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/2305.06174v2
Analysis of Climate Campaigns on Social Media using Bayesian Model Averaging
Tunazzina Islam, Ruqi Zhang, Dan Goldwasser
cs.CL, cs.AI, cs.CY, cs.LG, cs.SI
Climate change is the defining issue of our time, and we are at a defining moment. Various interest groups, social movement organizations, and individuals engage in collective action on this issue on social media. In addition, issue advocacy campaigns on social media often arise in response to ongoing societal concerns...
http://arxiv.org/pdf/2305.06174v2
2023-06-30T18:06:50
2023-05-06T16:43:29
{ "title": [ 0.02708435, 0.018798828, -0.028656006, -0.028518677, 0.02645874, -0.025405884, 0.011802673, -0.0099487305, -0.051483154, 0.03074646, -0.029724121, 0.046661377, -0.01525116, -0.01889038, 0.0006117821, -0.020446777, 0.0053138733, 0.037...
http://arxiv.org/abs/2305.04091v3
Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
Lei Wang, Wanyu Xu, Yihuai Lan, Zhiqiang Hu, Yunshi Lan, Roy Ka-Wei Lee, Ee-Peng Lim
cs.CL
Large language models (LLMs) have recently been shown to deliver impressive performance in various NLP tasks. To tackle multi-step reasoning tasks, few-shot chain-of-thought (CoT) prompting includes a few manually crafted step-by-step reasoning demonstrations which enable LLMs to explicitly generate reasoning steps and...
http://arxiv.org/pdf/2305.04091v3
2023-05-26T07:06:48
2023-05-06T16:34:37
{ "title": [ 0.0037879944, -0.03326416, -0.022338867, -0.06817627, -0.025497437, -0.006565094, -0.054504395, 0.0076560974, 0.0066833496, 0.0061569214, -0.016525269, 0.006511688, -0.0043182373, -0.028366089, -0.020523071, -0.0031147003, 0.00983429, ...
http://arxiv.org/abs/2305.04087v5
Self-Edit: Fault-Aware Code Editor for Code Generation
Kechi Zhang, Zhuo Li, Jia Li, Ge Li, Zhi Jin
cs.SE, cs.CL
Large language models (LLMs) have demonstrated an impressive ability to generate codes on competitive programming tasks. However, with limited sample numbers, LLMs still suffer from poor accuracy. Inspired by the process of human programming, we propose a generate-and-edit approach named Self-Edit that utilizes executi...
http://arxiv.org/pdf/2305.04087v5
2023-09-11T06:27:53
2023-05-06T16:12:19
{ "title": [ -0.024734497, -0.02178955, -0.040283203, -0.07574463, -0.0062217712, -0.0005259514, -0.038391113, 0.030044556, 0.000667572, 0.01550293, 0.01121521, -0.0124435425, -0.003944397, -0.047821045, -0.03894043, -0.00040555, -0.008583069, -0...
http://arxiv.org/abs/2305.04082v1
A Minimal Approach for Natural Language Action Space in Text-based Games
Dongwon Kelvin Ryu, Meng Fang, Shirui Pan, Gholamreza Haffari, Ehsan Shareghi
cs.LG, cs.CL
Text-based games (TGs) are language-based interactive environments for reinforcement learning. While language models (LMs) and knowledge graphs (KGs) are commonly used for handling large action space in TGs, it is unclear whether these techniques are necessary or overused. In this paper, we revisit the challenge of exp...
http://arxiv.org/pdf/2305.04082v1
2023-05-06T16:05:27
2023-05-06T16:05:27
{ "title": [ -0.02041626, -0.009735107, -0.016662598, -0.020996094, -0.00080776215, -0.008506775, -0.008049011, -0.009918213, 0.006275177, 0.031082153, -0.025161743, 0.017227173, -0.023010254, -0.003768921, -0.04434204, -0.014671326, -0.002840042, ...
http://arxiv.org/abs/2305.04076v2
SANTA: Separate Strategies for Inaccurate and Incomplete Annotation Noise in Distantly-Supervised Named Entity Recognition
Shuzheng Si, Zefan Cai, Shuang Zeng, Guoqiang Feng, Jiaxing Lin, Baobao Chang
cs.CL, cs.AI
Distantly-Supervised Named Entity Recognition effectively alleviates the burden of time-consuming and expensive annotation in the supervised setting. But the context-free matching process and the limited coverage of knowledge bases introduce inaccurate and incomplete annotation noise respectively. Previous studies eith...
http://arxiv.org/pdf/2305.04076v2
2023-07-28T06:51:56
2023-05-06T15:48:24
{ "title": [ -0.020568848, 0.004257202, -0.0340271, -0.062561035, 0.017532349, -0.005092621, -0.02998352, -0.00022506714, -0.00010317564, 0.012008667, -0.064086914, 0.035583496, -0.0009379387, 0.007522583, -0.012649536, -0.011985779, -0.010047913, ...
http://arxiv.org/abs/2305.04067v1
Reactive Perturbation Defocusing for Textual Adversarial Defense
Heng Yang, Ke Li
cs.CL
Recent studies have shown that large pre-trained language models are vulnerable to adversarial attacks. Existing methods attempt to reconstruct the adversarial examples. However, these methods usually have limited performance in defense against adversarial examples, while also negatively impacting the performance on na...
http://arxiv.org/pdf/2305.04067v1
2023-05-06T15:14:11
2023-05-06T15:14:11
{ "title": [ 0.0039253235, -0.0029678345, 0.005355835, -0.06628418, 0.0024814606, 0.018432617, -0.016952515, -0.0033435822, -0.009391785, 0.017211914, -0.023406982, 0.018600464, -0.016220093, -0.01890564, -0.02998352, -0.026977539, 0.0031986237, ...
http://arxiv.org/abs/2305.04049v1
Actively Discovering New Slots for Task-oriented Conversation
Yuxia Wu, Tianhao Dai, Zhedong Zheng, Lizi Liao
cs.CL
Existing task-oriented conversational search systems heavily rely on domain ontologies with pre-defined slots and candidate value sets. In practical applications, these prerequisites are hard to meet, due to the emerging new user requirements and ever-changing scenarios. To mitigate these issues for better interaction ...
http://arxiv.org/pdf/2305.04049v1
2023-05-06T13:33:33
2023-05-06T13:33:33
{ "title": [ 0.04043579, 0.0059547424, -0.021499634, -0.042022705, -0.017166138, -0.019042969, -0.015434265, -0.014266968, -0.019454956, 0.039916992, -0.0025634766, 0.009284973, 0.0087890625, -0.042816162, 0.012374878, -0.0030822754, 0.0069351196, ...
http://arxiv.org/abs/2305.04044v1
Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generation
Kun Zhou, Yifan Li, Wayne Xin Zhao, Ji-Rong Wen
cs.CL
Recently, continuous diffusion models (CDM) have been introduced into non-autoregressive (NAR) text-to-text generation. However, the discrete nature of text increases the difficulty of CDM to generate coherent and fluent texts, and also causes the incompatibility problem between CDM and advanced NLP techniques, especia...
http://arxiv.org/pdf/2305.04044v1
2023-05-06T13:20:31
2023-05-06T13:20:31
{ "title": [ -0.02609253, -0.005420685, -0.05783081, -0.05065918, 0.0030117035, -0.005821228, -0.03665161, 0.008506775, 0.011077881, 0.026473999, -0.029067993, 0.032684326, 0.003232956, -0.008964539, -0.042053223, 0.0024986267, -0.023513794, 0.03...
http://arxiv.org/abs/2305.04039v1
Refining the Responses of LLMs by Themselves
Tianqiang Yan, Tiansheng Xu
cs.CL, cs.AI
In this paper, we propose a simple yet efficient approach based on prompt engineering that leverages the large language model itself to optimize its answers without relying on auxiliary models. We introduce an iterative self-evaluating optimization mechanism, with the potential for improved output quality as iterations...
http://arxiv.org/pdf/2305.04039v1
2023-05-06T13:03:45
2023-05-06T13:03:45
{ "title": [ -0.01146698, -0.03878784, -0.0052948, -0.037109375, -0.06616211, 0.023330688, -0.04333496, 0.020721436, -0.03125, 0.022705078, -0.014427185, 0.008338928, -0.03262329, -0.04510498, 0.025543213, -0.020141602, 0.021621704, 0.012054443, ...
http://arxiv.org/abs/2305.04003v3
ANTONIO: Towards a Systematic Method of Generating NLP Benchmarks for Verification
Marco Casadio, Luca Arnaboldi, Matthew L. Daggitt, Omri Isac, Tanvi Dinkar, Daniel Kienitz, Verena Rieser, Ekaterina Komendantskaya
cs.CL, cs.AI, cs.LG
Verification of machine learning models used in Natural Language Processing (NLP) is known to be a hard problem. In particular, many known neural network verification methods that work for computer vision and other numeric datasets do not work for NLP. Here, we study technical reasons that underlie this problem. Based ...
http://arxiv.org/pdf/2305.04003v3
2023-08-15T15:09:02
2023-05-06T10:36:39
{ "title": [ -0.031677246, -0.015380859, -0.026870728, -0.07556152, -0.0002501011, 0.01676941, -0.06719971, -0.009880066, -0.011917114, 0.033416748, -0.015396118, 0.028533936, -0.019210815, -0.010643005, -0.058563232, 0.0038490295, -0.035583496, ...
http://arxiv.org/abs/2305.03987v1
Replicating Complex Dialogue Policy of Humans via Offline Imitation Learning with Supervised Regularization
Zhoujian Sun, Chenyang Zhao, Zhengxing Huang, Nai Ding
cs.CL, cs.AI
Policy learning (PL) is a module of a task-oriented dialogue system that trains an agent to make actions in each dialogue turn. Imitating human action is a fundamental problem of PL. However, both supervised learning (SL) and reinforcement learning (RL) frameworks cannot imitate humans well. Training RL models require ...
http://arxiv.org/pdf/2305.03987v1
2023-05-06T09:27:58
2023-05-06T09:27:58
{ "title": [ 0.0104522705, 0.013221741, -0.011070251, -0.058410645, 0.004463196, 0.004501343, -0.03781128, -0.010299683, -0.03894043, 0.0050849915, -0.0055122375, 0.0047912598, -0.013877869, -0.037994385, 0.0007047653, -0.033843994, -0.024429321, ...
http://arxiv.org/abs/2305.03981v1
Pre-training Language Model as a Multi-perspective Course Learner
Beiduo Chen, Shaohan Huang, Zihan Zhang, Wu Guo, Zhenhua Ling, Haizhen Huang, Furu Wei, Weiwei Deng, Qi Zhang
cs.CL
ELECTRA, the generator-discriminator pre-training framework, has achieved impressive semantic construction capability among various downstream tasks. Despite the convincing performance, ELECTRA still faces the challenges of monotonous training and deficient interaction. Generator with only masked language modeling (MLM...
http://arxiv.org/pdf/2305.03981v1
2023-05-06T09:02:10
2023-05-06T09:02:10
{ "title": [ -0.012306213, -0.0000062584877, -0.045562744, -0.05432129, -0.01739502, -0.019821167, -0.047058105, 0.008262634, -0.025726318, 0.028274536, -0.00819397, -0.013145447, -0.0032444, -0.04055786, -0.010612488, -0.020736694, -0.008331299, ...
http://arxiv.org/abs/2305.03977v1
An Adversarial Non-Autoregressive Model for Text Generation with Incomplete Information
Da Ren, Qing Li
cs.CL
Non-autoregressive models have been widely studied in the Complete Information Scenario (CIS), in which the models have complete input information to obtain corresponding output. However, their explorations in the Incomplete Information Scenario (IIS) are extremely limited. Our analyses reveal that the IIS's incomplete...
http://arxiv.org/pdf/2305.03977v1
2023-05-06T08:43:33
2023-05-06T08:43:33
{ "title": [ -0.035003662, 0.0007572174, -0.04006958, -0.03640747, -0.016708374, 0.016052246, -0.06958008, -0.017074585, -0.010932922, 0.032043457, -0.017501831, 0.034851074, -0.0071983337, -0.0030765533, -0.039520264, -0.0076904297, -0.011886597, ...
http://arxiv.org/abs/2305.03973v1
DiscoPrompt: Path Prediction Prompt Tuning for Implicit Discourse Relation Recognition
Chunkit Chan, Xin Liu, Jiayang Cheng, Zihan Li, Yangqiu Song, Ginny Y. Wong, Simon See
cs.CL
Implicit Discourse Relation Recognition (IDRR) is a sophisticated and challenging task to recognize the discourse relations between the arguments with the absence of discourse connectives. The sense labels for each discourse relation follow a hierarchical classification scheme in the annotation process (Prasad et al., ...
http://arxiv.org/pdf/2305.03973v1
2023-05-06T08:16:07
2023-05-06T08:16:07
{ "title": [ 0.01525116, -0.007881165, -0.039489746, -0.09588623, -0.016921997, -0.001367569, -0.06793213, 0.018432617, 0.034423828, 0.005908966, -0.022903442, -0.012565613, -0.0003592968, -0.025909424, -0.029251099, -0.012413025, -0.023635864, 0...
http://arxiv.org/abs/2305.03971v3
Adaptive loose optimization for robust question answering
Jie Ma, Pinghui Wang, Zewei Wang, Dechen Kong, Min Hu, Ting Han, Jun Liu
cs.CL, cs.CV, H.3.4
Question answering methods are well-known for leveraging data bias, such as the language prior in visual question answering and the position bias in machine reading comprehension (extractive question answering). Current debiasing methods often come at the cost of significant in-distribution performance to achieve favor...
http://arxiv.org/pdf/2305.03971v3
2023-10-31T01:45:39
2023-05-06T08:09:46
{ "title": [ -0.026428223, -0.0029335022, -0.024276733, -0.036010742, -0.011405945, 0.009254456, -0.03062439, -0.0052375793, 0.007446289, 0.047698975, -0.009552002, 0.0026054382, -0.00289917, -0.00111866, -0.06878662, 0.006214142, 0.0020217896, 0...
http://arxiv.org/abs/2305.03970v1
NER-to-MRC: Named-Entity Recognition Completely Solving as Machine Reading Comprehension
Yuxiang Zhang, Junjie Wang, Xinyu Zhu, Tetsuya Sakai, Hayato Yamana
cs.CL
Named-entity recognition (NER) detects texts with predefined semantic labels and is an essential building block for natural language processing (NLP). Notably, recent NER research focuses on utilizing massive extra data, including pre-training corpora and incorporating search engines. However, these methods suffer from...
http://arxiv.org/pdf/2305.03970v1
2023-05-06T08:05:22
2023-05-06T08:05:22
{ "title": [ -0.025375366, -0.01586914, -0.025527954, -0.059692383, -0.0024414062, 0.0045051575, -0.03378296, -0.006652832, 0.015930176, 0.01687622, -0.03744507, 0.01763916, 0.0178833, -0.0061149597, -0.030334473, 0.008728027, -0.011581421, 0.035...
http://arxiv.org/abs/2305.03960v2
Beyond Rule-based Named Entity Recognition and Relation Extraction for Process Model Generation from Natural Language Text
Julian Neuberger, Lars Ackermann, Stefan Jablonski
cs.CL, cs.LG
Process-aware information systems offer extensive advantages to companies, facilitating planning, operations, and optimization of day-to-day business activities. However, the time-consuming but required step of designing formal business process models often hampers the potential of these systems. To overcome this chall...
http://arxiv.org/pdf/2305.03960v2
2023-08-07T06:35:25
2023-05-06T07:06:47
{ "title": [ -0.041229248, -0.012641907, -0.061676025, -0.06781006, -0.000010788441, 0.01751709, -0.08392334, 0.0014305115, -0.0021362305, 0.0067634583, -0.022537231, 0.010147095, -0.009613037, -0.018997192, -0.030136108, -0.0052719116, -0.020202637,...
http://arxiv.org/abs/2305.03949v1
Label-Free Multi-Domain Machine Translation with Stage-wise Training
Fan Zhang, Mei Tu, Sangha Kim, Song Liu, Jinyao Yan
cs.CL
Most multi-domain machine translation models rely on domain-annotated data. Unfortunately, domain labels are usually unavailable in both training processes and real translation scenarios. In this work, we propose a label-free multi-domain machine translation model which requires only a few or no domain-annotated data i...
http://arxiv.org/pdf/2305.03949v1
2023-05-06T06:30:29
2023-05-06T06:30:29
{ "title": [ -0.021224976, -0.017028809, -0.020629883, -0.076660156, -0.0015687943, 0.0129852295, -0.027709961, 0.026947021, -0.02947998, 0.0491333, 0.016693115, 0.007827759, 0.0047035217, -0.024902344, -0.028762817, 0.016418457, -0.016708374, 0....
http://arxiv.org/abs/2305.03937v1
Residual Prompt Tuning: Improving Prompt Tuning with Residual Reparameterization
Anastasia Razdaibiedina, Yuning Mao, Rui Hou, Madian Khabsa, Mike Lewis, Jimmy Ba, Amjad Almahairi
cs.CL, cs.AI
Prompt tuning is one of the successful approaches for parameter-efficient tuning of pre-trained language models. Despite being arguably the most parameter-efficient (tuned soft prompts constitute <0.1% of total parameters), it typically performs worse than other efficient tuning methods and is quite sensitive to hyper-...
http://arxiv.org/pdf/2305.03937v1
2023-05-06T05:35:14
2023-05-06T05:35:14
{ "title": [ -0.045928955, -0.02609253, -0.035308838, -0.06958008, -0.045806885, 0.022720337, -0.09436035, 0.0003671646, -0.02532959, 0.03302002, 0.011688232, 0.0023937225, 0.007446289, -0.046142578, -0.0037841797, 0.012069702, 0.00737381, 0.0235...
http://arxiv.org/abs/2305.03923v1
Active Continual Learning: Labelling Queries in a Sequence of Tasks
Thuy-Trang Vu, Shahram Khadivi, Dinh Phung, Gholamreza Haffari
cs.LG, cs.CL
Acquiring new knowledge without forgetting what has been learned in a sequence of tasks is the central focus of continual learning (CL). While tasks arrive sequentially, the training data are often prepared and annotated independently, leading to CL of incoming supervised learning tasks. This paper considers the under-...
http://arxiv.org/pdf/2305.03923v1
2023-05-06T04:11:03
2023-05-06T04:11:03
{ "title": [ 0.025894165, 0.019363403, -0.04562378, -0.06060791, 0.03805542, -0.041870117, 0.008872986, -0.0071258545, 0.011924744, 0.041870117, -0.0007109642, 0.000688076, -0.024841309, -0.04296875, -0.042388916, 0.008895874, 0.020736694, 0.0060...
http://arxiv.org/abs/2305.06152v2
Structure-CLIP: Towards Scene Graph Knowledge to Enhance Multi-modal Structured Representations
Yufeng Huang, Jiji Tang, Zhuo Chen, Rongsheng Zhang, Xinfeng Zhang, Weijie Chen, Zeng Zhao, Zhou Zhao, Tangjie Lv, Zhipeng Hu, Wen Zhang
cs.CL, cs.AI, cs.MM
Large-scale vision-language pre-training has achieved significant performance in multi-modal understanding and generation tasks. However, existing methods often perform poorly on image-text matching tasks that require structured representations, i.e., representations of objects, attributes, and relations. Previous mode...
http://arxiv.org/pdf/2305.06152v2
2023-08-24T04:44:39
2023-05-06T03:57:05
{ "title": [ -0.0051078796, -0.002046585, -0.04135132, -0.070007324, -0.006538391, -0.0023860931, -0.06756592, -0.009788513, 0.011421204, -0.0039138794, -0.012077332, 0.0015544891, -0.026977539, -0.037139893, -0.0087890625, -0.011383057, 0.013397217,...
http://arxiv.org/abs/2305.03915v1
HateMM: A Multi-Modal Dataset for Hate Video Classification
Mithun Das, Rohit Raj, Punyajoy Saha, Binny Mathew, Manish Gupta, Animesh Mukherjee
cs.CV, cs.CL, cs.MM
Hate speech has become one of the most significant issues in modern society, having implications in both the online and the offline world. Due to this, hate speech research has recently gained a lot of traction. However, most of the work has primarily focused on text media with relatively little work on images and even...
http://arxiv.org/pdf/2305.03915v1
2023-05-06T03:39:00
2023-05-06T03:39:00
{ "title": [ 0.013069153, 0.0051193237, -0.048065186, -0.06072998, 0.014930725, -0.02684021, -0.021774292, -0.0093688965, 0.021560669, 0.03225708, -0.04156494, 0.000623703, 0.018295288, -0.009597778, -0.046081543, -0.01083374, -0.0052604675, 0.04...
http://arxiv.org/abs/2305.03881v2
Fairness in Image Search: A Study of Occupational Stereotyping in Image Retrieval and its Debiasing
Swagatika Dash
cs.IR, cs.CL, cs.CV
Multi-modal search engines have experienced significant growth and widespread use in recent years, making them the second most common internet use. While search engine systems offer a range of services, the image search field has recently become a focal point in the information retrieval community, as the adage goes, "...
http://arxiv.org/pdf/2305.03881v2
2023-08-22T16:09:59
2023-05-06T00:24:44
{ "title": [ 0.027954102, 0.021514893, -0.022903442, -0.02003479, 0.029541016, -0.030456543, 0.022720337, -0.04043579, 0.00087070465, 0.03451538, -0.043182373, 0.0077705383, 0.014709473, -0.0006427765, -0.057403564, -0.010429382, -0.017623901, 0....
http://arxiv.org/abs/2305.03880v1
NorBench -- A Benchmark for Norwegian Language Models
David Samuel, Andrey Kutuzov, Samia Touileb, Erik Velldal, Lilja Øvrelid, Egil Rønningstad, Elina Sigdel, Anna Palatkina
cs.CL
We present NorBench: a streamlined suite of NLP tasks and probes for evaluating Norwegian language models (LMs) on standardized data splits and evaluation metrics. We also introduce a range of new Norwegian language models (both encoder and encoder-decoder based). Finally, we compare and analyze their performance, alon...
http://arxiv.org/pdf/2305.03880v1
2023-05-06T00:20:24
2023-05-06T00:20:24
{ "title": [ -0.016571045, -0.01234436, -0.022079468, -0.06896973, -0.0009469986, 0.00592041, -0.064575195, -0.0028438568, -0.006263733, 0.035308838, -0.02330017, 0.00655365, 0.034179688, -0.018875122, -0.025131226, 0.01676941, -0.014350891, 0.04...
http://arxiv.org/abs/2305.03873v1
Train Global, Tailor Local: Minimalist Multilingual Translation into Endangered Languages
Zhong Zhou, Jan Niehues, Alex Waibel
cs.CL
In many humanitarian scenarios, translation into severely low resource languages often does not require a universal translation engine, but a dedicated text-specific translation engine. For example, healthcare records, hygienic procedures, government communication, emergency procedures and religious texts are all limit...
http://arxiv.org/pdf/2305.03873v1
2023-05-05T23:22:16
2023-05-05T23:22:16
{ "title": [ -0.045135498, -0.007926941, -0.025375366, -0.03289795, -0.023635864, 0.0072784424, -0.034454346, -0.0030498505, -0.0413208, 0.050628662, 0.0037670135, 0.022506714, 0.012191772, -0.030563354, -0.02293396, -0.017364502, -0.0072250366, ...
http://arxiv.org/abs/2305.03851v1
Large Language Models in Sport Science & Medicine: Opportunities, Risks and Considerations
Mark Connor, Michael O'Neill
cs.CL, I.2.7
This paper explores the potential opportunities, risks, and challenges associated with the use of large language models (LLMs) in sports science and medicine. LLMs are large neural networks with transformer style architectures trained on vast amounts of textual data, and typically refined with human feedback. LLMs can ...
http://arxiv.org/pdf/2305.03851v1
2023-05-05T21:20:02
2023-05-05T21:20:02
{ "title": [ -0.029388428, -0.013061523, -0.03062439, -0.045562744, 0.0019254684, -0.0005245209, -0.043060303, -0.020904541, -0.035186768, 0.025299072, 0.0059776306, -0.011734009, -0.02532959, -0.038909912, -0.022994995, -0.0051994324, 0.0146102905, ...
http://arxiv.org/abs/2305.03845v1
CLaC at SemEval-2023 Task 2: Comparing Span-Prediction and Sequence-Labeling approaches for NER
Harsh Verma, Sabine Bergler
cs.CL, cs.AI
This paper summarizes the CLaC submission for the MultiCoNER 2 task which concerns the recognition of complex, fine-grained named entities. We compare two popular approaches for NER, namely Sequence Labeling and Span Prediction. We find that our best Span Prediction system performs slightly better than our best Sequenc...
http://arxiv.org/pdf/2305.03845v1
2023-05-05T20:49:40
2023-05-05T20:49:40
{ "title": [ -0.0019235611, -0.025009155, -0.023132324, -0.074401855, -0.010063171, 0.020751953, -0.06402588, 0.004814148, -0.038726807, 0.02368164, -0.020019531, 0.024429321, 0.00944519, -0.04611206, 0.018493652, 0.010986328, -0.01071167, 0.0731...
http://arxiv.org/abs/2305.03827v2
Uncertainty-Aware Bootstrap Learning for Joint Extraction on Distantly-Supervised Data
Yufei Li, Xiao Yu, Yanchi Liu, Haifeng Chen, Cong Liu
cs.CL, cs.LG
Jointly extracting entity pairs and their relations is challenging when working on distantly-supervised data with ambiguous or noisy labels. To mitigate such impact, we propose uncertainty-aware bootstrap learning, which is motivated by the intuition that the higher uncertainty of an instance, the more likely the model...
http://arxiv.org/pdf/2305.03827v2
2023-06-08T22:22:24
2023-05-05T20:06:11
{ "title": [ -0.016113281, 0.04257202, -0.08294678, -0.06958008, 0.039764404, -0.018600464, -0.0574646, -0.0061454773, -0.014427185, -0.01689148, -0.04840088, 0.014640808, -0.041900635, -0.034820557, -0.0068855286, -0.03717041, -0.012580872, 0.04...
http://arxiv.org/abs/2305.03819v1
Adapting Transformer Language Models for Predictive Typing in Brain-Computer Interfaces
Shijia Liu, David A. Smith
cs.CL
Brain-computer interfaces (BCI) are an important mode of alternative and augmentative communication for many people. Unlike keyboards, many BCI systems do not display even the 26 letters of English at one time, let alone all the symbols in more complex systems. Using language models to make character-level predictions,...
http://arxiv.org/pdf/2305.03819v1
2023-05-05T19:47:41
2023-05-05T19:47:41
{ "title": [ -0.036376953, -0.0055236816, -0.01083374, -0.004245758, -0.025177002, 0.0082092285, -0.047088623, -0.028274536, 0.007385254, 0.04385376, -0.01864624, -0.012557983, 0.05050659, -0.032836914, -0.028640747, 0.012702942, -0.0132751465, 0...
http://arxiv.org/abs/2305.03796v1
Transformer Working Memory Enables Regular Language Reasoning and Natural Language Length Extrapolation
Ta-Chung Chi, Ting-Han Fan, Alexander I. Rudnicky, Peter J. Ramadge
cs.CL
Unlike recurrent models, conventional wisdom has it that Transformers cannot perfectly model regular languages. Inspired by the notion of working memory, we propose a new Transformer variant named RegularGPT. With its novel combination of Weight-Sharing, Adaptive-Depth, and Sliding-Dilated-Attention, RegularGPT constru...
http://arxiv.org/pdf/2305.03796v1
2023-05-05T18:54:40
2023-05-05T18:54:40
{ "title": [ -0.013000488, -0.044677734, -0.058532715, -0.03543091, -0.06213379, -0.026519775, -0.06726074, 0.029678345, -0.064208984, 0.022201538, -0.015541077, 0.0045700073, 0.024627686, -0.04055786, 0.050933838, -0.028518677, -0.00094890594, 0...
http://arxiv.org/abs/2305.03793v1
Towards Zero-Shot Frame Semantic Parsing with Task Agnostic Ontologies and Simple Labels
Danilo Ribeiro, Omid Abdar, Jack Goetz, Mike Ross, Annie Dong, Kenneth Forbus, Ahmed Mohamed
cs.CL, cs.LG, I.2.7; I.2.6
Frame semantic parsing is an important component of task-oriented dialogue systems. Current models rely on a significant amount training data to successfully identify the intent and slots in the user's input utterance. This creates a significant barrier for adding new domains to virtual assistant capabilities, as creat...
http://arxiv.org/pdf/2305.03793v1
2023-05-05T18:47:18
2023-05-05T18:47:18
{ "title": [ -0.009994507, -0.002664566, -0.044281006, -0.09118652, 0.0016880035, -0.0036811829, -0.058532715, 0.023834229, 0.0003402233, 0.032073975, -0.011657715, 0.005264282, -0.01939392, -0.010414124, -0.050201416, 0.00094890594, -0.012168884, ...
http://arxiv.org/abs/2305.03788v1
Harnessing the Power of BERT in the Turkish Clinical Domain: Pretraining Approaches for Limited Data Scenarios
Hazal Türkmen, Oğuz Dikenelli, Cenk Eraslan, Mehmet Cem Çallı, Süha Süreyya Özbek
cs.CL
In recent years, major advancements in natural language processing (NLP) have been driven by the emergence of large language models (LLMs), which have significantly revolutionized research and development within the field. Building upon this progress, our study delves into the effects of various pre-training methodolog...
http://arxiv.org/pdf/2305.03788v1
2023-05-05T18:39:07
2023-05-05T18:39:07
{ "title": [ 0.04824829, -0.02406311, -0.026687622, -0.09082031, -0.0019702911, -0.01687622, -0.043701172, 0.0028457642, -0.013687134, 0.02381897, 0.0052108765, -0.0005493164, 0.0012102127, -0.054748535, -0.003435135, -0.0053977966, -0.011413574, ...
http://arxiv.org/abs/2305.03726v1
Otter: A Multi-Modal Model with In-Context Instruction Tuning
Bo Li, Yuanhan Zhang, Liangyu Chen, Jinghao Wang, Jingkang Yang, Ziwei Liu
cs.CV, cs.CL
Large language models (LLMs) have demonstrated significant universal capabilities as few/zero-shot learners in various tasks due to their pre-training on vast amounts of text data, as exemplified by GPT-3, which boosted to InstrctGPT and ChatGPT, effectively following natural language instructions to accomplish real-wo...
http://arxiv.org/pdf/2305.03726v1
2023-05-05T17:59:46
2023-05-05T17:59:46
{ "title": [ -0.05142212, -0.004047394, -0.03265381, -0.048858643, -0.013801575, 0.0006017685, -0.08392334, -0.018173218, 0.0045776367, 0.03515625, -0.031463623, 0.014732361, 0.00969696, -0.016983032, -0.026443481, 0.0023784637, -0.015388489, 0.0...
http://arxiv.org/abs/2305.03715v1
Large Language Models in Ambulatory Devices for Home Health Diagnostics: A case study of Sickle Cell Anemia Management
Oluwatosin Ogundare, Subuola Sofolahan
cs.CL
This study investigates the potential of an ambulatory device that incorporates Large Language Models (LLMs) in cadence with other specialized ML models to assess anemia severity in sickle cell patients in real time. The device would rely on sensor data that measures angiogenic material levels to assess anemia severity...
http://arxiv.org/pdf/2305.03715v1
2023-05-05T17:55:49
2023-05-05T17:55:49
{ "title": [ -0.03173828, -0.02670288, -0.03173828, -0.05960083, -0.033050537, 0.014625549, -0.09069824, -0.0020332336, -0.007797241, 0.04031372, -0.0087509155, 0.0032997131, 0.008743286, -0.04269409, -0.0058555603, -0.005207062, 0.022399902, 0.0...
http://arxiv.org/abs/2305.03695v3
Vera: A General-Purpose Plausibility Estimation Model for Commonsense Statements
Jiacheng Liu, Wenya Wang, Dianzhuo Wang, Noah A. Smith, Yejin Choi, Hannaneh Hajishirzi
cs.CL, cs.AI
Despite the much discussed capabilities of today's language models, they are still prone to silly and unexpected commonsense failures. We consider a retrospective verification approach that reflects on the correctness of LM outputs, and introduce Vera, a general-purpose model that estimates the plausibility of declarat...
http://arxiv.org/pdf/2305.03695v3
2023-10-18T14:48:51
2023-05-05T17:15:32
{ "title": [ 0.0038146973, -0.023925781, -0.008132935, -0.035308838, -0.0011587143, -0.006427765, -0.0138168335, -0.040985107, 0.011436462, 0.0020942688, -0.009857178, -0.005554199, -0.023330688, -0.032348633, -0.032073975, -0.0048675537, -0.02980041...
http://arxiv.org/abs/2305.03688v3
DAMO-NLP at SemEval-2023 Task 2: A Unified Retrieval-augmented System for Multilingual Named Entity Recognition
Zeqi Tan, Shen Huang, Zixia Jia, Jiong Cai, Yinghui Li, Weiming Lu, Yueting Zhuang, Kewei Tu, Pengjun Xie, Fei Huang, Yong Jiang
cs.CL
The MultiCoNER \RNum{2} shared task aims to tackle multilingual named entity recognition (NER) in fine-grained and noisy scenarios, and it inherits the semantic ambiguity and low-context setting of the MultiCoNER \RNum{1} task. To cope with these problems, the previous top systems in the MultiCoNER \RNum{1} either inco...
http://arxiv.org/pdf/2305.03688v3
2023-05-17T03:14:00
2023-05-05T16:59:26
{ "title": [ -0.015899658, 0.0028095245, -0.046173096, -0.070617676, 0.018493652, 0.011009216, -0.059906006, -0.005268097, 0.014434814, 0.03845215, -0.025772095, 0.032409668, 0.022506714, 0.0070495605, -0.0340271, 0.027786255, -0.026443481, 0.056...
http://arxiv.org/abs/2305.03668v2
A Suite of Generative Tasks for Multi-Level Multimodal Webpage Understanding
Andrea Burns, Krishna Srinivasan, Joshua Ainslie, Geoff Brown, Bryan A. Plummer, Kate Saenko, Jianmo Ni, Mandy Guo
cs.CL, cs.CV
Webpages have been a rich, scalable resource for vision-language and language only tasks. Yet only pieces of webpages are kept in existing datasets: image-caption pairs, long text articles, or raw HTML, never all in one place. Webpage tasks have resultingly received little attention and structured image-text data left ...
http://arxiv.org/pdf/2305.03668v2
2023-10-20T13:18:06
2023-05-05T16:38:05
{ "title": [ -0.0018367767, -0.0090408325, -0.043151855, -0.062469482, 0.00217247, -0.016967773, -0.03540039, 0.014862061, 0.0284729, 0.008964539, -0.0012102127, 0.016601562, -0.0040626526, -0.031463623, -0.01889038, 0.0116119385, -0.0027503967, ...
http://arxiv.org/abs/2305.03661v1
Predicting COVID-19 and pneumonia complications from admission texts
Dmitriy Umerenkov, Oleg Cherkashin, Alexander Nesterov, Victor Gombolevskiy, Irina Demko, Alexander Yalunin, Vladimir Kokh
cs.CL, cs.AI
In this paper we present a novel approach to risk assessment for patients hospitalized with pneumonia or COVID-19 based on their admission reports. We applied a Longformer neural network to admission reports and other textual data available shortly after admission to compute risk scores for the patients. We used patien...
http://arxiv.org/pdf/2305.03661v1
2023-05-05T16:28:44
2023-05-05T16:28:44
{ "title": [ 0.0040512085, -0.0012683868, -0.03741455, -0.07733154, -0.014778137, 0.006587982, -0.049743652, 0.015960693, -0.03363037, 0.033599854, -0.026794434, 0.022949219, 0.021774292, -0.026519775, 0.01689148, -0.026367188, 0.00894165, 0.0236...
http://arxiv.org/abs/2305.03660v1
Retrieval Augmented Chest X-Ray Report Generation using OpenAI GPT models
Mercy Ranjit, Gopinath Ganapathy, Ranjit Manuel, Tanuja Ganu
cs.CL, cs.AI, cs.IR, cs.LG, I.2; J.3; H.3
We propose Retrieval Augmented Generation (RAG) as an approach for automated radiology report writing that leverages multimodally aligned embeddings from a contrastively pretrained vision language model for retrieval of relevant candidate radiology text for an input radiology image and a general domain generative model...
http://arxiv.org/pdf/2305.03660v1
2023-05-05T16:28:03
2023-05-05T16:28:03
{ "title": [ -0.0055618286, 0.020324707, -0.029922485, -0.004386902, -0.009216309, -0.010795593, -0.015464783, -0.045654297, -0.015602112, 0.03765869, -0.01914978, 0.00071144104, 0.012870789, -0.030014038, -0.042144775, -0.0029144287, -0.0049057007, ...
http://arxiv.org/abs/2305.03655v2
White-Box Multi-Objective Adversarial Attack on Dialogue Generation
Yufei Li, Zexin Li, Yingfan Gao, Cong Liu
cs.CL, cs.CR, cs.LG
Pre-trained transformers are popular in state-of-the-art dialogue generation (DG) systems. Such language models are, however, vulnerable to various adversarial samples as studied in traditional tasks such as text classification, which inspires our curiosity about their robustness in DG systems. One main challenge of at...
http://arxiv.org/pdf/2305.03655v2
2023-05-08T15:16:05
2023-05-05T16:21:24
{ "title": [ -0.007286072, -0.008491516, -0.03137207, -0.03692627, 0.009750366, -0.014465332, -0.0059928894, -0.00970459, -0.046325684, 0.037475586, 0.040283203, 0.01889038, -0.0098724365, -0.039093018, -0.034454346, -0.0038394928, -0.028625488, ...
http://arxiv.org/abs/2305.03642v3
Jointly Extracting Interventions, Outcomes, and Findings from RCT Reports with LLMs
Somin Wadhwa, Jay DeYoung, Benjamin Nye, Silvio Amir, Byron C. Wallace
cs.CL
Results from Randomized Controlled Trials (RCTs) establish the comparative effectiveness of interventions, and are in turn critical inputs for evidence-based care. However, results from RCTs are presented in (often unstructured) natural language articles describing the design, execution, and outcomes of trials; clinici...
http://arxiv.org/pdf/2305.03642v3
2023-07-18T01:36:42
2023-05-05T16:02:06
{ "title": [ -0.014556885, 0.002412796, -0.032165527, -0.062194824, -0.04119873, 0.011100769, -0.07836914, -0.0058670044, -0.04257202, -0.0035305023, 0.00074863434, 0.04345703, -0.04751587, -0.05291748, 0.04220581, -0.017364502, -0.012496948, 0.0...
http://arxiv.org/abs/2305.03610v1
Data Curation for Image Captioning with Text-to-Image Generative Models
Wenyan Li, Jonas F. Lotz, Chen Qiu, Desmond Elliott
cs.CV, cs.AI, cs.CL
Recent advances in image captioning are mainly driven by large-scale vision-language pretraining, relying heavily on computational resources and increasingly large multimodal datasets. Instead of scaling up pretraining data, we ask whether it is possible to improve performance by improving the quality of the samples in...
http://arxiv.org/pdf/2305.03610v1
2023-05-05T15:16:07
2023-05-05T15:16:07
{ "title": [ -0.001156807, -0.00091171265, -0.019744873, -0.02960205, -0.01576233, -0.013076782, -0.011192322, -0.021759033, 0.01928711, 0.0625, -0.0491333, 0.028244019, 0.0087890625, 0.008529663, -0.0602417, 0.0027999878, -0.01109314, 0.03512573...
http://arxiv.org/abs/2305.03602v2
A Dual Semantic-Aware Recurrent Global-Adaptive Network For Vision-and-Language Navigation
Liuyi Wang, Zongtao He, Jiagui Tang, Ronghao Dang, Naijia Wang, Chengju Liu, Qijun Chen
cs.CV, cs.CL
Vision-and-Language Navigation (VLN) is a realistic but challenging task that requires an agent to locate the target region using verbal and visual cues. While significant advancements have been achieved recently, there are still two broad limitations: (1) The explicit information mining for significant guiding semanti...
http://arxiv.org/pdf/2305.03602v2
2023-05-30T02:33:12
2023-05-05T15:06:08
{ "title": [ -0.012466431, -0.012840271, -0.012931824, -0.04269409, -0.013877869, -0.01663208, -0.016281128, 0.010246277, 0.024490356, 0.030944824, -0.014038086, -0.0026607513, -0.004257202, -0.022476196, -0.03604126, 0.012886047, 0.018112183, 0....
http://arxiv.org/abs/2305.03598v3
NLI4CT: Multi-Evidence Natural Language Inference for Clinical Trial Reports
Maël Jullien, Marco Valentino, Hannah Frost, Paul O'Regan, Donal Landers, André Freitas
cs.CL, cs.AI, cs.LG
How can we interpret and retrieve medical evidence to support clinical decisions? Clinical trial reports (CTR) amassed over the years contain indispensable information for the development of personalized medicine. However, it is practically infeasible to manually inspect over 400,000+ clinical trial reports in order to...
http://arxiv.org/pdf/2305.03598v3
2023-10-28T09:56:49
2023-05-05T15:03:01
{ "title": [ -0.04168701, -0.033843994, -0.052001953, -0.0692749, -0.020187378, 0.016906738, -0.06915283, 0.013961792, -0.00623703, 0.017440796, -0.02734375, 0.014884949, -0.0113220215, -0.011909485, -0.023025513, 0.0029335022, -0.023452759, 0.04...
http://arxiv.org/abs/2305.03584v2
Now It Sounds Like You: Learning Personalized Vocabulary On Device
Sid Wang, Ashish Shenoy, Pierce Chuang, John Nguyen
cs.CL, cs.AI
In recent years, Federated Learning (FL) has shown significant advancements in its ability to perform various natural language processing (NLP) tasks. This work focuses on applying personalized FL for on-device language modeling. Due to limitations of memory and latency, these models cannot support the complexity of su...
http://arxiv.org/pdf/2305.03584v2
2023-10-05T20:45:46
2023-05-05T14:44:20
{ "title": [ -0.015419006, -0.006023407, -0.033294678, -0.021347046, -0.010292053, -0.0052871704, -0.012809753, 0.009414673, -0.05255127, 0.033111572, -0.00011509657, 0.037139893, -0.01373291, -0.026794434, 0.008834839, -0.0068130493, -0.010757446, ...
http://arxiv.org/abs/2305.03573v1
In-context Learning as Maintaining Coherency: A Study of On-the-fly Machine Translation Using Large Language Models
Suzanna Sia, Kevin Duh
cs.CL, cs.AI
The phenomena of in-context learning has typically been thought of as "learning from examples". In this work which focuses on Machine Translation, we present a perspective of in-context learning as the desired generation task maintaining coherency with its context, i.e., the prompt examples. We first investigate random...
http://arxiv.org/pdf/2305.03573v1
2023-05-05T14:30:20
2023-05-05T14:30:20
{ "title": [ -0.033935547, -0.0015630722, -0.025756836, -0.0446167, -0.0064849854, -0.0055274963, -0.034820557, 0.0003836155, -0.030273438, 0.064819336, -0.01878357, 0.022521973, 0.032958984, -0.0040397644, -0.035461426, 0.00066041946, -0.0076408386,...
http://arxiv.org/abs/2305.03520v2
Context-Aware Semantic Similarity Measurement for Unsupervised Word Sense Disambiguation
Jorge Martinez-Gil
cs.CL, cs.AI
The issue of word sense ambiguity poses a significant challenge in natural language processing due to the scarcity of annotated data to feed machine learning models to face the challenge. Therefore, unsupervised word sense disambiguation methods have been developed to overcome that challenge without relying on annotate...
http://arxiv.org/pdf/2305.03520v2
2023-11-14T11:46:15
2023-05-05T13:50:04
{ "title": [ -0.03186035, 0.011291504, -0.048339844, -0.028549194, 0.022323608, -0.0012741089, -0.0124053955, -0.012809753, 0.02520752, 0.048706055, -0.05340576, 0.015151978, 0.03414917, 0.0035190582, -0.036712646, 0.009147644, -0.005493164, 0.02...
http://arxiv.org/abs/2305.04928v3
From Zero to Hero: Harnessing Transformers for Biomedical Named Entity Recognition in Zero- and Few-shot Contexts
Miloš Košprdić, Nikola Prodanović, Adela Ljajić, Bojana Bašaragin, Nikola Milošević
cs.CL, cs.AI
Supervised named entity recognition (NER) in the biomedical domain depends on large sets of annotated texts with the given named entities. The creation of such datasets can be time-consuming and expensive, while extraction of new entities requires additional annotation tasks and retraining the model. To address these c...
http://arxiv.org/pdf/2305.04928v3
2023-05-27T19:52:16
2023-05-05T12:14:22
{ "title": [ 0.0017557144, -0.01486969, -0.036590576, -0.07733154, 0.0056037903, 0.008842468, -0.04724121, 0.011062622, 0.003709793, 0.031021118, -0.04058838, 0.013442993, 0.020401001, -0.0047950745, -0.036987305, -0.0010986328, -0.005443573, 0.0...
http://arxiv.org/abs/2305.03461v1
Interactive Acquisition of Fine-grained Visual Concepts by Exploiting Semantics of Generic Characterizations in Discourse
Jonghyuk Park, Alex Lascarides, Subramanian Ramamoorthy
cs.CL
Interactive Task Learning (ITL) concerns learning about unforeseen domain concepts via natural interactions with human users. The learner faces a number of significant constraints: learning should be online, incremental and few-shot, as it is expected to perform tangible belief updates right after novel words denoting ...
http://arxiv.org/pdf/2305.03461v1
2023-05-05T12:06:01
2023-05-05T12:06:01
{ "title": [ 0.024963379, -0.012573242, -0.0045814514, -0.050811768, 0.0026779175, -0.009590149, 0.0072517395, -0.007774353, 0.01600647, 0.016616821, -0.029632568, 0.010627747, -0.025680542, -0.009819031, -0.05895996, -0.032409668, 0.0013723373, ...
http://arxiv.org/abs/2305.03458v1
Multi-View Graph Representation Learning for Answering Hybrid Numerical Reasoning Question
Yifan Wei, Fangyu Lei, Yuanzhe Zhang, Jun Zhao, Kang Liu
cs.CL, cs.AI
Hybrid question answering (HybridQA) over the financial report contains both textual and tabular data, and requires the model to select the appropriate evidence for the numerical reasoning task. Existing methods based on encoder-decoder framework employ a expression tree-based decoder to solve numerical reasoning probl...
http://arxiv.org/pdf/2305.03458v1
2023-05-05T12:00:58
2023-05-05T12:00:58
{ "title": [ -0.0049057007, -0.0035896301, -0.03955078, -0.044006348, -0.01399231, -0.015457153, -0.020690918, 0.02935791, -0.009811401, 0.010826111, -0.024627686, -0.0014476776, -0.013442993, -0.032073975, -0.0154953, -0.03543091, 0.009994507, 0...
http://arxiv.org/abs/2305.03453v3
T-SciQ: Teaching Multimodal Chain-of-Thought Reasoning via Large Language Model Signals for Science Question Answering
Lei Wang, Yi Hu, Jiabang He, Xing Xu, Ning Liu, Hui Liu, Heng Tao Shen
cs.CL
Large Language Models (LLMs) have recently demonstrated exceptional performance in various Natural Language Processing (NLP) tasks. They have also shown the ability to perform chain-of-thought (CoT) reasoning to solve complex problems. Recent studies have explored CoT reasoning in complex multimodal scenarios, such as ...
http://arxiv.org/pdf/2305.03453v3
2023-08-16T12:50:46
2023-05-05T11:56:30
{ "title": [ -0.018539429, -0.025970459, -0.0052948, -0.010932922, -0.03768921, 0.0010585785, -0.02305603, -0.011459351, 0.015426636, 0.002216339, -0.020645142, -0.017547607, 0.007255554, -0.028305054, -0.02809143, -0.019760132, -0.004432678, 0.0...
http://arxiv.org/abs/2305.03445v4
LMs stand their Ground: Investigating the Effect of Embodiment in Figurative Language Interpretation by Language Models
Philipp Wicke
cs.CL
Figurative language is a challenge for language models since its interpretation is based on the use of words in a way that deviates from their conventional order and meaning. Yet, humans can easily understand and interpret metaphors, similes or idioms as they can be derived from embodied metaphors. Language is a proxy ...
http://arxiv.org/pdf/2305.03445v4
2023-06-15T12:33:37
2023-05-05T11:44:12
{ "title": [ -0.014129639, 0.0029067993, -0.012672424, -0.016784668, -0.014419556, -0.016693115, -0.029678345, -0.034729004, -0.0060539246, 0.03567505, -0.0075263977, -0.014404297, 0.039916992, -0.022644043, -0.042633057, 0.011962891, -0.020202637, ...
http://arxiv.org/abs/2305.03429v1
Simulating H.P. Lovecraft horror literature with the ChatGPT large language model
Eduardo C. Garrido-Merchán, José Luis Arroyo-Barrigüete, Roberto Gozalo-Brizuela
cs.CL
In this paper, we present a novel approach to simulating H.P. Lovecraft's horror literature using the ChatGPT large language model, specifically the GPT-4 architecture. Our study aims to generate text that emulates Lovecraft's unique writing style and themes, while also examining the effectiveness of prompt engineering...
http://arxiv.org/pdf/2305.03429v1
2023-05-05T11:03:03
2023-05-05T11:03:03
{ "title": [ -0.030014038, -0.021072388, -0.001329422, 0.005214691, -0.023040771, -0.0009994507, 0.0053253174, -0.016403198, -0.0026817322, 0.038513184, 0.006088257, 0.021652222, 0.04727173, -0.029876709, -0.003124237, 0.0435791, -0.03387451, 0.0...
http://arxiv.org/abs/2305.03423v2
Using ChatGPT for Entity Matching
Ralph Peeters, Christian Bizer
cs.CL
Entity Matching is the task of deciding if two entity descriptions refer to the same real-world entity. State-of-the-art entity matching methods often rely on fine-tuning Transformer models such as BERT or RoBERTa. Two major drawbacks of using these models for entity matching are that (i) the models require significant...
http://arxiv.org/pdf/2305.03423v2
2023-06-22T14:09:56
2023-05-05T10:39:32
{ "title": [ 0.022842407, 0.0079956055, -0.02645874, -0.023544312, -0.013710022, -0.011886597, -0.014213562, -0.028533936, -0.0033473969, 0.041381836, -0.008506775, 0.009140015, 0.019485474, -0.036987305, -0.004558563, 0.022537231, -0.0093307495, ...
http://arxiv.org/abs/2305.03407v1
Online Gesture Recognition using Transformer and Natural Language Processing
G. C. M. Silvestre, F. Balado, O. Akinremi, M. Ramo
cs.CL
The Transformer architecture is shown to provide a powerful machine transduction framework for online handwritten gestures corresponding to glyph strokes of natural language sentences. The attention mechanism is successfully used to create latent representations of an end-to-end encoder-decoder model, solving multi-lev...
http://arxiv.org/pdf/2305.03407v1
2023-05-05T10:17:22
2023-05-05T10:17:22
{ "title": [ -0.012512207, 0.002960205, -0.03970337, -0.044677734, 0.01687622, -0.013046265, -0.01612854, -0.020019531, 0.0061187744, 0.023986816, -0.037597656, -0.015426636, 0.006542206, -0.019439697, -0.0390625, -0.018295288, 0.006084442, 0.036...
http://arxiv.org/abs/2305.06475v1
A Model for Translation of Text from Indian Languages to Bharti Braille Characters
Nisheeth Joshi, Pragya Katyayan
cs.CL, cs.AI
People who are visually impaired face a lot of difficulties while studying. One of the major causes to this is lack of available text in Bharti Braille script. In this paper, we have suggested a scheme to convert text in major Indian languages into Bharti Braille. The system uses a hybrid approach where at first the te...
http://arxiv.org/pdf/2305.06475v1
2023-05-05T09:21:13
2023-05-05T09:21:13
{ "title": [ -0.03289795, -0.01260376, -0.039886475, -0.0073013306, -0.04736328, 0.004142761, -0.050567627, 0.0020866394, -0.0017633438, 0.060180664, 0.006576538, -0.017974854, 0.06915283, -0.04486084, -0.016723633, 0.0064926147, 0.002286911, 0.0...
http://arxiv.org/abs/2305.06157v1
Implications of Multi-Word Expressions on English to Bharti Braille Machine Translation
Nisheeth Joshi, Pragya Katyayan
cs.CL, cs.AI
In this paper, we have shown the improvement of English to Bharti Braille machine translation system. We have shown how we can improve a baseline NMT model by adding some linguistic knowledge to it. This was done for five language pairs where English sentences were translated into five Indian languages and then subsequ...
http://arxiv.org/pdf/2305.06157v1
2023-05-05T08:59:03
2023-05-05T08:59:03
{ "title": [ -0.012557983, -0.015777588, -0.01763916, -0.024429321, -0.03845215, -0.007209778, -0.029541016, 0.013679504, -0.011581421, 0.053131104, 0.006259918, -0.014190674, 0.0552063, -0.04446411, -0.0143966675, 0.008918762, -0.0070762634, 0.0...
http://arxiv.org/abs/2305.03369v1
The MuSe 2023 Multimodal Sentiment Analysis Challenge: Mimicked Emotions, Cross-Cultural Humour, and Personalisation
Lukas Christ, Shahin Amiriparian, Alice Baird, Alexander Kathan, Niklas Müller, Steffen Klug, Chris Gagne, Panagiotis Tzirakis, Eva-Maria Meßner, Andreas König, Alan Cowen, Erik Cambria, Björn W. Schuller
cs.LG, cs.AI, cs.CL, cs.MM
The MuSe 2023 is a set of shared tasks addressing three different contemporary multimodal affect and sentiment analysis problems: In the Mimicked Emotions Sub-Challenge (MuSe-Mimic), participants predict three continuous emotion targets. This sub-challenge utilises the Hume-Vidmimic dataset comprising of user-generated...
http://arxiv.org/pdf/2305.03369v1
2023-05-05T08:53:57
2023-05-05T08:53:57
{ "title": [ 0.012794495, 0.008956909, -0.023147583, -0.033477783, -0.0044898987, -0.019302368, -0.036254883, -0.012023926, 0.03543091, 0.032562256, -0.0020561218, -0.007621765, 0.034332275, -0.033935547, -0.00907135, 0.020187378, -0.005138397, 0...
http://arxiv.org/abs/2305.04927v1
Detecting and Reasoning of Deleted Tweets before they are Posted
Hamdy Mubarak, Samir Abdaljalil, Azza Nassar, Firoj Alam
cs.CL, cs.AI, cs.CY, 68T50, F.2.2; I.2.7
Social media platforms empower us in several ways, from information dissemination to consumption. While these platforms are useful in promoting citizen journalism, public awareness etc., they have misuse potentials. Malicious users use them to disseminate hate-speech, offensive content, rumor etc. to gain social and po...
http://arxiv.org/pdf/2305.04927v1
2023-05-05T08:25:07
2023-05-05T08:25:07
{ "title": [ -0.0023441315, 0.007194519, -0.040527344, -0.039886475, 0.009117126, -0.016448975, -0.011962891, -0.009643555, -0.020950317, 0.029281616, -0.038726807, 0.018356323, 0.016937256, -0.013847351, -0.018478394, -0.0090408325, -0.007156372, ...
http://arxiv.org/abs/2305.03356v1
From Parse-Execute to Parse-Execute-Refine: Improving Semantic Parser for Complex Question Answering over Knowledge Base
Wangzhen Guo, Linyin Luo, Hanjiang Lai, Jian Yin
cs.CL, cs.AI, cs.LG
Parsing questions into executable logical forms has showed impressive results for knowledge-base question answering (KBQA). However, complex KBQA is a more challenging task that requires to perform complex multi-step reasoning. Recently, a new semantic parser called KoPL has been proposed to explicitly model the reason...
http://arxiv.org/pdf/2305.03356v1
2023-05-05T08:20:09
2023-05-05T08:20:09
{ "title": [ -0.016616821, -0.0040245056, -0.03463745, -0.07733154, -0.00038051605, 0.012664795, -0.053253174, 0.0046157837, 0.0027561188, 0.027236938, -0.030334473, 0.021942139, 0.0002696514, 0.0063705444, -0.05596924, -0.0031776428, -0.008270264, ...
http://arxiv.org/abs/2305.03353v2
MindGames: Targeting Theory of Mind in Large Language Models with Dynamic Epistemic Modal Logic
Damien Sileo, Antoine Lernould
cs.CL, cs.AI, 68T01, 68T27, 68T50, I.2.7
Theory of Mind (ToM) is a critical component of intelligence but its assessment remains the subject of heated debates. Prior research applied human ToM assessments to natural language processing models using either human-created standardized tests or rule-based templates. However, these methods primarily focus on simpl...
http://arxiv.org/pdf/2305.03353v2
2023-11-07T08:53:59
2023-05-05T08:14:48
{ "title": [ -0.020233154, -0.02468872, -0.023025513, -0.053588867, -0.015151978, 0.0008587837, -0.03527832, 0.002986908, -0.0015192032, 0.025283813, -0.009635925, 0.00060892105, 0.01676941, -0.027679443, -0.02746582, 0.0033512115, -0.019927979, ...
http://arxiv.org/abs/2305.03336v1
QCRI at SemEval-2023 Task 3: News Genre, Framing and Persuasion Techniques Detection using Multilingual Models
Maram Hasanain, Ahmed Oumar El-Shangiti, Rabindra Nath Nandi, Preslav Nakov, Firoj Alam
cs.CL, cs.AI, cs.CY, 68T50, F.2.2; I.2.7
Misinformation spreading in mainstream and social media has been misleading users in different ways. Manual detection and verification efforts by journalists and fact-checkers can no longer cope with the great scale and quick spread of misleading information. This motivated research and industry efforts to develop syst...
http://arxiv.org/pdf/2305.03336v1
2023-05-05T07:40:41
2023-05-05T07:40:41
{ "title": [ -0.013206482, 0.001326561, -0.038208008, 0.014122009, -0.0052375793, -0.024414062, -0.010757446, -0.039855957, -0.060943604, 0.04425049, 0.023727417, 0.027130127, -0.010360718, -0.0362854, -0.007286072, -0.001991272, -0.03237915, 0.0...
http://arxiv.org/abs/2305.03742v1
Improved Logical Reasoning of Language Models via Differentiable Symbolic Programming
Hanlin Zhang, Jiani Huang, Ziyang Li, Mayur Naik, Eric Xing
cs.AI, cs.CL, cs.LG
Pre-trained large language models (LMs) struggle to perform logical reasoning reliably despite advances in scale and compositionality. In this work, we tackle this challenge through the lens of symbolic programming. We propose DSR-LM, a Differentiable Symbolic Reasoning framework where pre-trained LMs govern the percep...
http://arxiv.org/pdf/2305.03742v1
2023-05-05T07:24:46
2023-05-05T07:24:46
{ "title": [ -0.055603027, -0.024154663, -0.029388428, -0.08239746, -0.01360321, 0.0340271, -0.09576416, 0.015045166, -0.011672974, 0.019470215, -0.029571533, 0.007549286, -0.009300232, -0.00881958, -0.02420044, -0.009803772, -0.025527954, 0.0361...
http://arxiv.org/abs/2305.03319v2
HiPool: Modeling Long Documents Using Graph Neural Networks
Irene Li, Aosong Feng, Dragomir Radev, Rex Ying
cs.CL
Encoding long sequences in Natural Language Processing (NLP) is a challenging problem. Though recent pretraining language models achieve satisfying performances in many NLP tasks, they are still restricted by a pre-defined maximum length, making them challenging to be extended to longer sequences. So some recent works ...
http://arxiv.org/pdf/2305.03319v2
2023-05-15T03:48:36
2023-05-05T06:58:24
{ "title": [ -0.049804688, -0.011131287, -0.0395813, -0.041778564, -0.020996094, 0.0048713684, -0.059631348, 0.0005259514, -0.004787445, 0.018600464, -0.000043570995, -0.011131287, -0.0050811768, -0.027297974, -0.028244019, -0.004886627, -0.007949829...
http://arxiv.org/abs/2305.03314v1
Block the Label and Noise: An N-Gram Masked Speller for Chinese Spell Checking
Haiyun Yang
cs.CL
Recently, Chinese Spell Checking(CSC), a task to detect erroneous characters in a sentence and correct them, has attracted extensive interest because of its wide applications in various NLP tasks. Most of the existing methods have utilized BERT to extract semantic information for CSC task. However, these methods direct...
http://arxiv.org/pdf/2305.03314v1
2023-05-05T06:43:56
2023-05-05T06:43:56
{ "title": [ -0.03237915, 0.0074386597, -0.01966858, 0.007675171, -0.010940552, -0.019302368, 0.0008277893, -0.02897644, 0.007736206, 0.0345459, -0.047698975, -0.010093689, 0.062927246, -0.024154663, -0.03201294, -0.010231018, -0.028320312, 0.039...
http://arxiv.org/abs/2305.03300v1
LLM-RM at SemEval-2023 Task 2: Multilingual Complex NER using XLM-RoBERTa
Rahul Mehta, Vasudeva Varma
cs.CL
Named Entity Recognition(NER) is a task of recognizing entities at a token level in a sentence. This paper focuses on solving NER tasks in a multilingual setting for complex named entities. Our team, LLM-RM participated in the recently organized SemEval 2023 task, Task 2: MultiCoNER II,Multilingual Complex Named Entity...
http://arxiv.org/pdf/2305.03300v1
2023-05-05T06:05:45
2023-05-05T06:05:45
{ "title": [ -0.03164673, -0.033081055, -0.012840271, -0.031021118, -0.046691895, 0.03652954, -0.09613037, -0.00077581406, -0.022109985, 0.03265381, 0.018569946, 0.018692017, 0.036865234, -0.058807373, 0.033843994, 0.038513184, -0.03366089, 0.047...
http://arxiv.org/abs/2305.03299v1
Open Information Extraction via Chunks
Kuicai Dong, Aixin Sun, Jung-Jae Kim, Xiaoli Li
cs.CL, cs.AI
Open Information Extraction (OIE) aims to extract relational tuples from open-domain sentences. Existing OIE systems split a sentence into tokens and recognize token spans as tuple relations and arguments. We instead propose Sentence as Chunk sequence (SaC) and recognize chunk spans as tuple relations and arguments. We...
http://arxiv.org/pdf/2305.03299v1
2023-05-05T06:03:54
2023-05-05T06:03:54
{ "title": [ -0.030380249, -0.013168335, -0.06311035, -0.095336914, 0.00049066544, 0.012275696, -0.06628418, 0.017608643, -0.016937256, 0.042663574, -0.03125, 0.01763916, -0.013671875, -0.004890442, -0.032592773, 0.0017642975, 0.018463135, 0.0143...
http://arxiv.org/abs/2305.03296v1
TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition
Weixiang Zhao, Yanyan Zhao, Shilong Wang, Bing Qin
cs.CL
Emotion Support Conversation (ESC) is an emerging and challenging task with the goal of reducing the emotional distress of people. Previous attempts fail to maintain smooth transitions between utterances in ESC because they ignore to grasp the fine-grained transition information at each dialogue turn. To solve this pro...
http://arxiv.org/pdf/2305.03296v1
2023-05-05T05:50:26
2023-05-05T05:50:26
{ "title": [ -0.0010604858, -0.014152527, -0.029373169, -0.050354004, -0.0016736984, 0.00088977814, -0.030044556, -0.0057411194, -0.015914917, 0.027664185, -0.0018529892, 0.002500534, 0.037353516, -0.04525757, -0.003894806, -0.006969452, -0.010528564...
http://arxiv.org/abs/2305.03287v1
Low-Resource Multi-Granularity Academic Function Recognition Based on Multiple Prompt Knowledge
Jiawei Liu, Zi Xiong, Yi Jiang, Yongqiang Ma, Wei Lu, Yong Huang, Qikai Cheng
cs.CL, cs.AI
Fine-tuning pre-trained language models (PLMs), e.g., SciBERT, generally requires large numbers of annotated data to achieve state-of-the-art performance on a range of NLP tasks in the scientific domain. However, obtaining the fine-tune data for scientific NLP task is still challenging and expensive. Inspired by recent...
http://arxiv.org/pdf/2305.03287v1
2023-05-05T05:32:50
2023-05-05T05:32:50
{ "title": [ -0.013244629, 0.018249512, -0.04901123, -0.06677246, 0.0049362183, 0.0050621033, -0.08782959, -0.028533936, 0.014129639, 0.023849487, -0.01146698, -0.0038890839, 0.0012311935, -0.023605347, -0.048950195, -0.023666382, 0.0013933182, 0...
http://arxiv.org/abs/2305.03276v2
Expository Text Generation: Imitate, Retrieve, Paraphrase
Nishant Balepur, Jie Huang, Kevin Chen-Chuan Chang
cs.CL
Expository documents are vital resources for conveying complex information to readers. Despite their usefulness, writing expository text by hand is a challenging process that requires careful content planning, obtaining facts from multiple sources, and the ability to clearly synthesize these facts. To ease these burden...
http://arxiv.org/pdf/2305.03276v2
2023-10-23T01:32:10
2023-05-05T04:26:29
{ "title": [ -0.023010254, -0.011428833, -0.039520264, -0.04458618, -0.023849487, -0.002943039, -0.035888672, -0.0050697327, -0.020019531, 0.06335449, -0.015914917, 0.01826477, 0.036987305, -0.013633728, -0.033172607, 0.017227173, -0.009719849, 0...
http://arxiv.org/abs/2305.03268v1
Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework
Ruochen Zhao, Xingxuan Li, Shafiq Joty, Chengwei Qin, Lidong Bing
cs.CL
As large language models (LLMs) have become the norm in NLP, demonstrating good performance in generation and reasoning tasks, one of its most fatal disadvantages is the lack of factual correctness. Generating unfactual texts not only leads to lower performances but also degrades the trust and validity of their applica...
http://arxiv.org/pdf/2305.03268v1
2023-05-05T03:49:14
2023-05-05T03:49:14
{ "title": [ 0.00043082237, -0.024719238, -0.03845215, -0.07165527, 0.0107040405, -0.0073127747, -0.033843994, -0.008163452, -0.017288208, 0.017807007, -0.010353088, -0.0007238388, -0.009399414, -0.026565552, -0.04006958, -0.008804321, -0.012825012, ...
http://arxiv.org/abs/2305.03262v1
Rescue Conversations from Dead-ends: Efficient Exploration for Task-oriented Dialogue Policy Optimization
Yangyang Zhao, Zhenyu Wang, Mehdi Dastani, Shihan Wang
cs.HC, cs.CL
Training a dialogue policy using deep reinforcement learning requires a lot of exploration of the environment. The amount of wasted invalid exploration makes their learning inefficient. In this paper, we find and define an important reason for the invalid exploration: dead-ends. When a conversation enters a dead-end st...
http://arxiv.org/pdf/2305.03262v1
2023-05-05T03:28:49
2023-05-05T03:28:49
{ "title": [ -0.014953613, -0.009613037, -0.017471313, -0.057403564, -0.0022830963, 0.00088071823, -0.03253174, -0.002046585, -0.017944336, 0.017745972, -0.0023117065, 0.012542725, -0.013259888, -0.026733398, -0.016326904, -0.00907135, -0.017211914, ...
http://arxiv.org/abs/2305.03256v1
Stylized Data-to-Text Generation: A Case Study in the E-Commerce Domain
Liqiang Jing, Xuemeng Song, Xuming Lin, Zhongzhou Zhao, Wei Zhou, Liqiang Nie
cs.CL
Existing data-to-text generation efforts mainly focus on generating a coherent text from non-linguistic input data, such as tables and attribute-value pairs, but overlook that different application scenarios may require texts of different styles. Inspired by this, we define a new task, namely stylized data-to-text gene...
http://arxiv.org/pdf/2305.03256v1
2023-05-05T03:02:41
2023-05-05T03:02:41
{ "title": [ -0.029037476, -0.015853882, -0.04623413, -0.056732178, -0.007537842, -0.0023899078, -0.052825928, 0.0074386597, 0.0005402565, 0.045440674, -0.018341064, 0.012260437, 0.016571045, 0.0010986328, -0.05267334, -0.0076828003, -0.011566162, ...
http://arxiv.org/abs/2305.03253v1
VicunaNER: Zero/Few-shot Named Entity Recognition using Vicuna
Bin Ji
cs.CL
Large Language Models (LLMs, e.g., ChatGPT) have shown impressive zero- and few-shot capabilities in Named Entity Recognition (NER). However, these models can only be accessed via online APIs, which may cause data leak and non-reproducible problems. In this paper, we propose VicunaNER, a zero/few-shot NER framework bas...
http://arxiv.org/pdf/2305.03253v1
2023-05-05T02:46:22
2023-05-05T02:46:22
{ "title": [ -0.019439697, -0.017578125, -0.059570312, -0.061340332, 0.004714966, -0.00995636, -0.03579712, 0.0070114136, 0.010025024, 0.028640747, -0.029205322, 0.0008955002, 0.028900146, -0.009841919, -0.036865234, 0.0015983582, -0.0077819824, ...
http://arxiv.org/abs/2305.03237v1
Out-of-Domain Intent Detection Considering Multi-turn Dialogue Contexts
Hao Lang, Yinhe Zheng, Binyuan Hui, Fei Huang, Yongbin Li
cs.CL, cs.AI, cs.LG
Out-of-Domain (OOD) intent detection is vital for practical dialogue systems, and it usually requires considering multi-turn dialogue contexts. However, most previous OOD intent detection approaches are limited to single dialogue turns. In this paper, we introduce a context-aware OOD intent detection (Caro) framework t...
http://arxiv.org/pdf/2305.03237v1
2023-05-05T01:39:21
2023-05-05T01:39:21
{ "title": [ 0.029876709, 0.01398468, -0.031311035, -0.03543091, 0.029647827, -0.032806396, 0.00856781, -0.016067505, 0.009048462, -0.00087070465, -0.008308411, 0.018539429, -0.010673523, -0.033935547, -0.0119018555, -0.018569946, -0.04147339, 0....
http://arxiv.org/abs/2305.03236v1
A Survey on Out-of-Distribution Detection in NLP
Hao Lang, Yinhe Zheng, Yixuan Li, Jian Sun, Fei Huang, Yongbin Li
cs.CL, cs.AI, cs.LG
Out-of-distribution (OOD) detection is essential for the reliable and safe deployment of machine learning systems in the real world. Great progress has been made over the past years. This paper presents the first review of recent advances in OOD detection with a particular focus on natural language processing approache...
http://arxiv.org/pdf/2305.03236v1
2023-05-05T01:38:49
2023-05-05T01:38:49
{ "title": [ 0.010597229, 0.0044555664, -0.007888794, -0.03375244, 0.024002075, -0.005432129, 0.0045318604, -0.026763916, 0.02432251, 0.006652832, -0.042022705, 0.007572174, -0.011749268, -0.014640808, -0.038391113, -0.015563965, -0.04550171, 0.0...
http://arxiv.org/abs/2305.03232v2
Neuromodulation Gated Transformer
Kobe Knowles, Joshua Bensemann, Diana Benavides-Prado, Vithya Yogarajan, Michael Witbrock, Gillian Dobbie, Yang Chen
cs.CL
We introduce a novel architecture, the Neuromodulation Gated Transformer (NGT), which is a simple implementation of neuromodulation in transformers via a multiplicative effect. We compare it to baselines and show that it results in the best average performance on the SuperGLUE benchmark validation sets.
http://arxiv.org/pdf/2305.03232v2
2023-05-11T21:28:30
2023-05-05T01:23:22
{ "title": [ 0.04650879, 0.035095215, -0.01576233, -0.0045814514, -0.0031108856, -0.04525757, 0.0062217712, -0.06488037, -0.024719238, 0.0335083, 0.009819031, -0.032287598, 0.054748535, -0.062683105, 0.029525757, 0.01322937, 0.0140686035, 0.03567...
http://arxiv.org/abs/2305.03210v2
AttentionViz: A Global View of Transformer Attention
Catherine Yeh, Yida Chen, Aoyu Wu, Cynthia Chen, Fernanda Viégas, Martin Wattenberg
cs.HC, cs.CL, cs.CV, cs.LG
Transformer models are revolutionizing machine learning, but their inner workings remain mysterious. In this work, we present a new visualization technique designed to help researchers understand the self-attention mechanism in transformers that allows these models to learn rich, contextual relationships between elemen...
http://arxiv.org/pdf/2305.03210v2
2023-08-09T06:24:55
2023-05-04T23:46:49
{ "title": [ 0.0024280548, 0.0021324158, -0.040924072, -0.052612305, 0.021469116, -0.013687134, -0.015602112, -0.025772095, -0.0032577515, 0.03591919, -0.028335571, -0.0042762756, 0.022979736, -0.024658203, -0.033477783, 0.014472961, 0.0027561188, ...
http://arxiv.org/abs/2305.03207v1
Investigating Lexical Sharing in Multilingual Machine Translation for Indian Languages
Sonal Sannigrahi, Rachel Bawden
cs.CL, cs.AI
Multilingual language models have shown impressive cross-lingual transfer ability across a diverse set of languages and tasks. To improve the cross-lingual ability of these models, some strategies include transliteration and finer-grained segmentation into characters as opposed to subwords. In this work, we investigate...
http://arxiv.org/pdf/2305.03207v1
2023-05-04T23:35:15
2023-05-04T23:35:15
{ "title": [ -0.047027588, -0.009483337, -0.027557373, -0.04449463, -0.02142334, 0.014678955, -0.05053711, 0.013244629, -0.027999878, 0.050231934, -0.033111572, 0.03866577, 0.0095825195, -0.0065727234, -0.031707764, 0.004940033, -0.027404785, 0.0...
http://arxiv.org/abs/2305.03204v1
VideoOFA: Two-Stage Pre-Training for Video-to-Text Generation
Xilun Chen, Lili Yu, Wenhan Xiong, Barlas Oğuz, Yashar Mehdad, Wen-tau Yih
cs.CV, cs.CL
We propose a new two-stage pre-training framework for video-to-text generation tasks such as video captioning and video question answering: A generative encoder-decoder model is first jointly pre-trained on massive image-text data to learn fundamental vision-language concepts, and then adapted to video data in an inter...
http://arxiv.org/pdf/2305.03204v1
2023-05-04T23:27:21
2023-05-04T23:27:21
{ "title": [ -0.01259613, -0.011810303, -0.025894165, -0.05807495, -0.0087890625, -0.00008738041, -0.02796936, -0.011489868, -0.0013885498, 0.05038452, -0.039733887, 0.039123535, 0.007484436, 0.009094238, -0.04373169, -0.005153656, -0.010566711, ...
http://arxiv.org/abs/2305.03201v1
Enhancing Pashto Text Classification using Language Processing Techniques for Single And Multi-Label Analysis
Mursal Dawodi, Jawid Ahmad Baktash
cs.CL, cs.LG
Text classification has become a crucial task in various fields, leading to a significant amount of research on developing automated text classification systems for national and international languages. However, there is a growing need for automated text classification systems that can handle local languages. This stud...
http://arxiv.org/pdf/2305.03201v1
2023-05-04T23:11:31
2023-05-04T23:11:31
{ "title": [ 0.010925293, -0.004421234, -0.02218628, -0.043518066, -0.0078086853, -0.00036740303, -0.024017334, -0.013336182, 0.01436615, 0.009529114, -0.026824951, 0.0009880066, -0.020599365, -0.023284912, -0.03881836, -0.013961792, -0.023712158, ...
http://arxiv.org/abs/2305.03200v1
Employing Hybrid Deep Neural Networks on Dari Speech
Jawid Ahmad Baktash, Mursal Dawodi
eess.AS, cs.CL, cs.SD
This paper is an extension of our previous conference paper. In recent years, there has been a growing interest among researchers in developing and improving speech recognition systems to facilitate and enhance human-computer interaction. Today, Automatic Speech Recognition (ASR) systems have become ubiquitous, used in...
http://arxiv.org/pdf/2305.03200v1
2023-05-04T23:10:53
2023-05-04T23:10:53
{ "title": [ -0.007083893, -0.002937317, -0.021575928, -0.054748535, -0.0007658005, -0.005554199, -0.047027588, -0.0036754608, -0.008583069, 0.033416748, -0.023834229, -0.028274536, 0.031951904, -0.035705566, -0.03314209, -0.0067749023, -0.014724731,...
http://arxiv.org/abs/2305.03737v1
Tuning Traditional Language Processing Approaches for Pashto Text Classification
Jawid Ahmad Baktash, Mursal Dawodi, Mohammad Zarif Joya, Nematullah Hassanzada
cs.CL, cs.AI, cs.LG
Today text classification becomes critical task for concerned individuals for numerous purposes. Hence, several researches have been conducted to develop automatic text classification for national and international languages. However, the need for an automatic text categorization system for local languages is felt. The...
http://arxiv.org/pdf/2305.03737v1
2023-05-04T22:57:45
2023-05-04T22:57:45
{ "title": [ -0.0043182373, 0.010421753, -0.0135650635, -0.03353882, -0.004562378, 0.00031232834, -0.020965576, -0.024505615, 0.024627686, 0.01537323, -0.045776367, -0.0056419373, -0.018356323, -0.011955261, -0.05783081, -0.022766113, -0.017471313, ...
http://arxiv.org/abs/2305.03195v1
Gpt-4: A Review on Advancements and Opportunities in Natural Language Processing
Jawid Ahmad Baktash, Mursal Dawodi
cs.CL, cs.AI, cs.CY
Generative Pre-trained Transformer 4 (GPT-4) is the fourth-generation language model in the GPT series, developed by OpenAI, which promises significant advancements in the field of natural language processing (NLP). In this research article, we have discussed the features of GPT-4, its potential applications, and the c...
http://arxiv.org/pdf/2305.03195v1
2023-05-04T22:46:43
2023-05-04T22:46:43
{ "title": [ -0.009216309, -0.0076179504, -0.024841309, -0.05871582, 0.022888184, 0.009552002, -0.026748657, -0.015853882, 0.007575989, 0.017822266, 0.0012130737, 0.006084442, -0.015037537, -0.033325195, -0.039367676, 0.02519226, -0.016418457, 0....
http://arxiv.org/abs/2305.03144v1
Influence of various text embeddings on clustering performance in NLP
Rohan Saha
cs.LG, cs.CL, cs.IR
With the advent of e-commerce platforms, reviews are crucial for customers to assess the credibility of a product. The star ratings do not always match the review text written by the customer. For example, a three star rating (out of five) may be incongruous with the review text, which may be more suitable for a five s...
http://arxiv.org/pdf/2305.03144v1
2023-05-04T20:53:19
2023-05-04T20:53:19
{ "title": [ -0.007843018, -0.0039482117, -0.025924683, -0.022644043, 0.0012054443, -0.016555786, -0.0070152283, -0.008590698, -0.008460999, 0.03286743, -0.041809082, 0.03741455, 0.006275177, 0.0013923645, -0.024490356, 0.005542755, -0.021102905, ...
http://arxiv.org/abs/2305.03132v2
The Role of Global and Local Context in Named Entity Recognition
Arthur Amalvy, Vincent Labatut, Richard Dufour
cs.CL
Pre-trained transformer-based models have recently shown great performance when applied to Named Entity Recognition (NER). As the complexity of their self-attention mechanism prevents them from processing long documents at once, these models are usually applied in a sequential fashion. Such an approach unfortunately on...
http://arxiv.org/pdf/2305.03132v2
2023-05-30T21:26:33
2023-05-04T20:22:18
{ "title": [ -0.011009216, 0.0036735535, -0.049560547, -0.08276367, 0.03213501, -0.0036144257, -0.028945923, 0.015792847, 0.0037593842, 0.02772522, -0.037322998, 0.013710022, 0.009346008, -0.0043754578, -0.031204224, 0.0031280518, -0.00060224533, ...
http://arxiv.org/abs/2305.03130v2
Chain-of-Skills: A Configurable Model for Open-domain Question Answering
Kaixin Ma, Hao Cheng, Yu Zhang, Xiaodong Liu, Eric Nyberg, Jianfeng Gao
cs.CL
The retrieval model is an indispensable component for real-world knowledge-intensive tasks, e.g., open-domain question answering (ODQA). As separate retrieval skills are annotated for different datasets, recent work focuses on customized methods, limiting the model transferability and scalability. In this work, we prop...
http://arxiv.org/pdf/2305.03130v2
2023-05-26T17:19:58
2023-05-04T20:19:39
{ "title": [ 0.007259369, -0.003932953, -0.02067566, -0.062072754, 0.0029144287, 0.0006427765, -0.021881104, -0.012123108, 0.0020828247, 0.023666382, -0.010826111, -0.0015211105, 0.00021636486, -0.009841919, -0.047027588, -0.013931274, -0.010223389, ...
http://arxiv.org/abs/2305.09782v1
Analysis of Visual Question Answering Algorithms with attention model
Param Ahir, Hiteishi M. Diwanji
cs.CV, cs.AI, cs.CL, cs.LG
Visual question answering (VQA) usesimage processing algorithms to process the image and natural language processing methods to understand and answer the question. VQA is helpful to a visually impaired person, can be used for the security surveillance system and online chatbots that learn from the web. It uses NLP meth...
http://arxiv.org/pdf/2305.09782v1
2023-05-04T20:10:37
2023-05-04T20:10:37
{ "title": [ -0.0057411194, 0.017471313, -0.019805908, -0.018722534, 0.000644207, -0.013656616, -0.002462387, -0.013977051, 0.037231445, 0.045440674, -0.035095215, -0.015960693, 0.021270752, 0.00881958, -0.08001709, 0.0031547546, 0.009315491, 0.0...
http://arxiv.org/abs/2305.03117v2
Are Human Explanations Always Helpful? Towards Objective Evaluation of Human Natural Language Explanations
Bingsheng Yao, Prithviraj Sen, Lucian Popa, James Hendler, Dakuo Wang
cs.CL
Human-annotated labels and explanations are critical for training explainable NLP models. However, unlike human-annotated labels whose quality is easier to calibrate (e.g., with a majority vote), human-crafted free-form explanations can be quite subjective. Before blindly using them as ground truth to train ML models, ...
http://arxiv.org/pdf/2305.03117v2
2023-05-22T05:20:04
2023-05-04T19:31:50
{ "title": [ -0.016311646, -0.013595581, -0.023452759, -0.06518555, -0.011299133, 0.0012741089, -0.042022705, -0.009918213, -0.018295288, 0.015220642, -0.0077400208, 0.0039863586, -0.019836426, -0.013435364, -0.043762207, -0.0007381439, -0.018676758,...
http://arxiv.org/abs/2305.03111v3
Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs
Jinyang Li, Binyuan Hui, Ge Qu, Jiaxi Yang, Binhua Li, Bowen Li, Bailin Wang, Bowen Qin, Rongyu Cao, Ruiying Geng, Nan Huo, Xuanhe Zhou, Chenhao Ma, Guoliang Li, Kevin C. C. Chang, Fei Huang, Reynold Cheng, Yongbin Li
cs.CL
Text-to-SQL parsing, which aims at converting natural language instructions into executable SQLs, has gained increasing attention in recent years. In particular, Codex and ChatGPT have shown impressive results in this task. However, most of the prevalent benchmarks, i.e., Spider, and WikiSQL, focus on database schema w...
http://arxiv.org/pdf/2305.03111v3
2023-11-15T04:56:25
2023-05-04T19:02:29
{ "title": [ -0.03451538, -0.0069122314, -0.058654785, -0.036956787, -0.00818634, -0.01889038, -0.038146973, 0.03475952, 0.020523071, 0.024612427, -0.04925537, 0.01939392, 0.0029411316, -0.03488159, 0.013763428, -0.021759033, 0.014450073, 0.00674...
http://arxiv.org/abs/2305.03101v1
Hybrid Transducer and Attention based Encoder-Decoder Modeling for Speech-to-Text Tasks
Yun Tang, Anna Y. Sun, Hirofumi Inaguma, Xinyue Chen, Ning Dong, Xutai Ma, Paden D. Tomasello, Juan Pino
cs.CL, cs.SD, eess.AS
Transducer and Attention based Encoder-Decoder (AED) are two widely used frameworks for speech-to-text tasks. They are designed for different purposes and each has its own benefits and drawbacks for speech-to-text tasks. In order to leverage strengths of both modeling methods, we propose a solution by combining Transdu...
http://arxiv.org/pdf/2305.03101v1
2023-05-04T18:34:50
2023-05-04T18:34:50
{ "title": [ -0.011177063, 0.015808105, -0.02444458, -0.043060303, -0.010772705, -0.0015487671, -0.033721924, -0.0034828186, 0.016036987, 0.056915283, -0.032104492, -0.013183594, 0.052642822, -0.0053863525, -0.050872803, 0.008392334, -0.014251709, ...
http://arxiv.org/abs/2305.03092v2
Curating corpora with classifiers: A case study of clean energy sentiment online
Michael V. Arnold, Peter Sheridan Dodds, Christopher M. Danforth
cs.CL, cs.CY
Well curated, large-scale corpora of social media posts containing broad public opinion offer an alternative data source to complement traditional surveys. While surveys are effective at collecting representative samples and are capable of achieving high accuracy, they can be both expensive to run and lag public opinio...
http://arxiv.org/pdf/2305.03092v2
2023-05-10T02:39:55
2023-05-04T18:15:45
{ "title": [ 0.02407837, 0.018295288, -0.009429932, 0.0028648376, 0.010398865, -0.019134521, 0.025344849, -0.029525757, -0.018630981, 0.022460938, -0.057678223, 0.011367798, -0.030929565, 0.0058021545, -0.049713135, -0.05731201, -0.030380249, 0.0...
http://arxiv.org/abs/2305.05589v1
DomainInv: Domain Invariant Fine Tuning and Adversarial Label Correction For QA Domain Adaptation
Anant Khandelwal
cs.CL
Existing Question Answering (QA) systems limited by the capability of answering questions from unseen domain or any out-of-domain distributions making them less reliable for deployment to real scenarios. Most importantly all the existing QA domain adaptation methods are either based on generating synthetic data or pseu...
http://arxiv.org/pdf/2305.05589v1
2023-05-04T18:13:17
2023-05-04T18:13:17
{ "title": [ -0.02407837, -0.006374359, -0.050109863, -0.02507019, 0.010681152, -0.030014038, -0.003446579, -0.0033721924, -0.014350891, 0.051086426, 0.016143799, -0.004463196, 0.015129089, -0.03274536, -0.047943115, 0.009849548, -0.016326904, 0....
http://arxiv.org/abs/2305.03088v1
Modeling What-to-ask and How-to-ask for Answer-unaware Conversational Question Generation
Xuan Long Do, Bowei Zou, Shafiq Joty, Anh Tai Tran, Liangming Pan, Nancy F. Chen, Ai Ti Aw
cs.CL, cs.AI
Conversational Question Generation (CQG) is a critical task for machines to assist humans in fulfilling their information needs through conversations. The task is generally cast into two different settings: answer-aware and answer-unaware. While the former facilitates the models by exposing the expected answer, the lat...
http://arxiv.org/pdf/2305.03088v1
2023-05-04T18:06:48
2023-05-04T18:06:48
{ "title": [ -0.0132751465, -0.0037956238, -0.039794922, -0.037139893, -0.015586853, -0.0096206665, -0.034179688, -0.00447464, 0.01322937, 0.0152282715, -0.020263672, 0.01335907, -0.020584106, -0.01675415, -0.0546875, -0.007888794, -0.010620117, ...
http://arxiv.org/abs/2305.03048v2
Personalize Segment Anything Model with One Shot
Renrui Zhang, Zhengkai Jiang, Ziyu Guo, Shilin Yan, Junting Pan, Xianzheng Ma, Hao Dong, Peng Gao, Hongsheng Li
cs.CV, cs.AI, cs.CL, cs.LG, cs.MM
Driven by large-data pre-training, Segment Anything Model (SAM) has been demonstrated as a powerful and promptable framework, revolutionizing the segmentation models. Despite the generality, customizing SAM for specific visual concepts without man-powered prompting is under explored, e.g., automatically segmenting your...
http://arxiv.org/pdf/2305.03048v2
2023-10-04T01:15:21
2023-05-04T17:59:36
{ "title": [ 0.024215698, -0.017974854, -0.011375427, -0.062347412, -0.0463562, 0.0020065308, -0.040130615, 0.017715454, -0.053894043, 0.052734375, -0.008705139, -0.0010194778, -0.018310547, -0.020263672, 0.02848816, -0.039123535, 0.025787354, -0...
http://arxiv.org/abs/2305.03047v1
Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision
Zhiqing Sun, Yikang Shen, Qinhong Zhou, Hongxin Zhang, Zhenfang Chen, David Cox, Yiming Yang, Chuang Gan
cs.LG, cs.AI, cs.CL, cs.CY
Recent AI-assistant agents, such as ChatGPT, predominantly rely on supervised fine-tuning (SFT) with human annotations and reinforcement learning from human feedback (RLHF) to align the output of large language models (LLMs) with human intentions, ensuring they are helpful, ethical, and reliable. However, this dependen...
http://arxiv.org/pdf/2305.03047v1
2023-05-04T17:59:28
2023-05-04T17:59:28
{ "title": [ -0.029953003, -0.010040283, -0.049041748, -0.072143555, -0.009170532, 0.0079422, -0.08062744, 0.022445679, -0.00054454803, 0.05090332, 0.016235352, 0.0048103333, 0.008605957, -0.015655518, -0.052703857, 0.022521973, -0.0011529922, 0....
http://arxiv.org/abs/2305.03029v1
What changes when you randomly choose BPE merge operations? Not much
Jonne Sälevä, Constantine Lignos
cs.CL, cs.AI
We introduce three simple randomized variants of byte pair encoding (BPE) and explore whether randomizing the selection of merge operations substantially affects a downstream machine translation task. We focus on translation into morphologically rich languages, hypothesizing that this task may show sensitivity to the m...
http://arxiv.org/pdf/2305.03029v1
2023-05-04T17:53:27
2023-05-04T17:53:27
{ "title": [ -0.007911682, 0.033325195, -0.06488037, -0.04321289, 0.016448975, -0.008056641, -0.026550293, 0.033599854, -0.027023315, 0.041900635, -0.01727295, 0.014930725, 0.018356323, -0.024398804, 0.013069153, -0.0082473755, 0.005378723, 0.053...
http://arxiv.org/abs/2305.03025v1
Panda LLM: Training Data and Evaluation for Open-Sourced Chinese Instruction-Following Large Language Models
Fangkai Jiao, Bosheng Ding, Tianze Luo, Zhanfeng Mo
cs.CL, cs.AI
This project focuses on enhancing open-source large language models through instruction-tuning and providing comprehensive evaluations of their performance. We explore how various training data factors, such as quantity, quality, and linguistic distribution, influence the performance of instruction-tuned models trained...
http://arxiv.org/pdf/2305.03025v1
2023-05-04T17:49:09
2023-05-04T17:49:09
{ "title": [ -0.041656494, 0.007007599, -0.048950195, -0.016998291, -0.0132369995, -0.015403748, -0.04373169, -0.029220581, -0.022323608, 0.019104004, -0.020141602, 0.013465881, 0.0006737709, -0.038635254, -0.007083893, -0.012718201, -0.019927979, ...
http://arxiv.org/abs/2305.03010v1
Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence
Haoran Li, Mingshi Xu, Yangqiu Song
cs.CL, cs.CR
Sentence-level representations are beneficial for various natural language processing tasks. It is commonly believed that vector representations can capture rich linguistic properties. Currently, large language models (LMs) achieve state-of-the-art performance on sentence embedding. However, some recent works suggest t...
http://arxiv.org/pdf/2305.03010v1
2023-05-04T17:31:41
2023-05-04T17:31:41
{ "title": [ 0.0066719055, -0.018066406, 0.003856659, 0.006690979, -0.03274536, -0.031402588, 0.02861023, -0.01499176, -0.022140503, 0.02848816, -0.036071777, 0.006259918, 0.025772095, -0.018630981, -0.017745972, -0.030059814, -0.0001424551, 0.02...
http://arxiv.org/abs/2305.03007v1
NatCS: Eliciting Natural Customer Support Dialogues
James Gung, Emily Moeng, Wesley Rose, Arshit Gupta, Yi Zhang, Saab Mansour
cs.CL
Despite growing interest in applications based on natural customer support conversations, there exist remarkably few publicly available datasets that reflect the expected characteristics of conversations in these settings. Existing task-oriented dialogue datasets, which were collected to benchmark dialogue systems main...
http://arxiv.org/pdf/2305.03007v1
2023-05-04T17:25:24
2023-05-04T17:25:24
{ "title": [ -0.013000488, -0.02520752, -0.035827637, -0.059906006, -0.0071525574, -0.0024414062, -0.042877197, 0.0042037964, 0.013710022, 0.019897461, -0.045288086, -0.00047969818, -0.021224976, -0.0052833557, -0.05923462, -0.024810791, -0.011878967...