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/2305.07766v1
NL2TL: Transforming Natural Languages to Temporal Logics using Large Language Models
Yongchao Chen, Rujul Gandhi, Yang Zhang, Chuchu Fan
cs.CL
Temporal Logic (TL) can be used to rigorously specify complex high-level specification for systems in many engineering applications. The translation between natural language (NL) and TL has been under-explored due to the lack of dataset and generalizable model across different application domains. In this paper, we pro...
http://arxiv.org/pdf/2305.07766v1
2023-05-12T21:22:08
2023-05-12T21:22:08
{ "title": [ -0.022628784, -0.021316528, -0.06008911, -0.109558105, 0.020187378, 0.009880066, -0.06561279, 0.03503418, 0.002943039, 0.02684021, -0.01033783, -0.0062713623, 0.008926392, -0.017532349, -0.044830322, -0.00027799606, -0.026977539, 0.0...
http://arxiv.org/abs/2305.07763v1
Knowledge Authoring for Rules and Actions
Yuheng Wang, Paul Fodor, Michael Kifer
cs.CL, cs.AI
Knowledge representation and reasoning (KRR) systems describe and reason with complex concepts and relations in the form of facts and rules. Unfortunately, wide deployment of KRR systems runs into the problem that domain experts have great difficulty constructing correct logical representations of their domain knowledg...
http://arxiv.org/pdf/2305.07763v1
2023-05-12T21:08:35
2023-05-12T21:08:35
{ "title": [ -0.0007419586, -0.011444092, -0.08294678, -0.06970215, -0.04611206, -0.027404785, -0.06451416, 0.03024292, -0.040863037, 0.043762207, -0.050231934, 0.0009899139, 0.012939453, -0.017669678, 0.0025405884, -0.022659302, 0.033203125, 0.0...
http://arxiv.org/abs/2305.07760v2
Unsupervised Melody-Guided Lyrics Generation
Yufei Tian, Anjali Narayan-Chen, Shereen Oraby, Alessandra Cervone, Gunnar Sigurdsson, Chenyang Tao, Wenbo Zhao, Tagyoung Chung, Jing Huang, Nanyun Peng
cs.AI, cs.CL, cs.MM
Automatic song writing is a topic of significant practical interest. However, its research is largely hindered by the lack of training data due to copyright concerns and challenged by its creative nature. Most noticeably, prior works often fall short of modeling the cross-modal correlation between melody and lyrics due...
http://arxiv.org/pdf/2305.07760v2
2023-05-26T00:56:58
2023-05-12T20:57:20
{ "title": [ -0.02861023, -0.013061523, -0.045013428, -0.051239014, -0.011878967, -0.0007753372, -0.037384033, 0.027404785, 0.0022964478, 0.04220581, -0.05105591, 0.015487671, 0.003370285, 0.0019483566, -0.036895752, -0.0018863678, -0.013671875, ...
http://arxiv.org/abs/2305.07759v2
TinyStories: How Small Can Language Models Be and Still Speak Coherent English?
Ronen Eldan, Yuanzhi Li
cs.CL, cs.AI, cs.LG
Language models (LMs) are powerful tools for natural language processing, but they often struggle to produce coherent and fluent text when they are small. Models with around 125M parameters such as GPT-Neo (small) or GPT-2 (small) can rarely generate coherent and consistent English text beyond a few words even after ex...
http://arxiv.org/pdf/2305.07759v2
2023-05-24T23:30:43
2023-05-12T20:56:48
{ "title": [ -0.05065918, -0.008354187, -0.05822754, -0.03189087, -0.026428223, -0.012199402, -0.055908203, 0.0038261414, -0.025650024, 0.03829956, -0.012382507, -0.0036792755, 0.023590088, -0.03201294, -0.04019165, 0.003566742, 0.003627777, 0.02...
http://arxiv.org/abs/2305.07717v1
Parallel Tree Kernel Computation
Souad Taouti, Hadda Cherroun, Djelloul Ziadi
cs.CL
Tree kernels are fundamental tools that have been leveraged in many applications, particularly those based on machine learning for Natural Language Processing tasks. In this paper, we devise a parallel implementation of the sequential algorithm for the computation of some tree kernels of two finite sets of trees (Ouali...
http://arxiv.org/pdf/2305.07717v1
2023-05-12T18:16:45
2023-05-12T18:16:45
{ "title": [ -0.063964844, 0.0010147095, -0.030853271, -0.051635742, -0.028396606, 0.0234375, -0.06335449, -0.0027809143, 0.019424438, 0.062286377, -0.009162903, -0.023040771, 0.0029506683, -0.026565552, -0.07080078, 0.014259338, 0.0079574585, 0....
http://arxiv.org/abs/2305.07709v1
Using Language Models to Detect Alarming Student Responses
Christopher M. Ormerod, Milan Patel, Harry Wang
cs.CL, cs.IR, cs.LG
This article details the advances made to a system that uses artificial intelligence to identify alarming student responses. This system is built into our assessment platform to assess whether a student's response indicates they are a threat to themselves or others. Such responses may include details concerning threats...
http://arxiv.org/pdf/2305.07709v1
2023-05-12T18:07:00
2023-05-12T18:07:00
{ "title": [ 0.022201538, 0.028259277, -0.0075149536, -0.016647339, -0.0043945312, -0.02204895, -0.017318726, -0.049591064, -0.00023853779, 0.028274536, -0.030838013, -0.018630981, 0.00024604797, -0.031341553, -0.038085938, -0.017669678, -0.006732940...
http://arxiv.org/abs/2305.07637v2
Text2Cohort: Democratizing the NCI Imaging Data Commons with Natural Language Cohort Discovery
Pranav Kulkarni, Adway Kanhere, Paul H. Yi, Vishwa S. Parekh
cs.LG, cs.CL, cs.HC, cs.IR
The Imaging Data Commons (IDC) is a cloud-based database that provides researchers with open access to cancer imaging data, with the goal of facilitating collaboration in medical imaging research. However, querying the IDC database for cohort discovery and access to imaging data has a significant learning curve for res...
http://arxiv.org/pdf/2305.07637v2
2023-05-16T13:50:33
2023-05-12T17:46:06
{ "title": [ -0.03955078, -0.010108948, -0.054382324, -0.03466797, -0.01902771, -0.005393982, -0.048461914, -0.0055503845, 0.01399231, 0.012962341, -0.054626465, 0.010246277, -0.015029907, -0.01940918, -0.019134521, -0.0079956055, -0.015304565, 0...
http://arxiv.org/abs/2305.07622v3
PALR: Personalization Aware LLMs for Recommendation
Fan Yang, Zheng Chen, Ziyan Jiang, Eunah Cho, Xiaojiang Huang, Yanbin Lu
cs.IR, cs.AI, cs.CL, I.2.6; I.2.7
Large language models (LLMs) have recently received significant attention for their exceptional capabilities. Despite extensive efforts in developing general-purpose LLMs that can be utilized in various natural language processing (NLP) tasks, there has been less research exploring their potential in recommender system...
http://arxiv.org/pdf/2305.07622v3
2023-06-07T17:55:58
2023-05-12T17:21:33
{ "title": [ -0.033081055, -0.006828308, -0.048461914, -0.052703857, -0.0018844604, 0.010673523, -0.03439331, 0.0041236877, -0.028015137, 0.022827148, -0.01675415, 0.041656494, -0.027633667, -0.01586914, -0.015960693, 0.0025691986, -0.016952515, ...
http://arxiv.org/abs/2305.07615v1
What are the Desired Characteristics of Calibration Sets? Identifying Correlates on Long Form Scientific Summarization
Griffin Adams, Bichlien H Nguyen, Jake Smith, Yingce Xia, Shufang Xie, Anna Ostropolets, Budhaditya Deb, Yuan-Jyue Chen, Tristan Naumann, Noémie Elhadad
cs.CL
Summarization models often generate text that is poorly calibrated to quality metrics because they are trained to maximize the likelihood of a single reference (MLE). To address this, recent work has added a calibration step, which exposes a model to its own ranked outputs to improve relevance or, in a separate line of...
http://arxiv.org/pdf/2305.07615v1
2023-05-12T17:08:47
2023-05-12T17:08:47
{ "title": [ -0.04260254, 0.05770874, -0.058532715, 0.019592285, 0.0017614365, -0.015266418, -0.022079468, -0.040863037, -0.041046143, 0.05368042, -0.0046043396, 0.0068206787, -0.017700195, -0.036315918, -0.0074310303, -0.023956299, 0.016723633, ...
http://arxiv.org/abs/2305.07614v1
NevIR: Negation in Neural Information Retrieval
Orion Weller, Dawn Lawrie, Benjamin Van Durme
cs.IR, cs.CL
Negation is a common everyday phenomena and has been a consistent area of weakness for language models (LMs). Although the Information Retrieval (IR) community has adopted LMs as the backbone of modern IR architectures, there has been little to no research in understanding how negation impacts neural IR. We therefore c...
http://arxiv.org/pdf/2305.07614v1
2023-05-12T17:05:54
2023-05-12T17:05:54
{ "title": [ 0.004501343, -0.0040664673, -0.012451172, -0.016693115, 0.0030403137, -0.013664246, 0.009674072, -0.042785645, 0.0005054474, 0.03414917, -0.040130615, -0.012573242, 0.025115967, -0.010986328, -0.045288086, -0.0022449493, -0.014045715, ...
http://arxiv.org/abs/2305.07611v3
Multimodal Sentiment Analysis: A Survey
Songning Lai, Xifeng Hu, Haoxuan Xu, Zhaoxia Ren, Zhi Liu
cs.CL, cs.CV
Multimodal sentiment analysis has become an important research area in the field of artificial intelligence. With the latest advances in deep learning, this technology has reached new heights. It has great potential for both application and research, making it a popular research topic. This review provides an overview ...
http://arxiv.org/pdf/2305.07611v3
2023-07-03T19:05:30
2023-05-12T16:56:13
{ "title": [ -0.013710022, 0.0079422, -0.032562256, -0.049621582, 0.0128479, -0.008605957, -0.02848816, -0.011428833, 0.018554688, 0.049346924, -0.015205383, -0.0035743713, 0.0031375885, -0.0067863464, -0.06713867, 0.0038318634, -0.017745972, 0.0...
http://arxiv.org/abs/2305.07609v3
Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation
Jizhi Zhang, Keqin Bao, Yang Zhang, Wenjie Wang, Fuli Feng, Xiangnan He
cs.IR, cs.CL, cs.CY
The remarkable achievements of Large Language Models (LLMs) have led to the emergence of a novel recommendation paradigm -- Recommendation via LLM (RecLLM). Nevertheless, it is important to note that LLMs may contain social prejudices, and therefore, the fairness of recommendations made by RecLLM requires further inves...
http://arxiv.org/pdf/2305.07609v3
2023-10-17T13:29:54
2023-05-12T16:54:36
{ "title": [ -0.009178162, -0.008270264, -0.029724121, -0.015213013, -0.004676819, -0.008766174, -0.011100769, -0.0082092285, -0.017700195, 0.027923584, -0.040008545, 0.014862061, 0.00957489, -0.014663696, -0.023010254, -0.0038852692, -0.035217285, ...
http://arxiv.org/abs/2305.07565v1
A Memory Model for Question Answering from Streaming Data Supported by Rehearsal and Anticipation of Coreference Information
Vladimir Araujo, Alvaro Soto, Marie-Francine Moens
cs.CL, cs.AI, cs.LG
Existing question answering methods often assume that the input content (e.g., documents or videos) is always accessible to solve the task. Alternatively, memory networks were introduced to mimic the human process of incremental comprehension and compression of the information in a fixed-capacity memory. However, these...
http://arxiv.org/pdf/2305.07565v1
2023-05-12T15:46:36
2023-05-12T15:46:36
{ "title": [ 0.009742737, -0.005683899, -0.0044784546, -0.07305908, -0.012680054, 0.0012865067, -0.033447266, 0.011665344, 0.016601562, 0.048614502, -0.039154053, 0.0074691772, 0.0031051636, 0.014480591, -0.08129883, 0.0027999878, 0.014709473, 0....
http://arxiv.org/abs/2305.07558v1
Measuring Progress in Fine-grained Vision-and-Language Understanding
Emanuele Bugliarello, Laurent Sartran, Aishwarya Agrawal, Lisa Anne Hendricks, Aida Nematzadeh
cs.CL, cs.CV
While pretraining on large-scale image-text data from the Web has facilitated rapid progress on many vision-and-language (V&L) tasks, recent work has demonstrated that pretrained models lack "fine-grained" understanding, such as the ability to recognise relationships, verbs, and numbers in images. This has resulted in ...
http://arxiv.org/pdf/2305.07558v1
2023-05-12T15:34:20
2023-05-12T15:34:20
{ "title": [ -0.0063819885, -0.008468628, -0.03579712, -0.037017822, -0.006187439, -0.02720642, -0.017837524, -0.0058555603, 0.014442444, 0.025222778, -0.046417236, -0.0008993149, -0.0035915375, -0.013511658, -0.056640625, -0.011802673, 0.016540527, ...
http://arxiv.org/abs/2305.07507v2
LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development
Ilias Chalkidis, Nicolas Garneau, Catalina Goanta, Daniel Martin Katz, Anders Søgaard
cs.CL
In this work, we conduct a detailed analysis on the performance of legal-oriented pre-trained language models (PLMs). We examine the interplay between their original objective, acquired knowledge, and legal language understanding capacities which we define as the upstream, probing, and downstream performance, respectiv...
http://arxiv.org/pdf/2305.07507v2
2023-05-22T18:20:54
2023-05-12T14:21:38
{ "title": [ -0.06793213, 0.0048561096, -0.08441162, -0.046325684, -0.002626419, 0.008918762, -0.0725708, 0.021850586, -0.017684937, 0.027114868, -0.041870117, 0.021408081, 0.020111084, -0.020904541, 0.0019626617, 0.0026245117, -0.017486572, 0.04...
http://arxiv.org/abs/2305.07491v1
A Comprehensive Analysis of Adapter Efficiency
Nandini Mundra, Sumanth Doddapaneni, Raj Dabre, Anoop Kunchukuttan, Ratish Puduppully, Mitesh M. Khapra
cs.CL
Adapters have been positioned as a parameter-efficient fine-tuning (PEFT) approach, whereby a minimal number of parameters are added to the model and fine-tuned. However, adapters have not been sufficiently analyzed to understand if PEFT translates to benefits in training/deployment efficiency and maintainability/exten...
http://arxiv.org/pdf/2305.07491v1
2023-05-12T14:05:45
2023-05-12T14:05:45
{ "title": [ 0.05807495, 0.016540527, -0.0011949539, -0.05090332, -0.02204895, -0.035308838, -0.01838684, -0.025726318, 0.036315918, 0.053863525, 0.011886597, -0.010566711, 0.03652954, -0.048583984, 0.0063705444, 0.044189453, 0.030166626, 0.01278...
http://arxiv.org/abs/2305.07490v2
ArtGPT-4: Artistic Vision-Language Understanding with Adapter-enhanced MiniGPT-4
Zhengqing Yuan, Huiwen Xue, Xinyi Wang, Yongming Liu, Zhuanzhe Zhao, Kun Wang
cs.CL, cs.CV
In recent years, large language models (LLMs) have made significant progress in natural language processing (NLP), with models like ChatGPT and GPT-4 achieving impressive capabilities in various linguistic tasks. However, training models on such a large scale is challenging, and finding datasets that match the model's ...
http://arxiv.org/pdf/2305.07490v2
2023-05-30T14:51:28
2023-05-12T14:04:30
{ "title": [ -0.0076522827, -0.01272583, -0.034606934, -0.037353516, 0.0050697327, -0.019760132, -0.0059432983, -0.0057296753, 0.021118164, 0.016616821, 0.014434814, -0.017623901, 0.013389587, -0.060516357, -0.015556335, 0.013442993, 0.004711151, ...
http://arxiv.org/abs/2305.07475v1
Comprehensive Solution Program Centric Pretraining for Table-and-Text Hybrid Numerical Reasoning
Qianying Liu, Dongsheng Yang, Wenjie Zhong, Fei Cheng, Sadao Kurohashi
cs.CL
Numerical reasoning over table-and-text hybrid passages, such as financial reports, poses significant challenges and has numerous potential applications. Noise and irrelevant variables in the model input have been a hindrance to its performance. Additionally, coarse-grained supervision of the whole solution program has...
http://arxiv.org/pdf/2305.07475v1
2023-05-12T13:44:40
2023-05-12T13:44:40
{ "title": [ -0.010353088, -0.009300232, -0.03793335, -0.045959473, -0.00045132637, -0.02003479, -0.011405945, 0.0138168335, -0.028182983, 0.04159546, -0.024932861, 0.03591919, 0.015052795, -0.017044067, -0.005203247, -0.013305664, 0.010253906, -...
http://arxiv.org/abs/2305.07457v2
Perturbation-based QE: An Explainable, Unsupervised Word-level Quality Estimation Method for Blackbox Machine Translation
Tu Anh Dinh, Jan Niehues
cs.CL, I.2.7
Quality Estimation (QE) is the task of predicting the quality of Machine Translation (MT) system output, without using any gold-standard translation references. State-of-the-art QE models are supervised: they require human-labeled quality of some MT system output on some datasets for training, making them domain-depend...
http://arxiv.org/pdf/2305.07457v2
2023-07-13T07:35:09
2023-05-12T13:10:57
{ "title": [ -0.02168274, 0.012023926, 0.0035896301, -0.023834229, -0.012397766, 0.003435135, -0.032073975, -0.025909424, 0.006958008, 0.041290283, -0.033935547, 0.013511658, 0.017959595, -0.024597168, -0.036865234, 0.0069885254, -0.012817383, 0....
http://arxiv.org/abs/2305.07455v1
Improving Cascaded Unsupervised Speech Translation with Denoising Back-translation
Yu-Kuan Fu, Liang-Hsuan Tseng, Jiatong Shi, Chen-An Li, Tsu-Yuan Hsu, Shinji Watanabe, Hung-yi Lee
cs.CL, cs.SD, eess.AS
Most of the speech translation models heavily rely on parallel data, which is hard to collect especially for low-resource languages. To tackle this issue, we propose to build a cascaded speech translation system without leveraging any kind of paired data. We use fully unpaired data to train our unsupervised systems and...
http://arxiv.org/pdf/2305.07455v1
2023-05-12T13:07:51
2023-05-12T13:07:51
{ "title": [ -0.013450623, 0.02218628, -0.020492554, -0.062683105, 0.008560181, 0.012123108, -0.0635376, -0.021636963, -0.007949829, 0.051330566, -0.011054993, 0.011817932, 0.03729248, -0.025894165, -0.038604736, 0.013465881, -0.0003926754, 0.017...
http://arxiv.org/abs/2305.07424v2
Instance Smoothed Contrastive Learning for Unsupervised Sentence Embedding
Hongliang He, Junlei Zhang, Zhenzhong Lan, Yue Zhang
cs.CL, cs.LG
Contrastive learning-based methods, such as unsup-SimCSE, have achieved state-of-the-art (SOTA) performances in learning unsupervised sentence embeddings. However, in previous studies, each embedding used for contrastive learning only derived from one sentence instance, and we call these embeddings instance-level embed...
http://arxiv.org/pdf/2305.07424v2
2023-05-18T08:31:51
2023-05-12T12:46:13
{ "title": [ 0.003408432, -0.00077199936, -0.020889282, -0.03829956, 0.004688263, -0.006137848, -0.025375366, -0.038848877, 0.036193848, -0.0042419434, -0.015014648, 0.013198853, -0.00077295303, -0.032928467, -0.011962891, -0.0066833496, -0.013259888...
http://arxiv.org/abs/2305.07406v1
Two-in-One: A Model Hijacking Attack Against Text Generation Models
Wai Man Si, Michael Backes, Yang Zhang, Ahmed Salem
cs.CR, cs.CL, cs.LG
Machine learning has progressed significantly in various applications ranging from face recognition to text generation. However, its success has been accompanied by different attacks. Recently a new attack has been proposed which raises both accountability and parasitic computing risks, namely the model hijacking attac...
http://arxiv.org/pdf/2305.07406v1
2023-05-12T12:13:27
2023-05-12T12:13:27
{ "title": [ -0.022338867, -0.024673462, -0.03567505, -0.02848816, -0.0049209595, -0.00055742264, -0.023590088, -0.012023926, -0.022125244, 0.040740967, 0.002603531, 0.026306152, 0.008560181, -0.020706177, -0.038726807, 0.0039367676, -0.02180481, ...
http://arxiv.org/abs/2305.07402v2
Knowledge Refinement via Interaction Between Search Engines and Large Language Models
Jiazhan Feng, Chongyang Tao, Xiubo Geng, Tao Shen, Can Xu, Guodong Long, Dongyan Zhao, Daxin Jiang
cs.CL, cs.IR
Information retrieval (IR) plays a crucial role in locating relevant resources from vast amounts of data, and its applications have evolved from traditional knowledge bases to modern search engines (SEs). The emergence of large language models (LLMs) has further revolutionized the IR field by enabling users to interact...
http://arxiv.org/pdf/2305.07402v2
2023-05-21T14:58:50
2023-05-12T11:58:15
{ "title": [ -0.019363403, -0.014961243, -0.034851074, -0.07080078, -0.010894775, 0.0017356873, -0.047576904, -0.007843018, 0.011894226, 0.0317688, -0.027404785, -0.008796692, 0.001625061, -0.010002136, -0.051727295, -0.009773254, 0.0019283295, 0...
http://arxiv.org/abs/2305.07393v2
Prompt Learning to Mitigate Catastrophic Forgetting in Cross-lingual Transfer for Open-domain Dialogue Generation
Lei Liu, Jimmy Xiangji Huang
cs.CL, I.2.7
Dialogue systems for non-English languages have long been under-explored. In this paper, we take the first step to investigate few-shot cross-lingual transfer learning (FS-XLT) and multitask learning (MTL) in the context of open-domain dialogue generation for non-English languages with limited data. We observed catastr...
http://arxiv.org/pdf/2305.07393v2
2023-05-29T19:04:00
2023-05-12T11:41:16
{ "title": [ 0.001455307, 0.0050697327, -0.03540039, -0.058685303, 0.005290985, -0.0029296875, -0.04550171, -0.008850098, -0.0102005005, 0.02998352, 0.011222839, 0.016189575, 0.010444641, -0.03262329, -0.013923645, 0.00081539154, -0.00793457, 0.0...
http://arxiv.org/abs/2305.07389v1
Investigating the Sensitivity of Automatic Speech Recognition Systems to Phonetic Variation in L2 Englishes
Emma O'Neill, Julie Carson-Berndsen
cs.CL, cs.SD, eess.AS
Automatic Speech Recognition (ASR) systems exhibit the best performance on speech that is similar to that on which it was trained. As such, underrepresented varieties including regional dialects, minority-speakers, and low-resource languages, see much higher word error rates (WERs) than those varieties seen as 'prestig...
http://arxiv.org/pdf/2305.07389v1
2023-05-12T11:29:13
2023-05-12T11:29:13
{ "title": [ -0.036376953, 0.010345459, -0.021743774, 0.014816284, -0.022415161, -0.009490967, -0.018981934, -0.012962341, -0.037231445, 0.049865723, -0.014884949, 0.005004883, -0.009346008, -0.025543213, -0.06756592, -0.023132324, -0.013664246, ...
http://arxiv.org/abs/2305.07378v1
Surfacing Biases in Large Language Models using Contrastive Input Decoding
Gal Yona, Or Honovich, Itay Laish, Roee Aharoni
cs.CL, cs.CY, cs.LG
Ensuring that large language models (LMs) are fair, robust and useful requires an understanding of how different modifications to their inputs impact the model's behaviour. In the context of open-text generation tasks, however, such an evaluation is not trivial. For example, when introducing a model with an input text ...
http://arxiv.org/pdf/2305.07378v1
2023-05-12T11:09:49
2023-05-12T11:09:49
{ "title": [ -0.013885498, -0.0076560974, -0.024002075, -0.039154053, -0.011962891, -0.012619019, -0.045959473, -0.016418457, 0.024520874, 0.036499023, -0.027618408, -0.011169434, 0.013687134, -0.017868042, -0.057006836, 0.0032920837, -0.025054932, ...
http://arxiv.org/abs/2305.07375v4
Is ChatGPT a Good Causal Reasoner? A Comprehensive Evaluation
Jinglong Gao, Xiao Ding, Bing Qin, Ting Liu
cs.CL, cs.AI
Causal reasoning ability is crucial for numerous NLP applications. Despite the impressive emerging ability of ChatGPT in various NLP tasks, it is unclear how well ChatGPT performs in causal reasoning. In this paper, we conduct the first comprehensive evaluation of the ChatGPT's causal reasoning capabilities. Experiment...
http://arxiv.org/pdf/2305.07375v4
2023-10-12T06:42:25
2023-05-12T10:54:13
{ "title": [ 0.010169983, -0.013763428, -0.020141602, -0.06378174, 0.0059127808, 0.004924774, -0.024230957, -0.0039596558, -0.029403687, 0.0016288757, 0.010612488, 0.012748718, -0.014099121, -0.035583496, 0.0032863617, 0.0028820038, -0.02973938, ...
http://arxiv.org/abs/2305.07374v1
Implications of Deep Circuits in Improving Quality of Quantum Question Answering
Pragya Katyayan, Nisheeth Joshi
cs.CL, cs.AI, quant-ph
Question Answering (QA) has proved to be an arduous challenge in the area of natural language processing (NLP) and artificial intelligence (AI). Many attempts have been made to develop complete solutions for QA as well as improving significant sub-modules of the QA systems to improve the overall performance through the...
http://arxiv.org/pdf/2305.07374v1
2023-05-12T10:52:13
2023-05-12T10:52:13
{ "title": [ 0.009536743, -0.01020813, -0.0053901672, -0.025802612, -0.0022144318, -0.020141602, 0.015350342, -0.014717102, 0.015777588, 0.033569336, 0.00007891655, -0.0042648315, 0.02368164, -0.01676941, -0.049438477, 0.02645874, -0.0009074211, ...
http://arxiv.org/abs/2305.07372v2
Interactive Text-to-SQL Generation via Editable Step-by-Step Explanations
Yuan Tian, Zheng Zhang, Zheng Ning, Toby Jia-Jun Li, Jonathan K. Kummerfeld, Tianyi Zhang
cs.DB, cs.CL, I.2.7
Relational databases play an important role in this Big Data era. However, it is challenging for non-experts to fully unleash the analytical power of relational databases, since they are not familiar with database languages such as SQL. Many techniques have been proposed to automatically generate SQL from natural langu...
http://arxiv.org/pdf/2305.07372v2
2023-06-14T03:33:03
2023-05-12T10:45:29
{ "title": [ -0.014511108, -0.01109314, -0.04196167, -0.0859375, 0.0009813309, 0.017547607, -0.049316406, 0.020004272, 0.013671875, 0.03640747, -0.039794922, 0.005455017, -0.0063667297, -0.009681702, -0.034423828, -0.02116394, 0.010940552, 0.0041...
http://arxiv.org/abs/2305.07365v1
Towards Transliteration between Sindhi Scripts from Devanagari to Perso-Arabic
Shivani Singh Rathore, Bharti Nathani, Nisheeth Joshi, Pragya Katyayan, Chander Prakash Dadlani
cs.CL, cs.AI
In this paper, we have shown a script conversion (transliteration) technique that converts Sindhi text in the Devanagari script to the Perso-Arabic script. We showed this by incorporating a hybrid approach where some part of the text is converted using a rule base and in case an ambiguity arises then a probabilistic mo...
http://arxiv.org/pdf/2305.07365v1
2023-05-12T10:29:37
2023-05-12T10:29:37
{ "title": [ -0.030548096, -0.020248413, -0.033691406, 0.014839172, -0.059783936, -0.009689331, -0.024841309, -0.008354187, -0.0038700104, 0.025299072, -0.040374756, -0.0149383545, 0.019470215, -0.04421997, 0.007671356, -0.0236969, 0.022399902, 0...
http://arxiv.org/abs/2305.07360v1
Improving the Quality of Neural Machine Translation Through Proper Translation of Name Entities
Radhika Sharma, Pragya Katyayan, Nisheeth Joshi
cs.CL, cs.AI
In this paper, we have shown a method of improving the quality of neural machine translation by translating/transliterating name entities as a preprocessing step. Through experiments we have shown the performance gain of our system. For evaluation we considered three types of name entities viz person names, location na...
http://arxiv.org/pdf/2305.07360v1
2023-05-12T10:20:13
2023-05-12T10:20:13
{ "title": [ -0.03137207, -0.0018205643, -0.025970459, -0.040283203, 0.005340576, 0.0019130707, -0.03353882, -0.012054443, -0.010185242, 0.04940796, 0.0014839172, 0.0061149597, 0.042144775, -0.022598267, -0.02822876, 0.03894043, -0.009376526, 0.0...
http://arxiv.org/abs/2305.07358v2
Towards Versatile and Efficient Visual Knowledge Integration into Pre-trained Language Models with Cross-Modal Adapters
Xinyun Zhang, Haochen Tan, Han Wu, Mingjie Zhan, Ding Liang, Bei Yu
cs.CL
Humans learn language via multi-modal knowledge. However, due to the text-only pre-training scheme, most existing pre-trained language models (PLMs) are hindered from the multi-modal information. To inject visual knowledge into PLMs, existing methods incorporate either the text or image encoder of vision-language mod...
http://arxiv.org/pdf/2305.07358v2
2023-08-28T11:07:56
2023-05-12T10:08:46
{ "title": [ -0.0008621216, -0.0035114288, -0.041625977, -0.06451416, -0.017471313, -0.018447876, -0.06439209, -0.010635376, 0.03564453, 0.030761719, -0.035064697, -0.006416321, 0.030395508, -0.012718201, -0.040527344, 0.0036029816, 0.01940918, 0...
http://arxiv.org/abs/2305.07355v2
ZARA: Improving Few-Shot Self-Rationalization for Small Language Models
Wei-Lin Chen, An-Zi Yen, Cheng-Kuang Wu, Hen-Hsen Huang, Hsin-Hsi Chen
cs.CL
Language models (LMs) that jointly generate end-task answers as well as free-text rationales are known as self-rationalization models. Recent works demonstrate great performance gain for self-rationalization by few-shot prompting LMs with rationale-augmented exemplars. However, the ability to benefit from explanations ...
http://arxiv.org/pdf/2305.07355v2
2023-10-23T15:04:20
2023-05-12T10:07:12
{ "title": [ -0.052001953, -0.019515991, -0.03778076, -0.049926758, -0.023376465, 0.022949219, -0.08294678, 0.0119018555, -0.00038552284, 0.032592773, 0.015113831, 0.009544373, -0.013313293, -0.029724121, -0.05493164, 0.0109939575, -0.0035381317, ...
http://arxiv.org/abs/2305.07341v1
Model-based Programming: Redefining the Atomic Unit of Programming for the Deep Learning Era
Meng Zheng
cs.LG, cs.CL, cs.SE
This paper introduces and explores a new programming paradigm, Model-based Programming, designed to address the challenges inherent in applying deep learning models to real-world applications. Despite recent significant successes of deep learning models across a range of tasks, their deployment in real business scenari...
http://arxiv.org/pdf/2305.07341v1
2023-05-12T09:38:11
2023-05-12T09:38:11
{ "title": [ -0.036987305, -0.02003479, -0.041992188, -0.045196533, -0.0067481995, -0.012184143, -0.03390503, -0.010856628, -0.020492554, 0.0284729, -0.0077934265, 0.015670776, -0.0047950745, -0.03314209, -0.027236938, 0.005542755, -0.0048103333, ...
http://arxiv.org/abs/2305.07340v1
MedGPTEval: A Dataset and Benchmark to Evaluate Responses of Large Language Models in Medicine
Jie Xu, Lu Lu, Sen Yang, Bilin Liang, Xinwei Peng, Jiali Pang, Jinru Ding, Xiaoming Shi, Lingrui Yang, Huan Song, Kang Li, Xin Sun, Shaoting Zhang
cs.CL
METHODS: First, a set of evaluation criteria is designed based on a comprehensive literature review. Second, existing candidate criteria are optimized for using a Delphi method by five experts in medicine and engineering. Third, three clinical experts design a set of medical datasets to interact with LLMs. Finally, ben...
http://arxiv.org/pdf/2305.07340v1
2023-05-12T09:37:13
2023-05-12T09:37:13
{ "title": [ -0.033111572, -0.005393982, -0.041900635, -0.05456543, 0.005897522, -0.0004954338, -0.06335449, -0.028381348, 0.0017814636, 0.031555176, 0.009124756, 0.0042266846, 0.0020961761, -0.03414917, -0.04058838, 0.015380859, 0.0019435883, 0....
http://arxiv.org/abs/2305.18304v1
Semantic-aware Digital Twin for Metaverse: A Comprehensive Review
Senthil Kumar Jagatheesaperumal, Zhaohui Yang, Qianqian Yang, Chongwen Huang, Wei Xu, Mohammad Shikh-Bahaei, Zhaoyang Zhang
cs.CY, cs.CL, cs.IR, cs.MM, A.1; H.5; I.6; J.7; F.4
To facilitate the deployment of digital twins in Metaverse, the paradigm with semantic awareness has been proposed as a means for enabling accurate and task-oriented information extraction with inherent intelligence. However, this framework requires all devices in the Metaverse environment to be directly linked with th...
http://arxiv.org/pdf/2305.18304v1
2023-05-12T09:19:30
2023-05-12T09:19:30
{ "title": [ -0.022979736, 0.0046958923, -0.06781006, -0.05770874, 0.016494751, -0.019546509, -0.029708862, 0.0057792664, -0.027313232, 0.06506348, 0.025985718, 0.013328552, 0.047729492, -0.030548096, -0.03137207, 0.034851074, 0.0052757263, 0.027...
http://arxiv.org/abs/2305.07310v1
Improving Zero-shot Multilingual Neural Machine Translation by Leveraging Cross-lingual Consistency Regularization
Pengzhi Gao, Liwen Zhang, Zhongjun He, Hua Wu, Haifeng Wang
cs.CL, cs.AI
The multilingual neural machine translation (NMT) model has a promising capability of zero-shot translation, where it could directly translate between language pairs unseen during training. For good transfer performance from supervised directions to zero-shot directions, the multilingual NMT model is expected to learn ...
http://arxiv.org/pdf/2305.07310v1
2023-05-12T08:32:18
2023-05-12T08:32:18
{ "title": [ -0.0146102905, -0.0013990402, -0.013053894, -0.062561035, -0.005760193, 0.012397766, -0.0524292, 0.0107040405, 0.0018558502, 0.05545044, 0.010360718, 0.00548172, 0.02003479, -0.019012451, -0.044281006, 0.033416748, -0.0028438568, 0.0...
http://arxiv.org/abs/2305.07303v1
Multi-Relational Hyperbolic Word Embeddings from Natural Language Definitions
Marco Valentino, Danilo S. Carvalho, André Freitas
cs.CL, cs.LG
Neural-based word embeddings using solely distributional information have consistently produced useful meaning representations for downstream tasks. However, existing approaches often result in representations that are hard to interpret and control. Natural language definitions, on the other side, possess a recursive, ...
http://arxiv.org/pdf/2305.07303v1
2023-05-12T08:16:06
2023-05-12T08:16:06
{ "title": [ -0.006046295, -0.0066337585, -0.025421143, -0.061920166, 0.004245758, 0.0045700073, -0.040039062, -0.0040893555, 0.012359619, 0.014930725, -0.02330017, 0.018814087, -0.016342163, -0.005218506, -0.02923584, -0.0061454773, 0.0028095245, ...
http://arxiv.org/abs/2305.07289v1
RepCL: Exploring Effective Representation for Continual Text Classification
Yifan Song, Peiyi Wang, Dawei Zhu, Tianyu Liu, Zhifang Sui, Sujian Li
cs.CL
Continual learning (CL) aims to constantly learn new knowledge over time while avoiding catastrophic forgetting on old tasks. In this work, we focus on continual text classification under the class-incremental setting. Recent CL studies find that the representations learned in one task may not be effective for other ta...
http://arxiv.org/pdf/2305.07289v1
2023-05-12T07:32:00
2023-05-12T07:32:00
{ "title": [ 0.0043945312, 0.007030487, -0.041809082, -0.05999756, 0.02015686, -0.013076782, -0.016799927, -0.00970459, 0.03475952, 0.028717041, -0.024887085, 0.015029907, -0.0014972687, -0.007537842, -0.05368042, 0.008712769, -0.021133423, 0.076...
http://arxiv.org/abs/2305.07288v1
Open-WikiTable: Dataset for Open Domain Question Answering with Complex Reasoning over Table
Sunjun Kweon, Yeonsu Kwon, Seonhee Cho, Yohan Jo, Edward Choi
cs.CL
Despite recent interest in open domain question answering (ODQA) over tables, many studies still rely on datasets that are not truly optimal for the task with respect to utilizing structural nature of table. These datasets assume answers reside as a single cell value and do not necessitate exploring over multiple cells...
http://arxiv.org/pdf/2305.07288v1
2023-05-12T07:24:16
2023-05-12T07:24:16
{ "title": [ -0.015960693, -0.0040893555, -0.051574707, -0.07446289, 0.008956909, 0.0032997131, -0.042236328, 0.0051231384, 0.010925293, 0.027389526, -0.019439697, 0.010276794, -0.0010843277, -0.004623413, -0.04699707, -0.0021438599, 0.00012177229, ...
http://arxiv.org/abs/2305.07280v1
Harvesting Event Schemas from Large Language Models
Jialong Tang, Hongyu Lin, Zhuoqun Li, Yaojie Lu, Xianpei Han, Le Sun
cs.CL
Event schema provides a conceptual, structural and formal language to represent events and model the world event knowledge. Unfortunately, it is challenging to automatically induce high-quality and high-coverage event schemas due to the open nature of real-world events, the diversity of event expressions, and the spars...
http://arxiv.org/pdf/2305.07280v1
2023-05-12T06:51:05
2023-05-12T06:51:05
{ "title": [ -0.02368164, -0.021896362, -0.047210693, -0.12219238, 0.013252258, 0.017349243, -0.089782715, 0.02293396, 0.026977539, 0.020431519, 0.0064048767, -0.029449463, 0.02154541, -0.033416748, -0.031677246, 0.023544312, -0.0079956055, 0.018...
http://arxiv.org/abs/2305.07266v1
Gaussian Prior Reinforcement Learning for Nested Named Entity Recognition
Yawen Yang, Xuming Hu, Fukun Ma, Shu'ang Li, Aiwei Liu, Lijie Wen, Philip S. Yu
cs.CL, cs.AI
Named Entity Recognition (NER) is a well and widely studied task in natural language processing. Recently, the nested NER has attracted more attention since its practicality and difficulty. Existing works for nested NER ignore the recognition order and boundary position relation of nested entities. To address these iss...
http://arxiv.org/pdf/2305.07266v1
2023-05-12T05:55:34
2023-05-12T05:55:34
{ "title": [ 0.0016479492, 0.0074806213, -0.033294678, -0.06185913, 0.018203735, -0.016525269, -0.024002075, -0.0056877136, -0.030456543, 0.015686035, -0.042999268, 0.021438599, -0.0287323, -0.006046295, -0.014312744, -0.03677368, -0.009239197, 0...
http://arxiv.org/abs/2305.10438v1
IMAGINATOR: Pre-Trained Image+Text Joint Embeddings using Word-Level Grounding of Images
Varuna Krishna, S Suryavardan, Shreyash Mishra, Sathyanarayanan Ramamoorthy, Parth Patwa, Megha Chakraborty, Aman Chadha, Amitava Das, Amit Sheth
cs.CL, cs.AI, cs.CV, cs.MM
Word embeddings, i.e., semantically meaningful vector representation of words, are largely influenced by the distributional hypothesis "You shall know a word by the company it keeps" (Harris, 1954), whereas modern prediction-based neural network embeddings rely on design choices and hyperparameter optimization. Word em...
http://arxiv.org/pdf/2305.10438v1
2023-05-12T05:34:52
2023-05-12T05:34:52
{ "title": [ -0.0018644333, 0.015991211, -0.028366089, -0.036193848, 0.0143966675, -0.012283325, -0.019638062, -0.04360962, 0.001906395, 0.026123047, -0.034362793, 0.027938843, -0.0026245117, -0.011993408, -0.016967773, -0.0053482056, -0.0025863647, ...
http://arxiv.org/abs/2305.07243v2
Better speech synthesis through scaling
James Betker
cs.SD, cs.CL, eess.AS
In recent years, the field of image generation has been revolutionized by the application of autoregressive transformers and DDPMs. These approaches model the process of image generation as a step-wise probabilistic processes and leverage large amounts of compute and data to learn the image distribution. This methodolo...
http://arxiv.org/pdf/2305.07243v2
2023-05-23T21:41:54
2023-05-12T04:19:49
{ "title": [ -0.0010194778, -0.013931274, -0.05532837, -0.07940674, -0.039733887, -0.014160156, -0.058044434, 0.043304443, -0.030517578, 0.052886963, -0.038208008, 0.0065307617, 0.014907837, -0.039031982, 0.011726379, -0.0077667236, 0.015914917, ...
http://arxiv.org/abs/2305.07230v2
When Giant Language Brains Just Aren't Enough! Domain Pizzazz with Knowledge Sparkle Dust
Minh-Tien Nguyen, Duy-Hung Nguyen, Shahab Sabahi, Hung Le, Jeff Yang, Hajime Hotta
cs.CL
Large language models (LLMs) have significantly advanced the field of natural language processing, with GPT models at the forefront. While their remarkable performance spans a range of tasks, adapting LLMs for real-world business scenarios still poses challenges warranting further investigation. This paper presents an ...
http://arxiv.org/pdf/2305.07230v2
2023-07-08T08:13:06
2023-05-12T03:49:59
{ "title": [ -0.009483337, -0.034423828, -0.057861328, -0.041107178, -0.028396606, -0.02482605, -0.012962341, 0.022003174, 0.004322052, 0.018844604, -0.06149292, 0.011703491, -0.011001587, -0.023971558, -0.030166626, -0.045898438, 0.019958496, -0...
http://arxiv.org/abs/2305.07224v4
Asymmetric feature interaction for interpreting model predictions
Xiaolei Lu, Jianghong Ma, Haode Zhang
cs.CL
In natural language processing (NLP), deep neural networks (DNNs) could model complex interactions between context and have achieved impressive results on a range of NLP tasks. Prior works on feature interaction attribution mainly focus on studying symmetric interaction that only explains the additional influence of a ...
http://arxiv.org/pdf/2305.07224v4
2023-10-26T02:32:54
2023-05-12T03:31:24
{ "title": [ -0.04296875, -0.008384705, -0.02519226, -0.054229736, -0.012466431, 0.009109497, -0.055389404, -0.024810791, -0.007183075, 0.02053833, 0.00349617, -0.04949951, 0.0023670197, -0.039123535, -0.05871582, -0.009033203, 0.017486572, 0.039...
http://arxiv.org/abs/2305.07677v2
Masked Audio Text Encoders are Effective Multi-Modal Rescorers
Jinglun Cai, Monica Sunkara, Xilai Li, Anshu Bhatia, Xiao Pan, Sravan Bodapati
cs.SD, cs.CL, cs.LG
Masked Language Models (MLMs) have proven to be effective for second-pass rescoring in Automatic Speech Recognition (ASR) systems. In this work, we propose Masked Audio Text Encoder (MATE), a multi-modal masked language model rescorer which incorporates acoustic representations into the input space of MLM. We adopt con...
http://arxiv.org/pdf/2305.07677v2
2023-05-24T18:18:00
2023-05-11T22:44:43
{ "title": [ -0.004825592, -0.0010251999, -0.020065308, -0.006336212, -0.04031372, -0.022857666, -0.031158447, -0.015960693, 0.0009279251, 0.07543945, 0.021820068, -0.010475159, 0.08148193, -0.0473938, 0.0001411438, 0.039520264, -0.027145386, 0.0...
http://arxiv.org/abs/2305.07167v1
OneCAD: One Classifier for All image Datasets using multimodal learning
Shakti N. Wadekar, Eugenio Culurciello
cs.CV, cs.CL, cs.LG, eess.IV
Vision-Transformers (ViTs) and Convolutional neural networks (CNNs) are widely used Deep Neural Networks (DNNs) for classification task. These model architectures are dependent on the number of classes in the dataset it was trained on. Any change in number of classes leads to change (partial or full) in the model's arc...
http://arxiv.org/pdf/2305.07167v1
2023-05-11T22:40:47
2023-05-11T22:40:47
{ "title": [ -0.0287323, 0.0010929108, -0.05569458, -0.04446411, 0.010658264, -0.016235352, -0.022994995, -0.006023407, 0.014663696, 0.01436615, -0.013084412, 0.0073127747, -0.008903503, -0.030838013, -0.026123047, 0.0067634583, 0.00026845932, 0....
http://arxiv.org/abs/2305.07157v1
Exploring Zero and Few-shot Techniques for Intent Classification
Soham Parikh, Quaizar Vohra, Prashil Tumbade, Mitul Tiwari
cs.CL, cs.AI
Conversational NLU providers often need to scale to thousands of intent-classification models where new customers often face the cold-start problem. Scaling to so many customers puts a constraint on storage space as well. In this paper, we explore four different zero and few-shot intent classification approaches with t...
http://arxiv.org/pdf/2305.07157v1
2023-05-11T22:07:27
2023-05-11T22:07:27
{ "title": [ 0.013473511, 0.0054512024, -0.042877197, -0.02017212, 0.012535095, -0.035339355, 0.011741638, -0.015449524, 0.020935059, 0.01763916, -0.011962891, -0.00793457, -0.0132369995, -0.035095215, -0.04095459, -0.0149383545, -0.010658264, 0....
http://arxiv.org/abs/2305.07151v1
Overinformative Question Answering by Humans and Machines
Polina Tsvilodub, Michael Franke, Robert D. Hawkins, Noah D. Goodman
cs.CL
When faced with a polar question, speakers often provide overinformative answers going beyond a simple "yes" or "no". But what principles guide the selection of additional information? In this paper, we provide experimental evidence from two studies suggesting that overinformativeness in human answering is driven by co...
http://arxiv.org/pdf/2305.07151v1
2023-05-11T21:41:41
2023-05-11T21:41:41
{ "title": [ 0.0126953125, 0.006614685, -0.017150879, -0.04537964, 0.015136719, -0.013969421, 0.0011663437, -0.01574707, 0.0146102905, 0.028396606, -0.026626587, -0.0020885468, -0.008560181, -0.0040779114, -0.053985596, 0.0026474, 0.0030078888, 0...
http://arxiv.org/abs/2305.10436v1
SmartPhone: Exploring Keyword Mnemonic with Auto-generated Verbal and Visual Cues
Jaewook Lee, Andrew Lan
cs.CL
In second language vocabulary learning, existing works have primarily focused on either the learning interface or scheduling personalized retrieval practices to maximize memory retention. However, the learning content, i.e., the information presented on flashcards, has mostly remained constant. Keyword mnemonic is a no...
http://arxiv.org/pdf/2305.10436v1
2023-05-11T20:58:10
2023-05-11T20:58:10
{ "title": [ -0.022354126, -0.010063171, -0.021392822, -0.017333984, -0.03878784, -0.008621216, -0.020629883, -0.0078048706, -0.006462097, 0.036468506, -0.023803711, 0.016723633, 0.01335907, -0.011810303, -0.026657104, 0.0033226013, -0.0051498413, ...
http://arxiv.org/abs/2305.10435v2
Generative Pre-trained Transformer: A Comprehensive Review on Enabling Technologies, Potential Applications, Emerging Challenges, and Future Directions
Gokul Yenduri, Ramalingam M, Chemmalar Selvi G, Supriya Y, Gautam Srivastava, Praveen Kumar Reddy Maddikunta, Deepti Raj G, Rutvij H Jhaveri, Prabadevi B, Weizheng Wang, Athanasios V. Vasilakos, Thippa Reddy Gadekallu
cs.CL, cs.AI
The Generative Pre-trained Transformer (GPT) represents a notable breakthrough in the domain of natural language processing, which is propelling us toward the development of machines that can understand and communicate using language in a manner that closely resembles that of humans. GPT is based on the transformer arc...
http://arxiv.org/pdf/2305.10435v2
2023-05-21T10:12:02
2023-05-11T19:20:38
{ "title": [ 0.027175903, 0.01260376, -0.04046631, -0.08709717, 0.031433105, -0.012931824, -0.037231445, -0.026779175, -0.013435364, 0.032562256, -0.005332947, 0.022644043, 0.03845215, -0.04397583, 0.0053367615, 0.028656006, 0.01512146, 0.0414428...
http://arxiv.org/abs/2305.07095v1
Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationales
Brihi Joshi, Ziyi Liu, Sahana Ramnath, Aaron Chan, Zhewei Tong, Shaoliang Nie, Qifan Wang, Yejin Choi, Xiang Ren
cs.CL, cs.AI, cs.LG
Among the remarkable emergent capabilities of large language models (LMs) is free-text rationalization; beyond a certain scale, large LMs are capable of generating seemingly useful rationalizations, which in turn, can dramatically enhance their performances on leaderboards. This phenomenon raises a question: can machin...
http://arxiv.org/pdf/2305.07095v1
2023-05-11T19:01:13
2023-05-11T19:01:13
{ "title": [ -0.02670288, -0.014694214, -0.023147583, -0.015075684, -0.027282715, -0.027648926, 0.0068244934, -0.027511597, -0.014137268, 0.028167725, -0.032592773, -0.0028438568, -0.0008049011, -0.009460449, -0.043670654, -0.013580322, -0.0004854202...
http://arxiv.org/abs/2305.07085v1
Enhancing Contrastive Learning with Noise-Guided Attack: Towards Continual Relation Extraction in the Wild
Ting Wu, Jingyi Liu, Rui Zheng, Qi Zhang, Tao Gui, Xuanjing Huang
cs.CL
The principle of continual relation extraction~(CRE) involves adapting to emerging novel relations while preserving od knowledge. While current endeavors in CRE succeed in preserving old knowledge, they tend to fail when exposed to contaminated data streams. We assume this is attributed to their reliance on an artifici...
http://arxiv.org/pdf/2305.07085v1
2023-05-11T18:48:18
2023-05-11T18:48:18
{ "title": [ 0.029388428, 0.009757996, -0.042022705, -0.061828613, 0.02406311, -0.01940918, -0.0037174225, -0.006893158, 0.035095215, 0.017532349, -0.033966064, 0.010025024, -0.013313293, -0.005504608, -0.055725098, -0.0026340485, -0.009025574, 0...
http://arxiv.org/abs/2305.07019v1
Musketeer (All for One, and One for All): A Generalist Vision-Language Model with Task Explanation Prompts
Zhaoyang Zhang, Yantao Shen, Kunyu Shi, Zhaowei Cai, Jun Fang, Siqi Deng, Hao Yang, Davide Modolo, Zhuowen Tu, Stefano Soatto
cs.CV, cs.AI, cs.CL
We present a sequence-to-sequence vision-language model whose parameters are jointly trained on all tasks (all for one) and fully shared among multiple tasks (one for all), resulting in a single model which we named Musketeer. The integration of knowledge across heterogeneous tasks is enabled by a novel feature called ...
http://arxiv.org/pdf/2305.07019v1
2023-05-11T17:57:49
2023-05-11T17:57:49
{ "title": [ -0.03338623, -0.026123047, -0.047821045, -0.04751587, -0.026931763, -0.022964478, -0.047851562, 0.024032593, 0.021347046, 0.002046585, -0.0041046143, -0.015396118, 0.033050537, -0.041900635, -0.0110321045, -0.001786232, 0.011772156, ...
http://arxiv.org/abs/2305.07016v1
A General-Purpose Multilingual Document Encoder
Onur Galoğlu, Robert Litschko, Goran Glavaš
cs.CL
Massively multilingual pretrained transformers (MMTs) have tremendously pushed the state of the art on multilingual NLP and cross-lingual transfer of NLP models in particular. While a large body of work leveraged MMTs to mine parallel data and induce bilingual document embeddings, much less effort has been devoted to t...
http://arxiv.org/pdf/2305.07016v1
2023-05-11T17:55:45
2023-05-11T17:55:45
{ "title": [ -0.022750854, -0.017593384, -0.029129028, -0.05102539, -0.044158936, -0.010673523, -0.06652832, 0.02645874, 0.028045654, 0.06774902, -0.00075626373, -0.024139404, 0.052825928, -0.026885986, -0.016693115, 0.045074463, 0.029953003, 0.0...
http://arxiv.org/abs/2305.07011v4
Region-Aware Pretraining for Open-Vocabulary Object Detection with Vision Transformers
Dahun Kim, Anelia Angelova, Weicheng Kuo
cs.CV, cs.AI, cs.CL
We present Region-aware Open-vocabulary Vision Transformers (RO-ViT) - a contrastive image-text pretraining recipe to bridge the gap between image-level pretraining and open-vocabulary object detection. At the pretraining phase, we propose to randomly crop and resize regions of positional embeddings instead of using th...
http://arxiv.org/pdf/2305.07011v4
2023-08-28T07:29:03
2023-05-11T17:53:29
{ "title": [ 0.022888184, 0.016433716, -0.015991211, -0.0619812, 0.017364502, -0.00970459, -0.019760132, -0.033599854, -0.012382507, 0.017822266, -0.061462402, 0.033355713, -0.040222168, 0.0009937286, -0.038208008, -0.033813477, 0.01727295, 0.028...
http://arxiv.org/abs/2305.10434v2
Learning the Visualness of Text Using Large Vision-Language Models
Gaurav Verma, Ryan A. Rossi, Christopher Tensmeyer, Jiuxiang Gu, Ani Nenkova
cs.CL, cs.AI, cs.LG
Visual text evokes an image in a person's mind, while non-visual text fails to do so. A method to automatically detect visualness in text will enable text-to-image retrieval and generation models to augment text with relevant images. This is particularly challenging with long-form text as text-to-image generation and r...
http://arxiv.org/pdf/2305.10434v2
2023-10-22T19:06:01
2023-05-11T17:45:16
{ "title": [ -0.02859497, 0.00051021576, -0.022842407, -0.043151855, -0.0058135986, -0.008918762, -0.040649414, -0.012466431, 0.040008545, 0.031341553, -0.051849365, -0.006904602, 0.018844604, 0.008399963, -0.068725586, -0.011680603, 0.0016546249, ...
http://arxiv.org/abs/2305.07005v1
Subword Segmental Machine Translation: Unifying Segmentation and Target Sentence Generation
Francois Meyer, Jan Buys
cs.CL
Subword segmenters like BPE operate as a preprocessing step in neural machine translation and other (conditional) language models. They are applied to datasets before training, so translation or text generation quality relies on the quality of segmentations. We propose a departure from this paradigm, called subword seg...
http://arxiv.org/pdf/2305.07005v1
2023-05-11T17:44:29
2023-05-11T17:44:29
{ "title": [ -0.033569336, -0.01739502, -0.0007829666, -0.039093018, -0.020736694, 0.017105103, -0.039489746, -0.006164551, -0.005027771, 0.056854248, -0.0018358231, 0.016540527, 0.03778076, -0.01638794, -0.040039062, 0.031143188, -0.006149292, 0...
http://arxiv.org/abs/2305.07004v2
Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting
Haoyang Huang, Tianyi Tang, Dongdong Zhang, Wayne Xin Zhao, Ting Song, Yan Xia, Furu Wei
cs.CL
Large language models (LLMs) demonstrate impressive multilingual capability, but their performance varies substantially across different languages. In this work, we introduce a simple yet effective method, called cross-lingual-thought prompting (XLT), to systematically improve the multilingual capability of LLMs. Speci...
http://arxiv.org/pdf/2305.07004v2
2023-10-22T13:43:27
2023-05-11T17:44:17
{ "title": [ -0.03479004, -0.022491455, -0.030593872, -0.009544373, -0.02772522, -0.01184845, -0.052001953, -0.010902405, -0.013923645, 0.00983429, -0.039245605, 0.024902344, 0.00894928, -0.03048706, 0.0107803345, -0.010971069, -0.01309967, 0.038...
http://arxiv.org/abs/2305.07001v1
Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach
Junjie Zhang, Ruobing Xie, Yupeng Hou, Wayne Xin Zhao, Leyu Lin, Ji-Rong Wen
cs.IR, cs.CL
In the past decades, recommender systems have attracted much attention in both research and industry communities, and a large number of studies have been devoted to developing effective recommendation models. Basically speaking, these models mainly learn the underlying user preference from historical behavior data, and...
http://arxiv.org/pdf/2305.07001v1
2023-05-11T17:39:07
2023-05-11T17:39:07
{ "title": [ -0.00044989586, -0.0063476562, -0.027694702, -0.052001953, -0.0023078918, -0.005039215, -0.028549194, -0.008598328, -0.0029335022, 0.021469116, -0.024978638, 0.012687683, -0.009262085, -0.025161743, -0.017044067, -0.0006155968, -0.014968...
http://arxiv.org/abs/2305.06993v1
SMATCH++: Standardized and Extended Evaluation of Semantic Graphs
Juri Opitz
cs.CL, cs.AI
The Smatch metric is a popular method for evaluating graph distances, as is necessary, for instance, to assess the performance of semantic graph parsing systems. However, we observe some issues in the metric that jeopardize meaningful evaluation. E.g., opaque pre-processing choices can affect results, and current graph...
http://arxiv.org/pdf/2305.06993v1
2023-05-11T17:29:47
2023-05-11T17:29:47
{ "title": [ -0.021316528, -0.016418457, -0.058044434, -0.072631836, 0.02003479, 0.007701874, -0.042144775, -0.009963989, 0.02331543, 0.013999939, -0.018493652, 0.015701294, -0.0236969, -0.01828003, -0.033813477, 0.009536743, -0.007270813, 0.0428...
http://arxiv.org/abs/2305.06988v1
Self-Chained Image-Language Model for Video Localization and Question Answering
Shoubin Yu, Jaemin Cho, Prateek Yadav, Mohit Bansal
cs.CV, cs.AI, cs.CL, cs.LG
Recent studies have shown promising results on utilizing pre-trained image-language models for video question answering. While these image-language models can efficiently bootstrap the representation learning of video-language models, they typically concatenate uniformly sampled video frames as visual inputs without ex...
http://arxiv.org/pdf/2305.06988v1
2023-05-11T17:23:00
2023-05-11T17:23:00
{ "title": [ -0.015419006, 0.021469116, -0.019317627, -0.01979065, -0.0012426376, -0.0116119385, -0.0026836395, -0.028335571, 0.0146484375, 0.04864502, -0.021026611, -0.017944336, -0.0065689087, -0.017532349, -0.06518555, -0.004951477, 0.026367188, ...
http://arxiv.org/abs/2305.13317v1
A Novel Dataset Towards Extracting Virus-Host Interactions
Rasha Alshawi, Atriya Sen, Nathan S. Upham, Beckett Sterner
cs.CL
We describe a novel dataset for the automated recognition of named taxonomic and other entities relevant to the association of viruses with their hosts. We further describe some initial results using pre-trained models on the named-entity recognition (NER) task on this novel dataset. We propose that our dataset of manu...
http://arxiv.org/pdf/2305.13317v1
2023-05-11T17:20:49
2023-05-11T17:20:49
{ "title": [ 0.0030117035, 0.01626587, -0.059936523, -0.10803223, 0.032836914, -0.011245728, -0.054901123, 0.012382507, -0.0056266785, 0.00023818016, -0.02319336, 0.0027561188, -0.048065186, -0.014297485, -0.04724121, -0.008758545, 0.009338379, 0...
http://arxiv.org/abs/2305.06984v3
Evaluating Open-Domain Question Answering in the Era of Large Language Models
Ehsan Kamalloo, Nouha Dziri, Charles L. A. Clarke, Davood Rafiei
cs.CL
Lexical matching remains the de facto evaluation method for open-domain question answering (QA). Unfortunately, lexical matching fails completely when a plausible candidate answer does not appear in the list of gold answers, which is increasingly the case as we shift from extractive to generative models. The recent suc...
http://arxiv.org/pdf/2305.06984v3
2023-07-06T18:52:08
2023-05-11T17:14:33
{ "title": [ -0.00919342, -0.00598526, -0.026870728, -0.054473877, 0.0037574768, -0.0012083054, -0.022277832, -0.0023097992, -0.00274086, 0.026535034, -0.020263672, -0.0014543533, -0.0006952286, -0.0036087036, -0.052734375, -0.0008211136, -0.00933074...
http://arxiv.org/abs/2305.06983v2
Active Retrieval Augmented Generation
Zhengbao Jiang, Frank F. Xu, Luyu Gao, Zhiqing Sun, Qian Liu, Jane Dwivedi-Yu, Yiming Yang, Jamie Callan, Graham Neubig
cs.CL, cs.LG
Despite the remarkable ability of large language models (LMs) to comprehend and generate language, they have a tendency to hallucinate and create factually inaccurate output. Augmenting LMs by retrieving information from external knowledge resources is one promising solution. Most existing retrieval augmented LMs emplo...
http://arxiv.org/pdf/2305.06983v2
2023-10-22T00:11:13
2023-05-11T17:13:40
{ "title": [ 0.030654907, 0.0074005127, -0.024932861, -0.029800415, -0.016647339, -0.02848816, -0.015327454, -0.022537231, 0.025421143, 0.072631836, 0.0031471252, -0.0055656433, 0.027420044, -0.045684814, -0.016555786, 0.043884277, 0.023269653, -...
http://arxiv.org/abs/2305.06908v4
CoMoSpeech: One-Step Speech and Singing Voice Synthesis via Consistency Model
Zhen Ye, Wei Xue, Xu Tan, Jie Chen, Qifeng Liu, Yike Guo
cs.SD, cs.AI, cs.CL, cs.LG, cs.MM, eess.AS
Denoising diffusion probabilistic models (DDPMs) have shown promising performance for speech synthesis. However, a large number of iterative steps are required to achieve high sample quality, which restricts the inference speed. Maintaining sample quality while increasing sampling speed has become a challenging task. I...
http://arxiv.org/pdf/2305.06908v4
2023-10-29T14:12:08
2023-05-11T15:51:46
{ "title": [ -0.05065918, -0.024841309, -0.07739258, -0.06964111, -0.008583069, -0.00374794, -0.062408447, 0.04220581, -0.003320694, 0.06402588, -0.037597656, -0.002981186, 0.05255127, -0.01928711, -0.036621094, -0.0014314651, -0.027404785, 0.058...
http://arxiv.org/abs/2305.06897v1
AfriQA: Cross-lingual Open-Retrieval Question Answering for African Languages
Odunayo Ogundepo, Tajuddeen R. Gwadabe, Clara E. Rivera, Jonathan H. Clark, Sebastian Ruder, David Ifeoluwa Adelani, Bonaventure F. P. Dossou, Abdou Aziz DIOP, Claytone Sikasote, Gilles Hacheme, Happy Buzaaba, Ignatius Ezeani, Rooweither Mabuya, Salomey Osei, Chris Emezue, Albert Njoroge Kahira, Shamsuddeen H. Muhammad...
cs.CL, cs.AI, cs.IR
African languages have far less in-language content available digitally, making it challenging for question answering systems to satisfy the information needs of users. Cross-lingual open-retrieval question answering (XOR QA) systems -- those that retrieve answer content from other languages while serving people in the...
http://arxiv.org/pdf/2305.06897v1
2023-05-11T15:34:53
2023-05-11T15:34:53
{ "title": [ -0.02003479, -0.020309448, 0.0088272095, -0.054504395, -0.023864746, 0.029251099, -0.07116699, -0.020248413, 0.0077819824, 0.038848877, -0.001537323, 0.015716553, 0.027313232, -0.015617371, -0.041778564, 0.024291992, 0.0056114197, 0....
http://arxiv.org/abs/2305.06892v1
IUST_NLP at SemEval-2023 Task 10: Explainable Detecting Sexism with Transformers and Task-adaptive Pretraining
Hadiseh Mahmoudi
cs.CL
This paper describes our system on SemEval-2023 Task 10: Explainable Detection of Online Sexism (EDOS). This work aims to design an automatic system for detecting and classifying sexist content in online spaces. We propose a set of transformer-based pre-trained models with task-adaptive pretraining and ensemble learnin...
http://arxiv.org/pdf/2305.06892v1
2023-05-11T15:29:04
2023-05-11T15:29:04
{ "title": [ 0.023422241, -0.00598526, -0.034820557, -0.038848877, 0.007545471, -0.02835083, -0.007347107, -0.013832092, 0.0041770935, 0.032043457, -0.025787354, 0.011192322, -0.008857727, -0.0057640076, -0.0385437, -0.013374329, -0.024230957, 0....
http://arxiv.org/abs/2305.06849v2
WebCPM: Interactive Web Search for Chinese Long-form Question Answering
Yujia Qin, Zihan Cai, Dian Jin, Lan Yan, Shihao Liang, Kunlun Zhu, Yankai Lin, Xu Han, Ning Ding, Huadong Wang, Ruobing Xie, Fanchao Qi, Zhiyuan Liu, Maosong Sun, Jie Zhou
cs.CL, cs.AI, cs.IR
Long-form question answering (LFQA) aims at answering complex, open-ended questions with detailed, paragraph-length responses. The de facto paradigm of LFQA necessitates two procedures: information retrieval, which searches for relevant supporting facts, and information synthesis, which integrates these facts into a co...
http://arxiv.org/pdf/2305.06849v2
2023-05-23T13:15:10
2023-05-11T14:47:29
{ "title": [ -0.011444092, -0.008255005, -0.018249512, -0.037963867, -0.0041275024, -0.011001587, -0.0067253113, -0.007820129, 0.019546509, 0.030288696, -0.028244019, -0.011184692, 0.00566864, -0.022338867, -0.062561035, -0.011810303, -0.013092041, ...
http://arxiv.org/abs/2305.06841v1
Think Twice: Measuring the Efficiency of Eliminating Prediction Shortcuts of Question Answering Models
Lukáš Mikula, Michal Štefánik, Marek Petrovič, Petr Sojka
cs.CL, cs.AI
While the Large Language Models (LLMs) dominate a majority of language understanding tasks, previous work shows that some of these results are supported by modelling spurious correlations of training datasets. Authors commonly assess model robustness by evaluating their models on out-of-distribution (OOD) datasets of t...
http://arxiv.org/pdf/2305.06841v1
2023-05-11T14:35:00
2023-05-11T14:35:00
{ "title": [ -0.0075187683, -0.0005970001, -0.032928467, -0.033111572, 0.0079422, -0.019348145, -0.00052404404, -0.012611389, 0.024002075, 0.022476196, -0.025299072, 0.008766174, -0.002998352, -0.007698059, -0.059417725, 0.0009665489, -0.0013198853, ...
http://arxiv.org/abs/2306.01743v1
Abugida Normalizer and Parser for Unicode texts
Nazmuddoha Ansary, Quazi Adibur Rahman Adib, Tahsin Reasat, Sazia Mehnaz, Asif Shahriyar Sushmit, Ahmed Imtiaz Humayun, Mohammad Mamun Or Rashid, Farig Sadeque
cs.CL
This paper proposes two libraries to address common and uncommon issues with Unicode-based writing schemes for Indic languages. The first is a normalizer that corrects inconsistencies caused by the encoding scheme https://pypi.org/project/bnunicodenormalizer/ . The second is a grapheme parser for Abugida text https://p...
http://arxiv.org/pdf/2306.01743v1
2023-05-11T14:34:08
2023-05-11T14:34:08
{ "title": [ -0.019332886, 0.01576233, -0.049621582, -0.055603027, -0.002784729, 0.01499939, -0.057678223, 0.005756378, -0.017150879, 0.06854248, -0.01574707, -0.0012674332, 0.040496826, -0.034301758, -0.024536133, 0.0065956116, 0.0138168335, -0....
http://arxiv.org/abs/2305.06818v1
Towards a Computational Analysis of Suspense: Detecting Dangerous Situations
Albin Zehe, Julian Schröter, Andreas Hotho
cs.CL, I.2.7
Suspense is an important tool in storytelling to keep readers engaged and wanting to read more. However, it has so far not been studied extensively in Computational Literary Studies. In this paper, we focus on one of the elements authors can use to build up suspense: dangerous situations. We introduce a corpus of texts...
http://arxiv.org/pdf/2305.06818v1
2023-05-11T14:12:55
2023-05-11T14:12:55
{ "title": [ 0.0053634644, 0.007621765, -0.0055274963, -0.02960205, 0.01890564, -0.01550293, 0.007987976, -0.033081055, -0.021697998, 0.03451538, 0.0026283264, -0.008842468, 0.011024475, -0.025482178, -0.028305054, 0.0052108765, -0.01751709, 0.02...
http://arxiv.org/abs/2305.06817v1
THUIR@COLIEE 2023: More Parameters and Legal Knowledge for Legal Case Entailment
Haitao Li, Changyue Wang, Weihang Su, Yueyue Wu, Qingyao Ai, Yiqun Liu
cs.CL, cs.IR
This paper describes the approach of the THUIR team at the COLIEE 2023 Legal Case Entailment task. This task requires the participant to identify a specific paragraph from a given supporting case that entails the decision for the query case. We try traditional lexical matching methods and pre-trained language models wi...
http://arxiv.org/pdf/2305.06817v1
2023-05-11T14:11:48
2023-05-11T14:11:48
{ "title": [ -0.0067863464, -0.026138306, -0.020065308, -0.05291748, -0.0262146, 0.00096797943, -0.037078857, 0.022003174, -0.038879395, 0.037078857, -0.016174316, 0.027267456, 0.025909424, -0.046173096, 0.03656006, -0.004673004, 0.020706177, 0.0...
http://arxiv.org/abs/2305.06812v1
THUIR@COLIEE 2023: Incorporating Structural Knowledge into Pre-trained Language Models for Legal Case Retrieval
Haitao Li, Weihang Su, Changyue Wang, Yueyue Wu, Qingyao Ai, Yiqun Liu
cs.IR, cs.CL
Legal case retrieval techniques play an essential role in modern intelligent legal systems. As an annually well-known international competition, COLIEE is aiming to achieve the state-of-the-art retrieval model for legal texts. This paper summarizes the approach of the championship team THUIR in COLIEE 2023. To be speci...
http://arxiv.org/pdf/2305.06812v1
2023-05-11T14:08:53
2023-05-11T14:08:53
{ "title": [ -0.004962921, -0.028167725, -0.027633667, -0.0602417, -0.02557373, 0.011520386, -0.054229736, 0.0052604675, 0.025497437, 0.04260254, -0.053009033, -0.012649536, 0.039001465, 0.0012025833, -0.051086426, -0.0014734268, -0.00920105, 0.0...
http://arxiv.org/abs/2305.06801v1
Detecting Idiomatic Multiword Expressions in Clinical Terminology using Definition-Based Representation Learning
François Remy, Alfiya Khabibullina, Thomas Demeester
cs.CL
This paper shines a light on the potential of definition-based semantic models for detecting idiomatic and semi-idiomatic multiword expressions (MWEs) in clinical terminology. Our study focuses on biomedical entities defined in the UMLS ontology and aims to help prioritize the translation efforts of these entities. In ...
http://arxiv.org/pdf/2305.06801v1
2023-05-11T13:42:58
2023-05-11T13:42:58
{ "title": [ 0.0284729, 0.0070343018, -0.016845703, -0.060943604, 0.012702942, 0.00021886826, -0.027359009, -0.020263672, -0.00724411, 0.04647827, -0.013542175, 0.015365601, 0.019180298, -0.012504578, -0.038360596, -0.0065307617, -0.021530151, 0....
http://arxiv.org/abs/2305.06754v2
COCKATIEL: COntinuous Concept ranKed ATtribution with Interpretable ELements for explaining neural net classifiers on NLP tasks
Fanny Jourdan, Agustin Picard, Thomas Fel, Laurent Risser, Jean Michel Loubes, Nicholas Asher
cs.CL, stat.ML
Transformer architectures are complex and their use in NLP, while it has engendered many successes, makes their interpretability or explainability challenging. Recent debates have shown that attention maps and attribution methods are unreliable (Pruthi et al., 2019; Brunner et al., 2019). In this paper, we present some...
http://arxiv.org/pdf/2305.06754v2
2023-05-14T14:38:41
2023-05-11T12:22:20
{ "title": [ -0.0034618378, -0.014823914, -0.004688263, -0.019088745, -0.0077819824, -0.005897522, -0.0064315796, -0.026260376, 0.0061912537, -0.011512756, -0.04244995, -0.006538391, -0.022247314, -0.021224976, -0.010429382, -0.027770996, -0.03311157...
http://arxiv.org/abs/2305.06747v1
The First Parallel Corpora for Kurdish Sign Language
Zina Kamal, Hossein Hassani
cs.CL
Kurdish Sign Language (KuSL) is the natural language of the Kurdish Deaf people. We work on automatic translation between spoken Kurdish and KuSL. Sign languages evolve rapidly and follow grammatical rules that differ from spoken languages. Consequently,those differences should be considered during any translation. We ...
http://arxiv.org/pdf/2305.06747v1
2023-05-11T12:10:20
2023-05-11T12:10:20
{ "title": [ -0.0289917, 0.0012760162, -0.06933594, -0.04232788, -0.04510498, 0.004890442, -0.08117676, 0.015220642, 0.0020828247, 0.03945923, -0.05038452, -0.022338867, 0.047729492, -0.03387451, 0.0014276505, -0.02407837, -0.016586304, 0.0475769...
http://arxiv.org/abs/2305.10433v1
Toxicity Inspector: A Framework to Evaluate Ground Truth in Toxicity Detection Through Feedback
Huriyyah Althunayan, Rahaf Bahlas, Manar Alharbi, Lena Alsuwailem, Abeer Aldayel, Rehab ALahmadi
cs.CL, cs.SI
Toxic language is difficult to define, as it is not monolithic and has many variations in perceptions of toxicity. This challenge of detecting toxic language is increased by the highly contextual and subjectivity of its interpretation, which can degrade the reliability of datasets and negatively affect detection model ...
http://arxiv.org/pdf/2305.10433v1
2023-05-11T11:56:42
2023-05-11T11:56:42
{ "title": [ 0.007621765, -0.0033855438, -0.014793396, -0.07128906, 0.021148682, -0.0064888, -0.03100586, -0.028533936, -0.012649536, 0.008621216, -0.033691406, 0.023956299, -0.039978027, -0.014526367, -0.040924072, -0.0234375, -0.01701355, 0.047...
http://arxiv.org/abs/2305.06721v2
Advancing Neural Encoding of Portuguese with Transformer Albertina PT-*
João Rodrigues, Luís Gomes, João Silva, António Branco, Rodrigo Santos, Henrique Lopes Cardoso, Tomás Osório
cs.CL
To advance the neural encoding of Portuguese (PT), and a fortiori the technological preparation of this language for the digital age, we developed a Transformer-based foundation model that sets a new state of the art in this respect for two of its variants, namely European Portuguese from Portugal (PT-PT) and American ...
http://arxiv.org/pdf/2305.06721v2
2023-06-20T15:22:58
2023-05-11T10:56:20
{ "title": [ -0.02217102, -0.014274597, -0.019638062, -0.010215759, -0.05319214, 0.012573242, -0.06173706, -0.023864746, -0.0357666, 0.057281494, -0.0020618439, -0.020141602, 0.05014038, -0.047027588, 0.010772705, 0.004142761, 0.009819031, 0.0315...
http://arxiv.org/abs/2305.06683v1
Cost-efficient Crowdsourcing for Span-based Sequence Labeling: Worker Selection and Data Augmentation
Yujie Wang, Chao Huang, Liner Yang, Zhixuan Fang, Yaping Huang, Yang Liu, Erhong Yang
cs.CL, cs.AI
This paper introduces a novel worker selection algorithm, enhancing annotation quality and reducing costs in challenging span-based sequence labeling tasks in Natural Language Processing (NLP). Unlike previous studies targeting simpler tasks, this study contends with the complexities of label interdependencies in seque...
http://arxiv.org/pdf/2305.06683v1
2023-05-11T09:40:24
2023-05-11T09:40:24
{ "title": [ 0.006095886, -0.006504059, -0.04550171, -0.072143555, 0.005302429, -0.012428284, -0.03152466, 0.029281616, -0.0362854, 0.028411865, -0.049560547, 0.033843994, -0.017440796, 0.008041382, -0.018249512, -0.022201538, -0.00081062317, 0.0...
http://arxiv.org/abs/2305.06677v2
INGENIOUS: Using Informative Data Subsets for Efficient Pre-Training of Language Models
H S V N S Kowndinya Renduchintala, Krishnateja Killamsetty, Sumit Bhatia, Milan Aggarwal, Ganesh Ramakrishnan, Rishabh Iyer, Balaji Krishnamurthy
cs.CL, cs.AI, cs.LG
A salient characteristic of pre-trained language models (PTLMs) is a remarkable improvement in their generalization capability and emergence of new capabilities with increasing model capacity and pre-training dataset size. Consequently, we are witnessing the development of enormous models pushing the state-of-the-art. ...
http://arxiv.org/pdf/2305.06677v2
2023-10-19T19:55:20
2023-05-11T09:24:41
{ "title": [ -0.0146102905, -0.012138367, -0.0513916, -0.08087158, -0.0046195984, -0.013198853, -0.062469482, 0.007484436, -0.01586914, 0.025619507, -0.020019531, 0.028060913, -0.00009715557, -0.011711121, -0.027252197, -0.0016851425, -0.0014276505, ...
http://arxiv.org/abs/2305.06655v2
QURG: Question Rewriting Guided Context-Dependent Text-to-SQL Semantic Parsing
Linzheng Chai, Dongling Xiao, Jian Yang, Liqun Yang, Qian-Wen Zhang, Yunbo Cao, Zhoujun Li, Zhao Yan
cs.CL
Context-dependent Text-to-SQL aims to translate multi-turn natural language questions into SQL queries. Despite various methods have exploited context-dependence information implicitly for contextual SQL parsing, there are few attempts to explicitly address the dependencies between current question and question context...
http://arxiv.org/pdf/2305.06655v2
2023-05-16T06:22:47
2023-05-11T08:45:55
{ "title": [ -0.008384705, -0.010848999, -0.035888672, -0.08685303, 0.014434814, 0.01802063, -0.045959473, 0.015808105, 0.009124756, 0.046539307, -0.049713135, 0.02079773, 0.01512146, 0.010253906, -0.061950684, -0.015808105, -0.006801605, 0.01112...
http://arxiv.org/abs/2305.06647v1
PROM: A Phrase-level Copying Mechanism with Pre-training for Abstractive Summarization
Xinbei Ma, Yeyun Gong, Pengcheng He, Hai Zhao, Nan Duan
cs.CL
Based on the remarkable achievements of pre-trained language models in abstractive summarization, the copying mechanism has proved helpful by improving the factuality, stability, and overall performance. This work proposes PROM, a new PhRase-level cOpying Mechanism that enhances attention on n-grams, which can be appli...
http://arxiv.org/pdf/2305.06647v1
2023-05-11T08:29:05
2023-05-11T08:29:05
{ "title": [ 0.007091522, -0.011505127, -0.010238647, -0.08886719, -0.018722534, 0.014778137, -0.062469482, -0.014030457, 0.0020103455, 0.048065186, -0.012916565, 0.031921387, 0.0058555603, -0.0013942719, -0.039276123, 0.014915466, -0.00065517426, ...
http://arxiv.org/abs/2305.06626v3
When the Majority is Wrong: Modeling Annotator Disagreement for Subjective Tasks
Eve Fleisig, Rediet Abebe, Dan Klein
cs.CL, cs.AI
Though majority vote among annotators is typically used for ground truth labels in natural language processing, annotator disagreement in tasks such as hate speech detection may reflect differences in opinion across groups, not noise. Thus, a crucial problem in hate speech detection is determining whether a statement i...
http://arxiv.org/pdf/2305.06626v3
2023-05-24T08:57:08
2023-05-11T07:55:20
{ "title": [ -0.01373291, 0.0069274902, -0.016235352, -0.035491943, -0.0064888, -0.009887695, -0.030380249, -0.038330078, 0.037872314, 0.024276733, -0.028915405, -0.02444458, -0.0020275116, -0.015716553, -0.07373047, -0.019119263, -0.0069007874, ...
http://arxiv.org/abs/2305.06620v1
Improving Continual Relation Extraction by Distinguishing Analogous Semantics
Wenzheng Zhao, Yuanning Cui, Wei Hu
cs.CL
Continual relation extraction (RE) aims to learn constantly emerging relations while avoiding forgetting the learned relations. Existing works store a small number of typical samples to re-train the model for alleviating forgetting. However, repeatedly replaying these samples may cause the overfitting problem. We condu...
http://arxiv.org/pdf/2305.06620v1
2023-05-11T07:32:20
2023-05-11T07:32:20
{ "title": [ 0.012710571, 0.018127441, -0.053131104, -0.093933105, 0.039855957, -0.005832672, -0.040863037, -0.0031051636, 0.03579712, 0.029907227, -0.01977539, 0.0057411194, -0.0041236877, 0.00018680096, -0.058502197, 0.010131836, -0.014976501, ...
http://arxiv.org/abs/2305.06616v1
Serial Contrastive Knowledge Distillation for Continual Few-shot Relation Extraction
Xinyi Wang, Zitao Wang, Wei Hu
cs.CL
Continual few-shot relation extraction (RE) aims to continuously train a model for new relations with few labeled training data, of which the major challenges are the catastrophic forgetting of old relations and the overfitting caused by data sparsity. In this paper, we propose a new model, namely SCKD, to accomplish t...
http://arxiv.org/pdf/2305.06616v1
2023-05-11T07:25:47
2023-05-11T07:25:47
{ "title": [ 0.029022217, -0.006385803, -0.052490234, -0.10296631, 0.018325806, -0.009376526, -0.038085938, 0.017425537, 0.07110596, 0.013618469, -0.03100586, -0.015266418, -0.014564514, -0.025161743, -0.06213379, -0.00894165, 0.012702942, 0.0049...
http://arxiv.org/abs/2305.06615v1
Autocorrelations Decay in Texts and Applicability Limits of Language Models
Nikolay Mikhaylovskiy, Ilya Churilov
cs.CL, I.2.7
We show that the laws of autocorrelations decay in texts are closely related to applicability limits of language models. Using distributional semantics we empirically demonstrate that autocorrelations of words in texts decay according to a power law. We show that distributional semantics provides coherent autocorrelati...
http://arxiv.org/pdf/2305.06615v1
2023-05-11T07:23:01
2023-05-11T07:23:01
{ "title": [ -0.038757324, 0.0059890747, -0.030441284, -0.029052734, -0.007472992, 0.007827759, -0.040771484, -0.022857666, 0.003376007, 0.06323242, -0.02142334, 0.0007696152, 0.022964478, 0.0021648407, -0.07849121, 0.016281128, -0.011306763, 0.0...
http://arxiv.org/abs/2305.06599v3
Structured Chain-of-Thought Prompting for Code Generation
Jia Li, Ge Li, Yongmin Li, Zhi Jin
cs.SE, cs.CL
Large Language Models (LLMs) (e.g., ChatGPT) have shown impressive performance in code generation. LLMs take prompts as inputs, and Chain-of-Thought (CoT) prompting is the state-of-the-art prompting technique. CoT prompting asks LLMs first to generate CoTs (i.e., intermediate natural language reasoning steps) and then ...
http://arxiv.org/pdf/2305.06599v3
2023-09-07T11:39:07
2023-05-11T06:43:37
{ "title": [ -0.015449524, -0.032928467, -0.016098022, -0.057617188, -0.02772522, 0.010757446, -0.039764404, 0.01574707, 0.0184021, -0.0064086914, -0.049621582, 0.0053863525, -0.025115967, -0.021484375, -0.01777649, -0.01940918, -0.0005645752, -0...
http://arxiv.org/abs/2305.06595v3
BanglaBook: A Large-scale Bangla Dataset for Sentiment Analysis from Book Reviews
Mohsinul Kabir, Obayed Bin Mahfuz, Syed Rifat Raiyan, Hasan Mahmud, Md Kamrul Hasan
cs.CL
The analysis of consumer sentiment, as expressed through reviews, can provide a wealth of insight regarding the quality of a product. While the study of sentiment analysis has been widely explored in many popular languages, relatively less attention has been given to the Bangla language, mostly due to a lack of relevan...
http://arxiv.org/pdf/2305.06595v3
2023-06-08T08:57:41
2023-05-11T06:27:38
{ "title": [ -0.024765015, -0.018737793, -0.03414917, -0.036376953, -0.0063056946, -0.010345459, -0.02456665, -0.010093689, 0.015365601, 0.028076172, -0.021835327, 0.011123657, -0.018615723, -0.008522034, -0.05682373, -0.015617371, -0.017196655, ...
http://arxiv.org/abs/2305.06590v2
FactKG: Fact Verification via Reasoning on Knowledge Graphs
Jiho Kim, Sungjin Park, Yeonsu Kwon, Yohan Jo, James Thorne, Edward Choi
cs.CL, cs.AI
In real world applications, knowledge graphs (KG) are widely used in various domains (e.g. medical applications and dialogue agents). However, for fact verification, KGs have not been adequately utilized as a knowledge source. KGs can be a valuable knowledge source in fact verification due to their reliability and broa...
http://arxiv.org/pdf/2305.06590v2
2023-05-19T03:50:52
2023-05-11T06:08:49
{ "title": [ -0.027267456, -0.0138168335, -0.060394287, -0.10821533, 0.010345459, 0.019561768, -0.073913574, 0.026184082, 0.0012073517, 0.01586914, -0.005203247, 0.008911133, -0.005138397, -0.023712158, -0.029708862, -0.004119873, -0.027648926, 0...
http://arxiv.org/abs/2305.06586v2
SemEval-2023 Task 2: Fine-grained Multilingual Named Entity Recognition (MultiCoNER 2)
Besnik Fetahu, Sudipta Kar, Zhiyu Chen, Oleg Rokhlenko, Shervin Malmasi
cs.CL, cs.AI
We present the findings of SemEval-2023 Task 2 on Fine-grained Multilingual Named Entity Recognition (MultiCoNER 2). Divided into 13 tracks, the task focused on methods to identify complex fine-grained named entities (like WRITTENWORK, VEHICLE, MUSICALGRP) across 12 languages, in both monolingual and multilingual scena...
http://arxiv.org/pdf/2305.06586v2
2023-05-25T17:54:06
2023-05-11T05:56:08
{ "title": [ -0.020736694, -0.005382538, -0.05230713, -0.05984497, 0.021377563, 0.0027809143, -0.050811768, 0.001663208, -0.0051727295, 0.042541504, -0.008331299, 0.020645142, 0.018875122, -0.011161804, -0.019363403, 0.024215698, -0.011497498, 0....
http://arxiv.org/abs/2305.06575v3
Chain-of-Dictionary Prompting Elicits Translation in Large Language Models
Hongyuan Lu, Haoyang Huang, Dongdong Zhang, Haoran Yang, Wai Lam, Furu Wei
cs.CL
Large language models (LLMs) have shown surprisingly good performance in multilingual neural machine translation (MNMT) even when trained without parallel data. Yet, despite the fact that the amount of training data is gigantic, they still struggle with translating rare words, particularly for low-resource languages. E...
http://arxiv.org/pdf/2305.06575v3
2023-05-24T05:01:52
2023-05-11T05:19:47
{ "title": [ -0.042022705, -0.012496948, -0.01928711, -0.039001465, -0.025512695, 0.0012331009, -0.04916382, -0.0035324097, -0.00048565865, 0.029373169, -0.0418396, 0.011383057, 0.021957397, -0.01776123, -0.04336548, 0.004425049, 0.012207031, -0....
http://arxiv.org/abs/2305.06574v1
A Fused Gromov-Wasserstein Framework for Unsupervised Knowledge Graph Entity Alignment
Jianheng Tang, Kangfei Zhao, Jia Li
cs.CL, cs.AI
Entity alignment is the task of identifying corresponding entities across different knowledge graphs (KGs). Although recent embedding-based entity alignment methods have shown significant advancements, they still struggle to fully utilize KG structural information. In this paper, we introduce FGWEA, an unsupervised ent...
http://arxiv.org/pdf/2305.06574v1
2023-05-11T05:17:54
2023-05-11T05:17:54
{ "title": [ -0.053649902, 0.019515991, -0.08532715, -0.050079346, 0.014579773, -0.0020523071, -0.064208984, 0.0016450882, 0.013404846, 0.046447754, 0.0038852692, 0.010261536, 0.027862549, -0.0069847107, -0.033477783, 0.023620605, -0.00024294853, ...
http://arxiv.org/abs/2305.06569v6
How to Index Item IDs for Recommendation Foundation Models
Wenyue Hua, Shuyuan Xu, Yingqiang Ge, Yongfeng Zhang
cs.IR, cs.AI, cs.CL, cs.LG
Recommendation foundation model utilizes large language models (LLM) for recommendation by converting recommendation tasks into natural language tasks. It enables generative recommendation which directly generates the item(s) to recommend rather than calculating a ranking score for each and every candidate item as in t...
http://arxiv.org/pdf/2305.06569v6
2023-09-26T01:40:11
2023-05-11T05:02:37
{ "title": [ -0.004722595, -0.01058197, -0.017745972, 0.004714966, -0.02949524, -0.013175964, -0.016235352, -0.01537323, -0.020401001, 0.039611816, -0.046020508, -0.0034999847, 0.008308411, -0.013000488, -0.039886475, -0.039978027, 0.010383606, 0...
http://arxiv.org/abs/2305.06566v4
ONCE: Boosting Content-based Recommendation with Both Open- and Closed-source Large Language Models
Qijiong Liu, Nuo Chen, Tetsuya Sakai, Xiao-Ming Wu
cs.IR, cs.CL
Personalized content-based recommender systems have become indispensable tools for users to navigate through the vast amount of content available on platforms like daily news websites and book recommendation services. However, existing recommenders face significant challenges in understanding the content of items. Larg...
http://arxiv.org/pdf/2305.06566v4
2023-08-31T13:43:43
2023-05-11T04:51:21
{ "title": [ -0.0030059814, -0.007419586, -0.02986145, -0.046173096, 0.014640808, -0.0015039444, -0.03677368, -0.018295288, -0.018753052, 0.04019165, -0.030410767, 0.034973145, 0.003572464, -0.0015974045, -0.02331543, 0.0015869141, -0.008178711, ...
http://arxiv.org/abs/2305.06557v1
Long-Tailed Question Answering in an Open World
Yi Dai, Hao Lang, Yinhe Zheng, Fei Huang, Yongbin Li
cs.CL, cs.AI, cs.LG
Real-world data often have an open long-tailed distribution, and building a unified QA model supporting various tasks is vital for practical QA applications. However, it is non-trivial to extend previous QA approaches since they either require access to seen tasks of adequate samples or do not explicitly model samples ...
http://arxiv.org/pdf/2305.06557v1
2023-05-11T04:28:58
2023-05-11T04:28:58
{ "title": [ 0.0050086975, 0.008468628, -0.012863159, -0.056121826, 0.012168884, -0.00057411194, -0.012535095, -0.0041122437, -0.0021266937, 0.011207581, -0.023269653, -0.0027599335, -0.04156494, -0.013046265, -0.041534424, -0.022872925, -0.003347396...
http://arxiv.org/abs/2305.06555v1
Domain Incremental Lifelong Learning in an Open World
Yi Dai, Hao Lang, Yinhe Zheng, Bowen Yu, Fei Huang, Yongbin Li
cs.CL, cs.AI, cs.LG
Lifelong learning (LL) is an important ability for NLP models to learn new tasks continuously. Architecture-based approaches are reported to be effective implementations for LL models. However, it is non-trivial to extend previous approaches to domain incremental LL scenarios since they either require access to task id...
http://arxiv.org/pdf/2305.06555v1
2023-05-11T04:19:08
2023-05-11T04:19:08
{ "title": [ 0.0096206665, 0.021636963, -0.061767578, -0.054870605, 0.0211792, -0.04547119, -0.000005185604, 0.0027427673, -0.028442383, 0.01727295, -0.019699097, 0.016311646, -0.03265381, -0.029144287, -0.0051956177, -0.013298035, -0.0034179688, ...
http://arxiv.org/abs/2305.06545v1
GeoGLUE: A GeoGraphic Language Understanding Evaluation Benchmark
Dongyang Li, Ruixue Ding, Qiang Zhang, Zheng Li, Boli Chen, Pengjun Xie, Yao Xu, Xin Li, Ning Guo, Fei Huang, Xiaofeng He
cs.CL, cs.AI
With a fast developing pace of geographic applications, automatable and intelligent models are essential to be designed to handle the large volume of information. However, few researchers focus on geographic natural language processing, and there has never been a benchmark to build a unified standard. In this work, we ...
http://arxiv.org/pdf/2305.06545v1
2023-05-11T03:21:56
2023-05-11T03:21:56
{ "title": [ -0.03479004, -0.02357483, -0.02418518, -0.046722412, 0.00028014183, 0.00061941147, -0.038208008, -0.00642395, 0.01084137, 0.020355225, -0.02104187, -0.0012979507, -0.0053520203, -0.01727295, -0.043792725, -0.0048217773, -0.00982666, ...
http://arxiv.org/abs/2305.06539v1
Semantic uncertainty guides the extension of conventions to new referents
Ron Eliav, Anya Ji, Yoav Artzi, Robert D. Hawkins
cs.CL
A long tradition of studies in psycholinguistics has examined the formation and generalization of ad hoc conventions in reference games, showing how newly acquired conventions for a given target transfer to new referential contexts. However, another axis of generalization remains understudied: how do conventions formed...
http://arxiv.org/pdf/2305.06539v1
2023-05-11T03:01:40
2023-05-11T03:01:40
{ "title": [ 0.02192688, 0.0057258606, -0.016571045, -0.029281616, -0.021255493, -0.013122559, -0.03692627, -0.041503906, -0.036743164, 0.03778076, -0.014022827, 0.034332275, 0.027023315, -0.04272461, 0.044006348, 0.0054473877, 0.015136719, 0.025...