LLMs) have been shown to be effective on a variety of tasks; however, no LLM specialized for the financial domain has bee...
http://arxiv.org/pdf/2303.17564v2
2023-05-09T16:06:35
2023-03-30T17:30:36
{ "title": [ -0.021392822, -0.024139404, -0.051635742, -0.05657959, -0.000056505203, -0.0018987656, -0.0435791, 0.0060768127, 0.004219055, 0.03805542, -0.016036987, -0.0058174133, 0.017425537, -0.0121154785, -0.0524292, 0.004722595, -0.01713562, ...
http://arxiv.org/abs/2303.17557v1
Recognition, recall, and retention of few-shot memories in large language models
A. Emin Orhan
cs.CL, cs.LG, cs.NE
The training of modern large language models (LLMs) takes place in a regime where most training examples are seen only a few times by the model during the course of training. What does a model remember about such examples seen only a few times during training and how long does that memory persist in the face of continu...
http://arxiv.org/pdf/2303.17557v1
2023-03-30T17:26:16
2023-03-30T17:26:16
{ "title": [ -0.013313293, -0.012458801, -0.034423828, -0.051483154, -0.007911682, -0.015472412, -0.030944824, 0.013710022, 0.0055618286, 0.0340271, -0.020980835, 0.00094366074, -0.0024852753, -0.0119018555, -0.05911255, 0.0013923645, -0.0032253265, ...
http://arxiv.org/abs/2303.17548v1
Whose Opinions Do Language Models Reflect?
Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee, Percy Liang, Tatsunori Hashimoto
cs.CL, cs.AI, cs.CY, cs.LG
Language models (LMs) are increasingly being used in open-ended contexts, where the opinions reflected by LMs in response to subjective queries can have a profound impact, both on user satisfaction, as well as shaping the views of society at large. In this work, we put forth a quantitative framework to investigate the ...
http://arxiv.org/pdf/2303.17548v1
2023-03-30T17:17:08
2023-03-30T17:17:08
{ "title": [ -0.031707764, -0.021972656, -0.00919342, 0.0016880035, -0.06878662, 0.010032654, -0.04815674, -0.012199402, 0.016555786, 0.015197754, -0.0016098022, -0.04095459, -0.023330688, -0.04888916, -0.017089844, -0.023834229, 0.0011777878, 0....
http://arxiv.org/abs/2303.17517v1
Hindi as a Second Language: Improving Visually Grounded Speech with Semantically Similar Samples
Hyeonggon Ryu, Arda Senocak, In So Kweon, Joon Son Chung
cs.CL, cs.CV, cs.SD, eess.AS
The objective of this work is to explore the learning of visually grounded speech models (VGS) from multilingual perspective. Bilingual VGS models are generally trained with an equal number of spoken captions from both languages. However, in reality, there can be an imbalance among the languages for the available spoke...
http://arxiv.org/pdf/2303.17517v1
2023-03-30T16:34:10
2023-03-30T16:34:10
{ "title": [ -0.010375977, 0.0032730103, -0.03302002, -0.036987305, -0.008026123, -0.008117676, -0.050598145, -0.010421753, -0.015563965, 0.037963867, -0.024261475, 0.033447266, 0.0014953613, -0.023071289, -0.02796936, -0.010543823, -0.007091522, ...
http://arxiv.org/abs/2303.17491v2
Language Models can Solve Computer Tasks
Geunwoo Kim, Pierre Baldi, Stephen McAleer
cs.CL, cs.AI, cs.HC, cs.LG
Agents capable of carrying out general tasks on a computer can improve efficiency and productivity by automating repetitive tasks and assisting in complex problem-solving. Ideally, such agents should be able to solve new computer tasks presented to them through natural language commands. However, previous approaches to...
http://arxiv.org/pdf/2303.17491v2
2023-06-07T17:50:44
2023-03-30T16:01:52
{ "title": [ -0.0044021606, -0.03048706, 0.025375366, -0.026031494, -0.06933594, 0.0049819946, -0.071777344, 0.00033712387, 0.021713257, 0.052825928, -0.0020332336, -0.055389404, 0.03173828, -0.035125732, -0.020599365, 0.0018777847, 0.017791748, ...
http://arxiv.org/abs/2303.17475v2
Efficient distributed representations beyond negative sampling
Lorenzo Dall'Amico, Enrico Maria Belliardo
cs.LG, cs.CL, stat.ML
This article describes an efficient method to learn distributed representations, also known as embeddings. This is accomplished minimizing an objective function similar to the one introduced in the Word2Vec algorithm and later adopted in several works. The optimization computational bottleneck is the calculation of the...
http://arxiv.org/pdf/2303.17475v2
2023-10-30T09:34:17
2023-03-30T15:48:26
{ "title": [ 0.005153656, 0.018676758, -0.014350891, -0.06048584, 0.026321411, 0.0028839111, -0.019882202, -0.03677368, 0.042266846, 0.030014038, -0.03237915, 0.00843811, -0.009017944, -0.001953125, -0.041748047, 0.01675415, 0.020629883, 0.017150...
http://arxiv.org/abs/2303.17466v2
Assessing Cross-Cultural Alignment between ChatGPT and Human Societies: An Empirical Study
Yong Cao, Li Zhou, Seolhwa Lee, Laura Cabello, Min Chen, Daniel Hershcovich
cs.CL
The recent release of ChatGPT has garnered widespread recognition for its exceptional ability to generate human-like responses in dialogue. Given its usage by users from various nations and its training on a vast multilingual corpus that incorporates diverse cultural and societal norms, it is crucial to evaluate its ef...
http://arxiv.org/pdf/2303.17466v2
2023-03-31T15:02:48
2023-03-30T15:43:39
{ "title": [ -0.011459351, -0.004737854, -0.006580353, 0.008094788, -0.0020809174, -0.01247406, -0.004085541, -0.024642944, -0.01991272, 0.03640747, -0.019104004, 0.028213501, -0.00674057, -0.0027942657, -0.03656006, 0.009117126, -0.027816772, 0....
http://arxiv.org/abs/2304.01005v3
Federated Learning Based Multilingual Emoji Prediction In Clean and Attack Scenarios
Karim Gamal, Ahmed Gaber, Hossam Amer
cs.CL
Federated learning is a growing field in the machine learning community due to its decentralized and private design. Model training in federated learning is distributed over multiple clients giving access to lots of client data while maintaining privacy. Then, a server aggregates the training done on these multiple cli...
http://arxiv.org/pdf/2304.01005v3
2023-07-07T00:51:43
2023-03-30T15:31:49
{ "title": [ 0.024215698, 0.009307861, -0.02720642, -0.07208252, 0.015045166, -0.002122879, -0.031280518, 0.012870789, -0.024169922, 0.06530762, 0.008171082, 0.0107421875, 0.018859863, -0.012054443, -0.029403687, 0.00017774105, -0.002368927, 0.02...
http://arxiv.org/abs/2303.17408v2
A Prompt-based Multimodal Tabular Transformer Encoder For Medical Intervention Duration Estimation
Yucheng Ruan, Xiang Lan, Daniel J. Tan, Hairil Rizal Abdullah, Mengling Feng
cs.CL
Objective: This study focuses on estimating the duration of medical interventions using electronic health records (EHRs) in clinical decision support. Most existing models were designed for structured tabular data only and often suffer from data corruption problem. The unstructured clinical free-text data that provides...
http://arxiv.org/pdf/2303.17408v2
2023-08-09T08:58:25
2023-03-30T14:25:44
{ "title": [ -0.008453369, 0.018981934, -0.038238525, -0.04208374, -0.0119018555, -0.012580872, -0.05807495, -0.03414917, -0.02368164, 0.030548096, -0.004043579, 0.0014839172, 0.03250122, -0.044281006, -0.0010061264, 0.0031967163, 0.00033450127, ...
http://arxiv.org/abs/2303.17395v1
WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Research
Xinhao Mei, Chutong Meng, Haohe Liu, Qiuqiang Kong, Tom Ko, Chengqi Zhao, Mark D. Plumbley, Yuexian Zou, Wenwu Wang
eess.AS, cs.CL, cs.MM, cs.SD
The advancement of audio-language (AL) multimodal learning tasks has been significant in recent years. However, researchers face challenges due to the costly and time-consuming collection process of existing audio-language datasets, which are limited in size. To address this data scarcity issue, we introduce WavCaps, t...
http://arxiv.org/pdf/2303.17395v1
2023-03-30T14:07:47
2023-03-30T14:07:47
{ "title": [ -0.004535675, -0.008003235, -0.034698486, -0.037231445, -0.0051879883, -0.0138168335, -0.0501709, -0.0015668869, -0.016693115, 0.05230713, -0.0062065125, 0.02368164, 0.023651123, -0.016723633, -0.024398804, 0.019180298, -0.031677246, ...
http://arxiv.org/abs/2303.17367v1
A BERT-based Unsupervised Grammatical Error Correction Framework
Nankai Lin, Hongbin Zhang, Menglan Shen, Yu Wang, Shengyi Jiang, Aimin Yang
cs.CL
Grammatical error correction (GEC) is a challenging task of natural language processing techniques. While more attempts are being made in this approach for universal languages like English or Chinese, relatively little work has been done for low-resource languages for the lack of large annotated corpora. In low-resourc...
http://arxiv.org/pdf/2303.17367v1
2023-03-30T13:29:49
2023-03-30T13:29:49
{ "title": [ 0.0008506775, 0.00006574392, -0.00932312, -0.060028076, -0.00090169907, 0.015342712, -0.04824829, -0.0045166016, 0.012374878, 0.038238525, -0.026947021, 0.008102417, 0.02545166, -0.002735138, -0.053955078, 0.0076141357, -0.04824829, ...
http://arxiv.org/abs/2303.18062v1
Solving morphological analogies: from retrieval to generation
Esteban Marquer, Miguel Couceiro
cs.CL, cs.AI, cs.LG
Analogical inference is a remarkable capability of human reasoning, and has been used to solve hard reasoning tasks. Analogy based reasoning (AR) has gained increasing interest from the artificial intelligence community and has shown its potential in multiple machine learning tasks such as classification, decision maki...
http://arxiv.org/pdf/2303.18062v1
2023-03-30T12:36:46
2023-03-30T12:36:46
{ "title": [ -0.013679504, 0.0007100105, 0.007987976, -0.038024902, 0.0012989044, 0.016921997, -0.03201294, -0.032592773, 0.027130127, 0.033935547, -0.003288269, -0.0058250427, 0.0039520264, -0.024108887, -0.050964355, 0.020721436, -0.011413574, ...
http://arxiv.org/abs/2303.17324v1
Topics in the Haystack: Extracting and Evaluating Topics beyond Coherence
Anton Thielmann, Quentin Seifert, Arik Reuter, Elisabeth Bergherr, Benjamin Säfken
cs.CL
Extracting and identifying latent topics in large text corpora has gained increasing importance in Natural Language Processing (NLP). Most models, whether probabilistic models similar to Latent Dirichlet Allocation (LDA) or neural topic models, follow the same underlying approach of topic interpretability and topic ext...
http://arxiv.org/pdf/2303.17324v1
2023-03-30T12:24:25
2023-03-30T12:24:25
{ "title": [ -0.019699097, 0.011001587, -0.06048584, -0.05441284, 0.022766113, 0.002210617, -0.041656494, -0.04534912, 0.016830444, 0.018920898, -0.0060806274, 0.026565552, -0.033294678, -0.0041656494, -0.039733887, 0.007019043, -0.012954712, 0.0...
http://arxiv.org/abs/2303.17322v1
Yes but.. Can ChatGPT Identify Entities in Historical Documents?
Carlos-Emiliano González-Gallardo, Emanuela Boros, Nancy Girdhar, Ahmed Hamdi, Jose G. Moreno, Antoine Doucet
cs.DL, cs.CL, cs.IR
Large language models (LLMs) have been leveraged for several years now, obtaining state-of-the-art performance in recognizing entities from modern documents. For the last few months, the conversational agent ChatGPT has "prompted" a lot of interest in the scientific community and public due to its capacity of generatin...
http://arxiv.org/pdf/2303.17322v1
2023-03-30T12:23:39
2023-03-30T12:23:39
{ "title": [ 0.018615723, 0.018341064, -0.0134887695, 0.012336731, 0.007007599, -0.01940918, 0.00630188, -0.04083252, 0.045410156, 0.021697998, -0.031585693, 0.02230835, 0.019897461, -0.017974854, -0.0021362305, 0.013320923, -0.03744507, 0.047668...
http://arxiv.org/abs/2303.17276v1
Humans in Humans Out: On GPT Converging Toward Common Sense in both Success and Failure
Philipp Koralus, Vincent Wang-Maścianica
cs.AI, cs.CL, cs.HC, cs.LG, 00, 68, I.2.0; I.2.6
Increase in computational scale and fine-tuning has seen a dramatic improvement in the quality of outputs of large language models (LLMs) like GPT. Given that both GPT-3 and GPT-4 were trained on large quantities of human-generated text, we might ask to what extent their outputs reflect patterns of human thinking, both...
http://arxiv.org/pdf/2303.17276v1
2023-03-30T10:32:18
2023-03-30T10:32:18
{ "title": [ 0.011184692, -0.015335083, -0.011886597, -0.012313843, -0.011306763, -0.03152466, 0.008087158, -0.022460938, -0.022125244, 0.030776978, -0.017242432, 0.021194458, -0.010353088, -0.03164673, -0.01449585, -0.018722534, -0.0115737915, 0...
http://arxiv.org/abs/2303.17183v1
The Nordic Pile: A 1.2TB Nordic Dataset for Language Modeling
Joey Öhman, Severine Verlinden, Ariel Ekgren, Amaru Cuba Gyllensten, Tim Isbister, Evangelia Gogoulou, Fredrik Carlsson, Magnus Sahlgren
cs.CL, cs.AI
Pre-training Large Language Models (LLMs) require massive amounts of text data, and the performance of the LLMs typically correlates with the scale and quality of the datasets. This means that it may be challenging to build LLMs for smaller languages such as Nordic ones, where the availability of text corpora is limite...
http://arxiv.org/pdf/2303.17183v1
2023-03-30T06:42:22
2023-03-30T06:42:22
{ "title": [ -0.016052246, -0.021530151, -0.046539307, -0.0725708, 0.01158905, -0.004840851, -0.048858643, 0.024902344, 0.008041382, 0.033416748, -0.010108948, 0.003282547, 0.045013428, -0.031188965, -0.02558899, 0.026229858, -0.002439499, 0.0373...
http://arxiv.org/abs/2304.03287v1
Synthesis of Mathematical programs from Natural Language Specifications
Ganesh Prasath, Shirish Karande
cs.AI, cs.CL, cs.LG
Several decision problems that are encountered in various business domains can be modeled as mathematical programs, i.e. optimization problems. The process of conducting such modeling often requires the involvement of experts trained in operations research and advanced algorithms. Surprisingly, despite the significant ...
http://arxiv.org/pdf/2304.03287v1
2023-03-30T06:10:00
2023-03-30T06:10:00
{ "title": [ -0.03363037, 0.0014848709, -0.019363403, -0.057403564, 0.00037002563, 0.022323608, -0.06506348, -0.017974854, -0.010139465, 0.01374054, -0.03125, 0.006790161, -0.02947998, -0.0030078888, -0.030136108, -0.013618469, -0.008674622, 0.00...
http://arxiv.org/abs/2303.17161v1
TreePiece: Faster Semantic Parsing via Tree Tokenization
Sid Wang, Akshat Shrivastava, Sasha Livshits
cs.CL, cs.AI, I.2.7
Autoregressive (AR) encoder-decoder neural networks have proved successful in many NLP problems, including Semantic Parsing -- a task that translates natural language to machine-readable parse trees. However, the sequential prediction process of AR models can be slow. To accelerate AR for semantic parsing, we introduce...
http://arxiv.org/pdf/2303.17161v1
2023-03-30T05:44:44
2023-03-30T05:44:44
{ "title": [ -0.04019165, -0.018081665, -0.05078125, -0.09326172, 0.008232117, 0.021896362, -0.062683105, 0.015914917, -0.0050239563, 0.026367188, -0.04156494, 0.018051147, -0.014190674, 0.00459671, -0.04647827, -0.0075950623, -0.017868042, 0.051...
http://arxiv.org/abs/2304.06034v1
Social Biases through the Text-to-Image Generation Lens
Ranjita Naik, Besmira Nushi
cs.CY, cs.AI, cs.CL, cs.CV
Text-to-Image (T2I) generation is enabling new applications that support creators, designers, and general end users of productivity software by generating illustrative content with high photorealism starting from a given descriptive text as a prompt. Such models are however trained on massive amounts of web data, which...
http://arxiv.org/pdf/2304.06034v1
2023-03-30T05:29:13
2023-03-30T05:29:13
{ "title": [ 0.01763916, 0.002286911, -0.0002939701, -0.013084412, -0.013694763, -0.021392822, 0.00774765, -0.042541504, 0.0011453629, 0.033569336, -0.040802002, 0.017471313, -0.0064353943, -0.009132385, -0.025848389, -0.010406494, -0.013702393, ...
http://arxiv.org/abs/2303.17119v1
TLAG: An Informative Trigger and Label-Aware Knowledge Guided Model for Dialogue-based Relation Extraction
Hao An, Dongsheng Chen, Weiyuan Xu, Zhihong Zhu, Yuexian Zou
cs.CL
Dialogue-based Relation Extraction (DRE) aims to predict the relation type of argument pairs that are mentioned in dialogue. The latest trigger-enhanced methods propose trigger prediction tasks to promote DRE. However, these methods are not able to fully leverage the trigger information and even bring noise to relation...
http://arxiv.org/pdf/2303.17119v1
2023-03-30T03:10:28
2023-03-30T03:10:28
{ "title": [ -0.006576538, -0.012130737, -0.054779053, -0.09106445, -0.00440979, 0.009941101, -0.06933594, 0.023666382, 0.02218628, 0.00073099136, -0.01109314, 0.0017414093, -0.004295349, -0.020736694, -0.045806885, -0.00315094, -0.014350891, 0.0...
http://arxiv.org/abs/2303.17612v3
oBERTa: Improving Sparse Transfer Learning via improved initialization, distillation, and pruning regimes
Daniel Campos, Alexandre Marques, Mark Kurtz, ChengXiang Zhai
cs.CL, cs.AI, cs.LG
In this paper, we introduce the range of oBERTa language models, an easy-to-use set of language models which allows Natural Language Processing (NLP) practitioners to obtain between 3.8 and 24.3 times faster models without expertise in model compression. Specifically, oBERTa extends existing work on pruning, knowledge ...
http://arxiv.org/pdf/2303.17612v3
2023-06-06T16:30:09
2023-03-30T01:37:19
{ "title": [ -0.028579712, -0.010803223, -0.034729004, -0.042755127, 0.0041885376, 0.009094238, -0.030532837, -0.008979797, 0.024169922, 0.01109314, -0.010673523, 0.024505615, -0.01399231, -0.02142334, -0.021713257, 0.020523071, 0.0076675415, 0.0...
http://arxiv.org/abs/2304.01003v1
QUADRo: Dataset and Models for QUestion-Answer Database Retrieval
Stefano Campese, Ivano Lauriola, Alessandro Moschitti
cs.CL, cs.LG
An effective paradigm for building Automated Question Answering systems is the re-use of previously answered questions, e.g., for FAQs or forum applications. Given a database (DB) of question/answer (q/a) pairs, it is possible to answer a target question by scanning the DB for similar questions. In this paper, we scale...
http://arxiv.org/pdf/2304.01003v1
2023-03-30T00:42:07
2023-03-30T00:42:07
{ "title": [ -0.013259888, -0.012557983, -0.03387451, -0.05593872, 0.0071525574, -0.00006312132, -0.010131836, 0.0017490387, 0.044128418, 0.03552246, -0.027404785, 0.004016876, -0.0032444, 0.0012874603, -0.08129883, 0.008018494, -0.002023697, 0.0...
http://arxiv.org/abs/2303.17071v1
DERA: Enhancing Large Language Model Completions with Dialog-Enabled Resolving Agents
Varun Nair, Elliot Schumacher, Geoffrey Tso, Anitha Kannan
cs.CL, cs.AI, cs.LG
Large language models (LLMs) have emerged as valuable tools for many natural language understanding tasks. In safety-critical applications such as healthcare, the utility of these models is governed by their ability to generate outputs that are factually accurate and complete. In this work, we present dialog-enabled re...
http://arxiv.org/pdf/2303.17071v1
2023-03-30T00:30:19
2023-03-30T00:30:19
{ "title": [ -0.0055007935, -0.020721436, -0.03955078, -0.10083008, 0.000385046, 0.0061416626, -0.08081055, 0.024795532, 0.01889038, 0.04534912, -0.02279663, 0.008201599, 0.03717041, -0.0023403168, -0.05291748, 0.024169922, -0.027786255, 0.045867...
http://arxiv.org/abs/2303.17006v1
How do decoding algorithms distribute information in dialogue responses?
Saranya Venkatraman, He He, David Reitter
cs.CL
Humans tend to follow the Uniform Information Density (UID) principle by distributing information evenly in utterances. We study if decoding algorithms implicitly follow this UID principle, and under what conditions adherence to UID might be desirable for dialogue generation. We generate responses using different decod...
http://arxiv.org/pdf/2303.17006v1
2023-03-29T20:21:45
2023-03-29T20:21:45
{ "title": [ -0.034179688, -0.010414124, -0.019714355, 0.016464233, -0.033447266, -0.0060653687, -0.0058135986, -0.014984131, 0.00248909, 0.039642334, -0.0037994385, -0.006324768, -0.00856781, -0.03201294, -0.023895264, 0.0027999878, 0.00484848, ...
http://arxiv.org/abs/2303.17003v1
Evaluating GPT-3.5 and GPT-4 Models on Brazilian University Admission Exams
Desnes Nunes, Ricardo Primi, Ramon Pires, Roberto Lotufo, Rodrigo Nogueira
cs.CL, cs.AI, cs.LG
The present study aims to explore the capabilities of Language Models (LMs) in tackling high-stakes multiple-choice tests, represented here by the Exame Nacional do Ensino M\'edio (ENEM), a multidisciplinary entrance examination widely adopted by Brazilian universities. This exam poses challenging tasks for LMs, since ...
http://arxiv.org/pdf/2303.17003v1
2023-03-29T20:10:13
2023-03-29T20:10:13
{ "title": [ 0.0009851456, 0.017120361, 0.0017080307, 0.008476257, -0.008270264, -0.0015230179, 0.00017261505, -0.050201416, -0.009399414, 0.027404785, -0.012367249, -0.013847351, -0.012458801, -0.032043457, -0.046569824, -0.010810852, -0.035003662, ...
http://arxiv.org/abs/2303.16992v2
ContraSim -- A Similarity Measure Based on Contrastive Learning
Adir Rahamim, Yonatan Belinkov
cs.CL
Recent work has compared neural network representations via similarity-based analyses to improve model interpretation. The quality of a similarity measure is typically evaluated by its success in assigning a high score to representations that are expected to be matched. However, existing similarity measures perform med...
http://arxiv.org/pdf/2303.16992v2
2023-05-30T09:47:33
2023-03-29T19:43:26
{ "title": [ -0.01184845, 0.013473511, -0.03668213, -0.007194519, 0.014907837, -0.025802612, 0.0032138824, -0.066345215, 0.03753662, 0.004573822, -0.04434204, 0.0044555664, -0.027374268, -0.038360596, -0.03744507, -0.0032844543, -0.0078086853, 0....
http://arxiv.org/abs/2303.16985v1
Adapting to the Low-Resource Double-Bind: Investigating Low-Compute Methods on Low-Resource African Languages
Colin Leong, Herumb Shandilya, Bonaventure F. P. Dossou, Atnafu Lambebo Tonja, Joel Mathew, Abdul-Hakeem Omotayo, Oreen Yousuf, Zainab Akinjobi, Chris Chinenye Emezue, Shamsudeen Muhammad, Steven Kolawole, Younwoo Choi, Tosin Adewumi
cs.CL, cs.AI
Many natural language processing (NLP) tasks make use of massively pre-trained language models, which are computationally expensive. However, access to high computational resources added to the issue of data scarcity of African languages constitutes a real barrier to research experiments on these languages. In this wor...
http://arxiv.org/pdf/2303.16985v1
2023-03-29T19:25:43
2023-03-29T19:25:43
{ "title": [ -0.03543091, -0.032440186, 0.0035648346, -0.0463562, -0.040618896, 0.018371582, -0.051361084, 0.015258789, -0.015029907, 0.055877686, 0.011734009, 0.014343262, 0.027496338, -0.027389526, -0.040527344, 0.02911377, -0.0015735626, 0.016...
http://arxiv.org/abs/2303.16974v1
BEVERS: A General, Simple, and Performant Framework for Automatic Fact Verification
Mitchell DeHaven, Stephen Scott
cs.CL
Automatic fact verification has become an increasingly popular topic in recent years and among datasets the Fact Extraction and VERification (FEVER) dataset is one of the most popular. In this work we present BEVERS, a tuned baseline system for the FEVER dataset. Our pipeline uses standard approaches for document retri...
http://arxiv.org/pdf/2303.16974v1
2023-03-29T19:16:19
2023-03-29T19:16:19
{ "title": [ -0.009605408, -0.0019931793, -0.033843994, -0.08148193, 0.02507019, 0.012458801, -0.03729248, -0.012054443, -0.0099105835, 0.03994751, -0.008728027, 0.02709961, -0.0055351257, -0.008094788, -0.04196167, 0.0017576218, -0.03387451, 0.0...
http://arxiv.org/abs/2303.16894v3
ViewRefer: Grasp the Multi-view Knowledge for 3D Visual Grounding with GPT and Prototype Guidance
Zoey Guo, Yiwen Tang, Ray Zhang, Dong Wang, Zhigang Wang, Bin Zhao, Xuelong Li
cs.CV, cs.AI, cs.CL
Understanding 3D scenes from multi-view inputs has been proven to alleviate the view discrepancy issue in 3D visual grounding. However, existing methods normally neglect the view cues embedded in the text modality and fail to weigh the relative importance of different views. In this paper, we propose ViewRefer, a multi...
http://arxiv.org/pdf/2303.16894v3
2023-08-24T18:45:43
2023-03-29T17:59:10
{ "title": [ 0.006500244, -0.00957489, -0.06604004, -0.0597229, -0.019165039, -0.028793335, -0.044921875, 0.011238098, 0.0025043488, -0.00008177757, -0.055267334, 0.017868042, 0.0025177002, -0.03201294, 0.0045433044, -0.03375244, 0.016098022, 0.0...
http://arxiv.org/abs/2303.16886v1
End-to-End $n$-ary Relation Extraction for Combination Drug Therapies
Yuhang Jiang, Ramakanth Kavuluru
cs.CL
Combination drug therapies are treatment regimens that involve two or more drugs, administered more commonly for patients with cancer, HIV, malaria, or tuberculosis. Currently there are over 350K articles in PubMed that use the "combination drug therapy" MeSH heading with at least 10K articles published per year over t...
http://arxiv.org/pdf/2303.16886v1
2023-03-29T17:55:50
2023-03-29T17:55:50
{ "title": [ 0.009246826, 0.0045928955, -0.04928589, -0.10888672, 0.022903442, -0.007663727, -0.04486084, 0.013496399, -0.011650085, 0.018692017, 0.010635376, 0.011566162, -0.029693604, -0.038513184, -0.029708862, 0.0024108887, -0.007949829, 0.02...
http://arxiv.org/abs/2303.16857v3
Did You Mean...? Confidence-based Trade-offs in Semantic Parsing
Elias Stengel-Eskin, Benjamin Van Durme
cs.CL
We illustrate how a calibrated model can help balance common trade-offs in task-oriented parsing. In a simulated annotator-in-the-loop experiment, we show that well-calibrated confidence scores allow us to balance cost with annotator load, improving accuracy with a small number of interactions. We then examine how conf...
http://arxiv.org/pdf/2303.16857v3
2023-10-20T12:54:30
2023-03-29T17:07:26
{ "title": [ -0.023880005, 0.0026168823, -0.061920166, -0.04309082, -0.012954712, -0.030181885, -0.04083252, -0.0064468384, -0.0317688, 0.017166138, -0.032348633, 0.011131287, -0.017852783, -0.024017334, -0.014076233, -0.036010742, -0.017440796, ...
http://arxiv.org/abs/2303.16854v1
AnnoLLM: Making Large Language Models to Be Better Crowdsourced Annotators
Xingwei He, Zhenghao Lin, Yeyun Gong, A-Long Jin, Hang Zhang, Chen Lin, Jian Jiao, Siu Ming Yiu, Nan Duan, Weizhu Chen
cs.CL
Many natural language processing (NLP) tasks rely on labeled data to train machine learning models to achieve high performance. However, data annotation can be a time-consuming and expensive process, especially when the task involves a large amount of data or requires specialized domains. Recently, GPT-3.5 series model...
http://arxiv.org/pdf/2303.16854v1
2023-03-29T17:03:21
2023-03-29T17:03:21
{ "title": [ -0.021743774, -0.0073547363, -0.028411865, -0.075805664, 0.0016231537, 0.014846802, -0.058624268, 0.008026123, 0.0052948, 0.033477783, -0.013389587, 0.0058174133, 0.008705139, 0.002664566, -0.050476074, 0.011482239, -0.007865906, 0.0...
http://arxiv.org/abs/2303.16839v3
MaMMUT: A Simple Architecture for Joint Learning for MultiModal Tasks
Weicheng Kuo, AJ Piergiovanni, Dahun Kim, Xiyang Luo, Ben Caine, Wei Li, Abhijit Ogale, Luowei Zhou, Andrew Dai, Zhifeng Chen, Claire Cui, Anelia Angelova
cs.CV, cs.CL, cs.LG
The development of language models have moved from encoder-decoder to decoder-only designs. In addition, we observe that the two most popular multimodal tasks, the generative and contrastive tasks, are nontrivial to accommodate in one architecture, and further need adaptations for downstream tasks. We propose a novel p...
http://arxiv.org/pdf/2303.16839v3
2023-08-09T05:39:34
2023-03-29T16:42:30
{ "title": [ -0.039886475, 0.0043678284, -0.047546387, -0.044433594, 0.016860962, -0.007385254, -0.03756714, -0.017593384, -0.0115356445, 0.030685425, -0.008079529, 0.0053253174, -0.0021305084, -0.018539429, -0.032287598, 0.0003259182, -0.016418457, ...
http://arxiv.org/abs/2303.16835v1
Zero-shot Entailment of Leaderboards for Empirical AI Research
Salomon Kabongo, Jennifer D'Souza, Sören Auer
cs.CL, cs.AI, cs.LG
We present a large-scale empirical investigation of the zero-shot learning phenomena in a specific recognizing textual entailment (RTE) task category, i.e. the automated mining of leaderboards for Empirical AI Research. The prior reported state-of-the-art models for leaderboards extraction formulated as an RTE task, in...
http://arxiv.org/pdf/2303.16835v1
2023-03-29T16:28:43
2023-03-29T16:28:43
{ "title": [ 0.013214111, -0.016357422, -0.016983032, -0.017654419, -0.016662598, -0.028259277, 0.046051025, 0.0032215118, 0.0090789795, 0.033477783, -0.030303955, 0.035247803, -0.005886078, -0.012794495, -0.012702942, 0.009536743, -0.013656616, ...
http://arxiv.org/abs/2303.16742v1
Evaluating NLG systems: A brief introduction
Emiel van Miltenburg
cs.CL
This year the International Conference on Natural Language Generation (INLG) will feature an award for the paper with the best evaluation. The purpose of this award is to provide an incentive for NLG researchers to pay more attention to the way they assess the output of their systems. This essay provides a short introd...
http://arxiv.org/pdf/2303.16742v1
2023-03-29T14:49:29
2023-03-29T14:49:29
{ "title": [ -0.028182983, -0.026382446, -0.014038086, -0.037353516, 0.000992775, 0.0060272217, -0.017364502, -0.0028190613, -0.0064353943, 0.036834717, -0.04257202, 0.0010910034, -0.01979065, 0.0048103333, -0.08215332, -0.01612854, -0.019302368, ...
http://arxiv.org/abs/2303.16726v1
Text revision in Scientific Writing Assistance: An Overview
Léane Jourdan, Florian Boudin, Richard Dufour, Nicolas Hernandez
cs.CL
Writing a scientific article is a challenging task as it is a highly codified genre. Good writing skills are essential to properly convey ideas and results of research work. Since the majority of scientific articles are currently written in English, this exercise is all the more difficult for non-native English speaker...
http://arxiv.org/pdf/2303.16726v1
2023-03-29T14:25:30
2023-03-29T14:25:30
{ "title": [ -0.05770874, -0.012176514, -0.029922485, -0.03640747, -0.021408081, 0.018157959, -0.053588867, 0.003080368, -0.016098022, 0.046569824, -0.018539429, 0.004508972, 0.017440796, -0.0062675476, -0.05609131, 0.0031089783, -0.015731812, 0....
http://arxiv.org/abs/2303.16694v1
Using Semantic Similarity and Text Embedding to Measure the Social Media Echo of Strategic Communications
Tristan J. B. Cann, Ben Dennes, Travis Coan, Saffron O'Neill, Hywel T. P. Williams
cs.SI, cs.CL
Online discourse covers a wide range of topics and many actors tailor their content to impact online discussions through carefully crafted messages and targeted campaigns. Yet the scale and diversity of online media content make it difficult to evaluate the impact of a particular message. In this paper, we present a ne...
http://arxiv.org/pdf/2303.16694v1
2023-03-29T13:46:07
2023-03-29T13:46:07
{ "title": [ -0.009597778, 0.01512146, -0.019119263, -0.007888794, 0.011680603, -0.0129470825, -0.009117126, -0.040985107, -0.024536133, 0.041259766, -0.018035889, 0.045837402, -0.0061302185, -0.008285522, -0.023864746, -0.018844604, -0.009292603, ...
http://arxiv.org/abs/2303.16657v1
Summarizing Indian Languages using Multilingual Transformers based Models
Dhaval Taunk, Vasudeva Varma
cs.CL
With the advent of multilingual models like mBART, mT5, IndicBART etc., summarization in low resource Indian languages is getting a lot of attention now a days. But still the number of datasets is low in number. In this work, we (Team HakunaMatata) study how these multilingual models perform on the datasets which have ...
http://arxiv.org/pdf/2303.16657v1
2023-03-29T13:05:17
2023-03-29T13:05:17
{ "title": [ -0.0011177063, 0.010238647, -0.012901306, -0.08850098, -0.01159668, 0.027862549, -0.10144043, -0.0070648193, -0.016174316, 0.050598145, -0.019424438, 0.010932922, 0.014823914, -0.0061302185, -0.052734375, -0.0012025833, -0.0017662048, ...
http://arxiv.org/abs/2303.16634v3
G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment
Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, Chenguang Zhu
cs.CL, cs.AI
The quality of texts generated by natural language generation (NLG) systems is hard to measure automatically. Conventional reference-based metrics, such as BLEU and ROUGE, have been shown to have relatively low correlation with human judgments, especially for tasks that require creativity and diversity. Recent studies ...
http://arxiv.org/pdf/2303.16634v3
2023-05-23T22:12:16
2023-03-29T12:46:54
{ "title": [ -0.015579224, 0.0046539307, -0.04220581, -0.05847168, 0.020767212, -0.009613037, -0.031280518, -0.012207031, -0.01637268, 0.022567749, 0.000024616718, -0.014144897, -0.006290436, -0.038879395, -0.046173096, -0.0066070557, -0.01689148, ...
http://arxiv.org/abs/2303.16621v1
AraSpot: Arabic Spoken Command Spotting
Mahmoud Salhab, Haidar Harmanani
cs.CL, cs.AI
Spoken keyword spotting (KWS) is the task of identifying a keyword in an audio stream and is widely used in smart devices at the edge in order to activate voice assistants and perform hands-free tasks. The task is daunting as there is a need, on the one hand, to achieve high accuracy while at the same time ensuring tha...
http://arxiv.org/pdf/2303.16621v1
2023-03-29T12:22:17
2023-03-29T12:22:17
{ "title": [ -0.0060653687, -0.0067977905, -0.046722412, -0.021438599, -0.03793335, -0.013252258, -0.046325684, -0.0143966675, -0.016494751, 0.02986145, -0.0045166016, -0.029266357, 0.030441284, -0.060180664, 0.009735107, -0.015594482, 0.016235352, ...