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/2303.16618v2
Personalised Language Modelling of Screen Characters Using Rich Metadata Annotations
Sebastian Vincent, Rowanne Sumner, Alice Dowek, Charlotte Blundell, Emily Preston, Chris Bayliss, Chris Oakley, Carolina Scarton
cs.CL, cs.AI, cs.LG
Language models that are sensitive to external context can more effectively capture the speaking patterns of individuals with specific characteristics or in particular environments. However, obtaining and leveraging such annotations can be challenging. In this work, we show how to leverage rich character and film annot...
http://arxiv.org/pdf/2303.16618v2
2023-08-11T10:01:35
2023-03-29T12:19:23
{ "title": [ -0.024353027, -0.022232056, -0.026260376, -0.06100464, -0.020523071, 0.017562866, -0.046020508, 0.022567749, 0.013687134, 0.058776855, 0.023788452, -0.0046463013, 0.036132812, -0.034606934, -0.048950195, 0.03591919, -0.004753113, 0.0...
http://arxiv.org/abs/2305.06148v1
A semi-automatic method for document classification in the shipping industry
Narayanan Arvind
cs.CL, cs.LG
In the shipping industry, document classification plays a crucial role in ensuring that the necessary documents are properly identified and processed for customs clearance. OCR technology is being used to automate the process of document classification, which involves identifying important documents such as Commercial ...
http://arxiv.org/pdf/2305.06148v1
2023-03-29T10:00:43
2023-03-29T10:00:43
{ "title": [ -0.024597168, 0.0076942444, -0.043640137, 0.0022621155, -0.011482239, -0.038208008, -0.014053345, -0.025314331, 0.008514404, 0.02078247, -0.066833496, 0.00712204, -0.0020828247, -0.023040771, -0.01828003, -0.022003174, 0.005508423, 0...
http://arxiv.org/abs/2303.16537v2
LMExplainer: a Knowledge-Enhanced Explainer for Language Models
Zichen Chen, Ambuj K Singh, Misha Sra
cs.CL
Large language models (LLMs) such as GPT-4 are very powerful and can process different kinds of natural language processing (NLP) tasks. However, it can be difficult to interpret the results due to the multi-layer nonlinear model structure and millions of parameters. A lack of clarity and understanding of how the langu...
http://arxiv.org/pdf/2303.16537v2
2023-08-03T23:23:43
2023-03-29T08:59:44
{ "title": [ -0.037963867, -0.021026611, -0.034362793, -0.07788086, -0.005874634, 0.01637268, -0.06976318, 0.013641357, -0.0070991516, 0.016067505, -0.009277344, 0.00079250336, 0.014099121, -0.016647339, -0.025924683, 0.0077438354, -0.011482239, ...
http://arxiv.org/abs/2303.16528v1
Building a Knowledge Graph of Distributed Ledger Technologies
Lukas König, Sebastian Neumaier
cs.CL, cs.AI, cs.CR
Distributed ledger systems have become more prominent and successful in recent years, with a focus on blockchains and cryptocurrency. This has led to various misunderstandings about both the technology itself and its capabilities, as in many cases blockchain and cryptocurrency is used synonymously and other application...
http://arxiv.org/pdf/2303.16528v1
2023-03-29T08:34:01
2023-03-29T08:34:01
{ "title": [ -0.06439209, -0.0029067993, -0.0993042, -0.0635376, -0.005607605, -0.0021591187, -0.05999756, 0.029953003, -0.02748108, 0.035095215, -0.028457642, -0.008483887, -0.012573242, -0.010543823, -0.035888672, -0.025466919, 0.020080566, -0....
http://arxiv.org/abs/2303.16445v3
Larger Probes Tell a Different Story: Extending Psycholinguistic Datasets Via In-Context Learning
Namrata Shivagunde, Vladislav Lialin, Anna Rumshisky
cs.CL
Language model probing is often used to test specific capabilities of models. However, conclusions from such studies may be limited when the probing benchmarks are small and lack statistical power. In this work, we introduce new, larger datasets for negation (NEG-1500-SIMP) and role reversal (ROLE-1500) inspired by psy...
http://arxiv.org/pdf/2303.16445v3
2023-11-14T17:24:28
2023-03-29T04:00:53
{ "title": [ 0.021453857, -0.01171875, -0.01109314, -0.04345703, -0.0024528503, -0.028762817, -0.0007171631, -0.0036201477, -0.019805908, 0.024230957, -0.04852295, 0.02041626, -0.0178833, -0.010681152, -0.022369385, -0.036743164, -0.010185242, 0....
http://arxiv.org/abs/2303.17579v2
Multimodal Image-Text Matching Improves Retrieval-based Chest X-Ray Report Generation
Jaehwan Jeong, Katherine Tian, Andrew Li, Sina Hartung, Fardad Behzadi, Juan Calle, David Osayande, Michael Pohlen, Subathra Adithan, Pranav Rajpurkar
cs.CL, cs.AI, cs.CV
Automated generation of clinically accurate radiology reports can improve patient care. Previous report generation methods that rely on image captioning models often generate incoherent and incorrect text due to their lack of relevant domain knowledge, while retrieval-based attempts frequently retrieve reports that are...
http://arxiv.org/pdf/2303.17579v2
2023-05-02T21:03:40
2023-03-29T04:00:47
{ "title": [ -0.011451721, 0.025756836, -0.047821045, -0.02532959, 0.00027108192, -0.016113281, -0.022354126, -0.028747559, 0.0021858215, 0.043151855, -0.020080566, 0.0074424744, 0.0013685226, -0.01965332, -0.047698975, 0.0018129349, -0.002937317, ...
http://arxiv.org/abs/2303.16434v1
TaskMatrix.AI: Completing Tasks by Connecting Foundation Models with Millions of APIs
Yaobo Liang, Chenfei Wu, Ting Song, Wenshan Wu, Yan Xia, Yu Liu, Yang Ou, Shuai Lu, Lei Ji, Shaoguang Mao, Yun Wang, Linjun Shou, Ming Gong, Nan Duan
cs.AI, cs.CL
Artificial Intelligence (AI) has made incredible progress recently. On the one hand, advanced foundation models like ChatGPT can offer powerful conversation, in-context learning and code generation abilities on a broad range of open-domain tasks. They can also generate high-level solution outlines for domain-specific t...
http://arxiv.org/pdf/2303.16434v1
2023-03-29T03:30:38
2023-03-29T03:30:38
{ "title": [ -0.016906738, -0.01512146, -0.078125, -0.05505371, -0.011116028, -0.03451538, -0.037384033, -0.0016212463, 0.000110566616, 0.020568848, -0.010040283, 0.008148193, -0.014671326, -0.0256958, -0.02267456, -0.033416748, 0.0016641617, 0.0...
http://arxiv.org/abs/2303.16421v1
ChatGPT is a Knowledgeable but Inexperienced Solver: An Investigation of Commonsense Problem in Large Language Models
Ning Bian, Xianpei Han, Le Sun, Hongyu Lin, Yaojie Lu, Ben He
cs.CL
Large language models (LLMs) such as ChatGPT and GPT-4 have made significant progress in NLP. However, their ability to memorize, represent, and leverage commonsense knowledge has been a well-known pain point for LLMs. It remains unclear that: (1) Can GPTs effectively answer commonsense questions? (2) Are GPTs knowledg...
http://arxiv.org/pdf/2303.16421v1
2023-03-29T03:05:43
2023-03-29T03:05:43
{ "title": [ 0.013175964, -0.02494812, 0.008354187, -0.009429932, -0.038604736, -0.0036449432, -0.018157959, -0.028289795, 0.006576538, 0.022964478, -0.027359009, 0.0043640137, 0.007675171, -0.01763916, -0.016281128, -0.00040388107, -0.022613525, ...
http://arxiv.org/abs/2303.16416v2
Zero-shot Clinical Entity Recognition using ChatGPT
Yan Hu, Iqra Ameer, Xu Zuo, Xueqing Peng, Yujia Zhou, Zehan Li, Yiming Li, Jianfu Li, Xiaoqian Jiang, Hua Xu
cs.CL
In this study, we investigated the potential of ChatGPT, a large language model developed by OpenAI, for the clinical named entity recognition task defined in the 2010 i2b2 challenge, in a zero-shot setting with two different prompt strategies. We compared its performance with GPT-3 in a similar zero-shot setting, as w...
http://arxiv.org/pdf/2303.16416v2
2023-05-15T18:17:17
2023-03-29T02:46:18
{ "title": [ -0.002855301, -0.004924774, -0.036468506, -0.06451416, 0.007247925, 0.0040664673, -0.038635254, 0.008041382, -0.005077362, 0.03274536, 0.0019330978, 0.0069999695, 0.0025749207, -0.019577026, -0.027679443, 0.022598267, -0.01512146, 0....
http://arxiv.org/abs/2303.16406v1
Hierarchical Video-Moment Retrieval and Step-Captioning
Abhay Zala, Jaemin Cho, Satwik Kottur, Xilun Chen, Barlas Oğuz, Yasher Mehdad, Mohit Bansal
cs.CV, cs.AI, cs.CL, cs.LG
There is growing interest in searching for information from large video corpora. Prior works have studied relevant tasks, such as text-based video retrieval, moment retrieval, video summarization, and video captioning in isolation, without an end-to-end setup that can jointly search from video corpora and generate summ...
http://arxiv.org/pdf/2303.16406v1
2023-03-29T02:33:54
2023-03-29T02:33:54
{ "title": [ -0.0060157776, 0.015167236, -0.06341553, -0.0552063, 0.026367188, -0.024108887, -0.0018568039, -0.013404846, -0.00017750263, 0.06994629, -0.019180298, 0.032287598, 0.014915466, 0.0050201416, -0.061462402, 0.009857178, -0.005279541, 0...
http://arxiv.org/abs/2303.16352v1
ChatGPT or academic scientist? Distinguishing authorship with over 99% accuracy using off-the-shelf machine learning tools
Heather Desaire, Aleesa E. Chua, Madeline Isom, Romana Jarosova, David Hua
cs.LG, cs.CL
ChatGPT has enabled access to AI-generated writing for the masses, and within just a few months, this product has disrupted the knowledge economy, initiating a culture shift in the way people work, learn, and write. The need to discriminate human writing from AI is now both critical and urgent, particularly in domains ...
http://arxiv.org/pdf/2303.16352v1
2023-03-28T23:16:00
2023-03-28T23:16:00
{ "title": [ -0.034729004, 0.009239197, -0.05392456, -0.008178711, -0.03302002, -0.014450073, -0.038879395, -0.031280518, -0.042175293, 0.032196045, 0.019073486, 0.014289856, -0.008598328, -0.032592773, -0.004421234, -0.021438599, -0.023880005, 0...
http://arxiv.org/abs/2303.16342v2
Language-Guided Audio-Visual Source Separation via Trimodal Consistency
Reuben Tan, Arijit Ray, Andrea Burns, Bryan A. Plummer, Justin Salamon, Oriol Nieto, Bryan Russell, Kate Saenko
cs.CV, cs.AI, cs.CL
We propose a self-supervised approach for learning to perform audio source separation in videos based on natural language queries, using only unlabeled video and audio pairs as training data. A key challenge in this task is learning to associate the linguistic description of a sound-emitting object to its visual featur...
http://arxiv.org/pdf/2303.16342v2
2023-09-23T15:05:33
2023-03-28T22:45:40
{ "title": [ -0.005218506, -0.0069084167, -0.05529785, -0.06707764, 0.010635376, -0.003490448, -0.047821045, 0.01725769, -0.013076782, 0.041229248, -0.022628784, 0.005191803, 0.008605957, -0.021697998, -0.02923584, -0.009979248, 0.011726379, 0.01...
http://arxiv.org/abs/2303.16281v2
A Perspectival Mirror of the Elephant: Investigating Language Bias on Google, ChatGPT, Wikipedia, and YouTube
Queenie Luo, Michael J. Puett, Michael D. Smith
cs.CY, cs.AI, cs.CL, cs.LG, cs.SI
Contrary to Google Search's mission of delivering information from "many angles so you can form your own understanding of the world," we find that Google and its most prominent returned results - Wikipedia and YouTube - simply reflect a narrow set of cultural stereotypes tied to the search language for complex topics l...
http://arxiv.org/pdf/2303.16281v2
2023-05-23T07:14:46
2023-03-28T19:49:58
{ "title": [ -0.01638794, -0.010131836, -0.04864502, -0.0025234222, -0.0020046234, -0.038238525, 0.012359619, -0.009025574, -0.02508545, 0.020874023, -0.017349243, 0.025650024, 0.0030460358, -0.019836426, -0.0035743713, 0.0013036728, -0.024307251, ...
http://arxiv.org/abs/2303.16275v1
Writing Assistants Should Model Social Factors of Language
Vivek Kulkarni, Vipul Raheja
cs.CL, cs.HC, I.2.7; K.4.2; H.5.0
Intelligent writing assistants powered by large language models (LLMs) are more popular today than ever before, but their further widespread adoption is precluded by sub-optimal performance. In this position paper, we argue that a major reason for this sub-optimal performance and adoption is a singular focus on the inf...
http://arxiv.org/pdf/2303.16275v1
2023-03-28T19:38:57
2023-03-28T19:38:57
{ "title": [ -0.01399231, -0.012069702, 0.002155304, 0.005218506, -0.07678223, 0.0026245117, -0.047668457, -0.004447937, -0.028182983, 0.05908203, -0.0062942505, -0.020126343, 0.023284912, -0.01499176, 0.007980347, 0.003709793, 0.00063610077, 0.0...
http://arxiv.org/abs/2303.16256v1
Scalable handwritten text recognition system for lexicographic sources of under-resourced languages and alphabets
Jan Idziak, Artjoms Šeļa, Michał Woźniak, Albert Leśniak, Joanna Byszuk, Maciej Eder
cs.CL, cs.CV
The paper discusses an approach to decipher large collections of handwritten index cards of historical dictionaries. Our study provides a working solution that reads the cards, and links their lemmas to a searchable list of dictionary entries, for a large historical dictionary entitled the Dictionary of the 17th- and 1...
http://arxiv.org/pdf/2303.16256v1
2023-03-28T19:06:27
2023-03-28T19:06:27
{ "title": [ -0.02180481, 0.0013809204, -0.038513184, -0.044067383, -0.021774292, -0.013092041, -0.040130615, 0.006351471, 0.006793976, 0.035614014, -0.09277344, 0.026931763, 0.006828308, 0.006752014, -0.04031372, -0.029571533, 0.016021729, 0.029...
http://arxiv.org/abs/2303.16252v1
Zero-Shot Generalizable End-to-End Task-Oriented Dialog System using Context Summarization and Domain Schema
Adib Mosharrof, M. H. Maqbool, A. B. Siddique
cs.CL, cs.LG
Task-oriented dialog systems empower users to accomplish their goals by facilitating intuitive and expressive natural language interactions. State-of-the-art approaches in task-oriented dialog systems formulate the problem as a conditional sequence generation task and fine-tune pre-trained causal language models in the...
http://arxiv.org/pdf/2303.16252v1
2023-03-28T18:56:31
2023-03-28T18:56:31
{ "title": [ 0.014564514, -0.008583069, -0.070373535, -0.10961914, 0.019210815, -0.013877869, -0.046691895, 0.04989624, 0.022369385, 0.061309814, 0.02432251, -0.012275696, 0.018295288, -0.020080566, -0.055511475, 0.016448975, -0.0040512085, 0.029...
http://arxiv.org/abs/2303.16199v2
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
Renrui Zhang, Jiaming Han, Chris Liu, Peng Gao, Aojun Zhou, Xiangfei Hu, Shilin Yan, Pan Lu, Hongsheng Li, Yu Qiao
cs.CV, cs.AI, cs.CL, cs.LG, cs.MM
We present LLaMA-Adapter, a lightweight adaption method to efficiently fine-tune LLaMA into an instruction-following model. Using 52K self-instruct demonstrations, LLaMA-Adapter only introduces 1.2M learnable parameters upon the frozen LLaMA 7B model, and costs less than one hour for fine-tuning on 8 A100 GPUs. Specifi...
http://arxiv.org/pdf/2303.16199v2
2023-06-14T17:31:32
2023-03-28T17:59:12
{ "title": [ -0.01676941, -0.011161804, -0.033996582, -0.072265625, -0.0046157837, 0.012481689, -0.060302734, 0.009498596, 0.01637268, 0.03741455, -0.009681702, 0.0070724487, 0.012962341, -0.008224487, -0.04916382, 0.02658081, -0.012908936, 0.039...
http://arxiv.org/abs/2303.16173v1
Towards Countering Essentialism through Social Bias Reasoning
Emily Allaway, Nina Taneja, Sarah-Jane Leslie, Maarten Sap
cs.CL
Essentialist beliefs (i.e., believing that members of the same group are fundamentally alike) play a central role in social stereotypes and can lead to harm when left unchallenged. In our work, we conduct exploratory studies into the task of countering essentialist beliefs (e.g., ``liberals are stupid''). Drawing on pr...
http://arxiv.org/pdf/2303.16173v1
2023-03-28T17:34:59
2023-03-28T17:34:59
{ "title": [ 0.037719727, 0.0013589859, -0.022323608, -0.050109863, 0.019439697, -0.028778076, 0.016174316, -0.016616821, -0.014587402, 0.031921387, -0.014503479, 0.0019893646, -0.021942139, -0.016021729, -0.045318604, 0.0019254684, -0.0074882507, ...
http://arxiv.org/abs/2303.16166v4
When Good and Reproducible Results are a Giant with Feet of Clay: The Importance of Software Quality in NLP
Sara Papi, Marco Gaido, Andrea Pilzer, Matteo Negri
cs.CL, cs.AI
Despite its crucial role in research experiments, code correctness is often presumed only on the basis of the perceived quality of results. This assumption comes with the risk of erroneous outcomes and potentially misleading findings. To address this issue, we posit that the current focus on reproducibility should go h...
http://arxiv.org/pdf/2303.16166v4
2023-08-15T21:53:56
2023-03-28T17:28:52
{ "title": [ -0.029449463, -0.025756836, -0.032714844, -0.052093506, -0.012611389, -0.0025596619, -0.037109375, -0.0018758774, -0.012283325, 0.011260986, -0.028717041, 0.0075645447, -0.016708374, -0.024353027, -0.03463745, -0.0068206787, -0.02859497,...
http://arxiv.org/abs/2303.16755v2
Training Language Models with Language Feedback at Scale
Jérémy Scheurer, Jon Ander Campos, Tomasz Korbak, Jun Shern Chan, Angelica Chen, Kyunghyun Cho, Ethan Perez
cs.CL, cs.AI, cs.LG
Pretrained language models often generate outputs that are not in line with human preferences, such as harmful text or factually incorrect summaries. Recent work approaches the above issues by learning from a simple form of human feedback: comparisons between pairs of model-generated outputs. However, comparison feedba...
http://arxiv.org/pdf/2303.16755v2
2023-04-09T14:10:44
2023-03-28T17:04:15
{ "title": [ -0.041259766, -0.008178711, -0.029754639, -0.049987793, -0.014831543, 0.013633728, -0.07080078, -0.0029754639, 0.0012302399, 0.021957397, -0.035736084, 0.0007724762, 0.007762909, -0.016601562, -0.028961182, 0.0047416687, -0.00048303604, ...
http://arxiv.org/abs/2303.16133v1
Exposing and Addressing Cross-Task Inconsistency in Unified Vision-Language Models
Adyasha Maharana, Amita Kamath, Christopher Clark, Mohit Bansal, Aniruddha Kembhavi
cs.CV, cs.AI, cs.CL, cs.LG
As general purpose vision models get increasingly effective at a wide set of tasks, it is imperative that they be consistent across the tasks they support. Inconsistent AI models are considered brittle and untrustworthy by human users and are more challenging to incorporate into larger systems that take dependencies on...
http://arxiv.org/pdf/2303.16133v1
2023-03-28T16:57:12
2023-03-28T16:57:12
{ "title": [ -0.03756714, -0.009811401, -0.03866577, -0.04248047, -0.011245728, -0.008430481, -0.054473877, -0.026489258, 0.005218506, 0.03475952, -0.017684937, -0.0038452148, 0.0102005005, -0.01776123, -0.03363037, -0.0037384033, -0.0016469955, ...
http://arxiv.org/abs/2303.16104v1
Hallucinations in Large Multilingual Translation Models
Nuno M. Guerreiro, Duarte Alves, Jonas Waldendorf, Barry Haddow, Alexandra Birch, Pierre Colombo, André F. T. Martins
cs.CL
Large-scale multilingual machine translation systems have demonstrated remarkable ability to translate directly between numerous languages, making them increasingly appealing for real-world applications. However, when deployed in the wild, these models may generate hallucinated translations which have the potential to ...
http://arxiv.org/pdf/2303.16104v1
2023-03-28T16:17:59
2023-03-28T16:17:59
{ "title": [ -0.035858154, 0.0069389343, -0.00598526, -0.036315918, -0.017654419, 0.010307312, -0.03930664, -0.01322937, 0.0045814514, 0.052520752, -0.012878418, 0.011550903, 0.020812988, -0.011856079, -0.050750732, 0.03048706, -0.0077209473, 0.0...
http://arxiv.org/abs/2303.16749v1
Improving Code Generation by Training with Natural Language Feedback
Angelica Chen, Jérémy Scheurer, Tomasz Korbak, Jon Ander Campos, Jun Shern Chan, Samuel R. Bowman, Kyunghyun Cho, Ethan Perez
cs.SE, cs.AI, cs.CL, cs.LG
The potential for pre-trained large language models (LLMs) to use natural language feedback at inference time has been an exciting recent development. We build upon this observation by formalizing an algorithm for learning from natural language feedback at training time instead, which we call Imitation learning from La...
http://arxiv.org/pdf/2303.16749v1
2023-03-28T16:15:31
2023-03-28T16:15:31
{ "title": [ -0.042114258, -0.036315918, -0.030044556, -0.051483154, -0.015533447, 0.014411926, -0.05810547, 0.012458801, 0.011657715, 0.008720398, -0.028137207, 0.0040779114, -0.020721436, -0.010658264, -0.026275635, -0.015731812, -0.014579773, ...
http://arxiv.org/abs/2303.16098v1
Carolina: a General Corpus of Contemporary Brazilian Portuguese with Provenance, Typology and Versioning Information
Maria Clara Ramos Morales Crespo, Maria Lina de Souza Jeannine Rocha, Mariana Lourenço Sturzeneker, Felipe Ribas Serras, Guilherme Lamartine de Mello, Aline Silva Costa, Mayara Feliciano Palma, Renata Morais Mesquita, Raquel de Paula Guets, Mariana Marques da Silva, Marcelo Finger, Maria Clara Paixão de Sousa, Cristian...
cs.CL, cs.AI, 68T50, I.2.7
This paper presents the first publicly available version of the Carolina Corpus and discusses its future directions. Carolina is a large open corpus of Brazilian Portuguese texts under construction using web-as-corpus methodology enhanced with provenance, typology, versioning, and text integrality. The corpus aims at b...
http://arxiv.org/pdf/2303.16098v1
2023-03-28T16:09:40
2023-03-28T16:09:40
{ "title": [ -0.072509766, -0.023956299, -0.08081055, -0.028778076, -0.03805542, 0.017440796, -0.06854248, 0.032836914, -0.0061912537, 0.048828125, -0.0065460205, -0.0074272156, 0.0317688, -0.04144287, -0.020065308, 0.00034046173, -0.012191772, 0...
http://arxiv.org/abs/2303.16028v1
Synthetically generated text for supervised text analysis
Andrew Halterman
cs.CL
Supervised text models are a valuable tool for political scientists but present several obstacles to their use, including the expense of hand-labeling documents, the difficulty of retrieving rare relevant documents for annotation, and copyright and privacy concerns involved in sharing annotated documents. This article ...
http://arxiv.org/pdf/2303.16028v1
2023-03-28T14:55:13
2023-03-28T14:55:13
{ "title": [ -0.03048706, -0.003932953, -0.036499023, -0.058410645, 0.00069236755, 0.010139465, -0.041259766, -0.009315491, -0.007434845, 0.02078247, -0.027862549, 0.017700195, -0.02229309, -0.0073547363, -0.038085938, -0.0039634705, -0.028640747, ...
http://arxiv.org/abs/2303.15987v2
Sentiment Analysis Dataset in Moroccan Dialect: Bridging the Gap Between Arabic and Latin Scripted dialect
Mouad Jbel, Imad Hafidi, Abdulmutallib Metrane
cs.CL
Sentiment analysis, the automated process of determining emotions or opinions expressed in text, has seen extensive exploration in the field of natural language processing. However, one aspect that has remained underrepresented is the sentiment analysis of the Moroccan dialect, which boasts a unique linguistic landscap...
http://arxiv.org/pdf/2303.15987v2
2023-11-06T18:38:55
2023-03-28T14:02:42
{ "title": [ -0.018066406, -0.008399963, -0.021453857, -0.02192688, -0.018157959, 0.000032424927, -0.03656006, -0.0032367706, 0.02468872, 0.032043457, -0.03567505, -0.026504517, 0.018295288, -0.029205322, -0.05593872, -0.0066719055, 0.02357483, 0...
http://arxiv.org/abs/2303.15973v1
Do Neural Topic Models Really Need Dropout? Analysis of the Effect of Dropout in Topic Modeling
Suman Adhya, Avishek Lahiri, Debarshi Kumar Sanyal
cs.CL, cs.LG
Dropout is a widely used regularization trick to resolve the overfitting issue in large feedforward neural networks trained on a small dataset, which performs poorly on the held-out test subset. Although the effectiveness of this regularization trick has been extensively studied for convolutional neural networks, there...
http://arxiv.org/pdf/2303.15973v1
2023-03-28T13:45:39
2023-03-28T13:45:39
{ "title": [ -0.013595581, 0.019607544, -0.027053833, -0.026000977, 0.007320404, 0.0017766953, -0.035369873, -0.047424316, 0.025665283, 0.026184082, -0.014152527, 0.030273438, -0.012054443, -0.015701294, -0.037139893, 0.030700684, -0.012573242, 0...
http://arxiv.org/abs/2303.15870v1
A Multi-Granularity Matching Attention Network for Query Intent Classification in E-commerce Retrieval
Chunyuan Yuan, Yiming Qiu, Mingming Li, Haiqing Hu, Songlin Wang, Sulong Xu
cs.IR, cs.CL
Query intent classification, which aims at assisting customers to find desired products, has become an essential component of the e-commerce search. Existing query intent classification models either design more exquisite models to enhance the representation learning of queries or explore label-graph and multi-task to ...
http://arxiv.org/pdf/2303.15870v1
2023-03-28T10:25:17
2023-03-28T10:25:17
{ "title": [ 0.022109985, 0.024307251, -0.048431396, -0.044067383, 0.02293396, -0.030944824, -0.017471313, -0.008766174, 0.041015625, 0.04537964, 0.0023956299, -0.01084137, 0.046844482, -0.02319336, -0.042175293, 0.023208618, -0.0023326874, 0.015...
http://arxiv.org/abs/2303.15846v1
Soft-prompt tuning to predict lung cancer using primary care free-text Dutch medical notes
Auke Elfrink, Iacopo Vagliano, Ameen Abu-Hanna, Iacer Calixto
cs.CL, cs.AI, cs.LG, I.2.7
We investigate different natural language processing (NLP) approaches based on contextualised word representations for the problem of early prediction of lung cancer using free-text patient medical notes of Dutch primary care physicians. Because lung cancer has a low prevalence in primary care, we also address the prob...
http://arxiv.org/pdf/2303.15846v1
2023-03-28T09:36:53
2023-03-28T09:36:53
{ "title": [ 0.018875122, -0.032196045, -0.0019464493, -0.051696777, -0.038360596, 0.0025348663, -0.04638672, 0.026641846, -0.07122803, 0.03161621, -0.042144775, 0.0070266724, -0.009635925, -0.020248413, 0.0012931824, -0.05505371, 0.044036865, -0...
http://arxiv.org/abs/2303.16757v1
How can Deep Learning Retrieve the Write-Missing Additional Diagnosis from Chinese Electronic Medical Record For DRG
Shaohui Liu, Xien Liu, Ji Wu
cs.CL, cs.AI
The purpose of write-missing diagnosis detection is to find diseases that have been clearly diagnosed from medical records but are missed in the discharge diagnosis. Unlike the definition of missed diagnosis, the write-missing diagnosis is clearly manifested in the medical record without further reasoning. The write-mi...
http://arxiv.org/pdf/2303.16757v1
2023-03-28T08:56:31
2023-03-28T08:56:31
{ "title": [ -0.0013341904, 0.009231567, -0.04449463, -0.03616333, -0.013809204, -0.014137268, -0.029159546, -0.0021038055, -0.030929565, 0.020385742, -0.057434082, 0.014137268, -0.0053367615, -0.027008057, 0.025314331, -0.055419922, 0.020462036, ...
http://arxiv.org/abs/2303.16779v1
Language Models Trained on Media Diets Can Predict Public Opinion
Eric Chu, Jacob Andreas, Stephen Ansolabehere, Deb Roy
cs.CL, cs.LG
Public opinion reflects and shapes societal behavior, but the traditional survey-based tools to measure it are limited. We introduce a novel approach to probe media diet models -- language models adapted to online news, TV broadcast, or radio show content -- that can emulate the opinions of subpopulations that have con...
http://arxiv.org/pdf/2303.16779v1
2023-03-28T06:08:25
2023-03-28T06:08:25
{ "title": [ 0.033447266, -0.0022678375, -0.0035247803, -0.0063552856, -0.048065186, -0.022338867, -0.0074043274, -0.010253906, -0.01777649, 0.056243896, -0.027389526, -0.009033203, 0.014305115, -0.029541016, -0.0034999847, -0.0121536255, 0.001374244...
http://arxiv.org/abs/2303.15727v1
Evaluation of ChatGPT for NLP-based Mental Health Applications
Bishal Lamichhane
cs.CL, cs.AI
Large language models (LLM) have been successful in several natural language understanding tasks and could be relevant for natural language processing (NLP)-based mental health application research. In this work, we report the performance of LLM-based ChatGPT (with gpt-3.5-turbo backend) in three text-based mental heal...
http://arxiv.org/pdf/2303.15727v1
2023-03-28T04:47:43
2023-03-28T04:47:43
{ "title": [ -0.010772705, -0.011123657, -0.015327454, -0.031433105, 0.008018494, 0.009292603, -0.021896362, -0.012489319, 0.0015611649, 0.033966064, -0.011001587, 0.007156372, 0.0070533752, -0.029968262, -0.03781128, 0.013000488, -0.025924683, 0...
http://arxiv.org/abs/2303.15714v4
Explicit Planning Helps Language Models in Logical Reasoning
Hongyu Zhao, Kangrui Wang, Mo Yu, Hongyuan Mei
cs.CL, cs.AI, cs.LG
Language models have been shown to perform remarkably well on a wide range of natural language processing tasks. In this paper, we propose LEAP, a novel system that uses language models to perform multi-step logical reasoning and incorporates explicit planning into the inference procedure. Explicit planning enables the...
http://arxiv.org/pdf/2303.15714v4
2023-11-07T18:12:20
2023-03-28T03:55:03
{ "title": [ 0.012580872, -0.019470215, -0.027404785, -0.08734131, -0.0513916, 0.0030460358, -0.097839355, 0.007434845, -0.03488159, 0.022338867, -0.035217285, -0.002035141, 0.03237915, -0.0423584, 0.047912598, -0.0061149597, 0.015350342, 0.02343...
http://arxiv.org/abs/2303.15708v2
Bias or Diversity? Unraveling Fine-Grained Thematic Discrepancy in U.S. News Headlines
Jinsheng Pan, Weihong Qi, Zichen Wang, Hanjia Lyu, Jiebo Luo
cs.CL
There is a broad consensus that news media outlets incorporate ideological biases in their news articles. However, prior studies on measuring the discrepancies among media outlets and further dissecting the origins of thematic differences suffer from small sample sizes and limited scope and granularity. In this study, ...
http://arxiv.org/pdf/2303.15708v2
2023-05-06T03:57:30
2023-03-28T03:31:37
{ "title": [ -0.01776123, 0.0077438354, -0.039215088, -0.026168823, 0.034210205, -0.012489319, 0.009391785, -0.024887085, -0.031341553, 0.031585693, -0.022857666, 0.009750366, -0.023544312, 0.0047912598, -0.066589355, -0.036895752, -0.017425537, ...
http://arxiv.org/abs/2303.15705v1
Translate the Beauty in Songs: Jointly Learning to Align Melody and Translate Lyrics
Chengxi Li, Kai Fan, Jiajun Bu, Boxing Chen, Zhongqiang Huang, Zhi Yu
cs.CL, cs.SD, eess.AS
Song translation requires both translation of lyrics and alignment of music notes so that the resulting verse can be sung to the accompanying melody, which is a challenging problem that has attracted some interests in different aspects of the translation process. In this paper, we propose Lyrics-Melody Translation with...
http://arxiv.org/pdf/2303.15705v1
2023-03-28T03:17:59
2023-03-28T03:17:59
{ "title": [ -0.025939941, -0.016494751, -0.056854248, -0.013496399, -0.026763916, -0.033813477, -0.016357422, 0.012962341, -0.0178833, 0.04586792, -0.019363403, -0.010025024, 0.026107788, -0.032104492, -0.03817749, -0.020935059, 0.004840851, 0.0...
http://arxiv.org/abs/2304.02468v1
Comparative Analysis of CHATGPT and the evolution of language models
Oluwatosin Ogundare, Gustavo Quiros Araya
cs.CL
Interest in Large Language Models (LLMs) has increased drastically since the emergence of ChatGPT and the outstanding positive societal response to the ease with which it performs tasks in Natural Language Processing (NLP). The triumph of ChatGPT, however, is how it seamlessly bridges the divide between language genera...
http://arxiv.org/pdf/2304.02468v1
2023-03-28T03:11:28
2023-03-28T03:11:28
{ "title": [ -0.033691406, -0.008361816, -0.026275635, -0.019378662, -0.015777588, 0.0030975342, -0.03970337, -0.0013942719, -0.015525818, 0.044952393, -0.0075302124, 0.016967773, 0.0141067505, -0.01638794, -0.027160645, 0.027511597, -0.012954712, ...
http://arxiv.org/abs/2303.15697v1
Model and Evaluation: Towards Fairness in Multilingual Text Classification
Nankai Lin, Junheng He, Zhenghang Tang, Dong Zhou, Aimin Yang
cs.CL
Recently, more and more research has focused on addressing bias in text classification models. However, existing research mainly focuses on the fairness of monolingual text classification models, and research on fairness for multilingual text classification is still very limited. In this paper, we focus on the task of ...
http://arxiv.org/pdf/2303.15697v1
2023-03-28T03:00:01
2023-03-28T03:00:01
{ "title": [ -0.025100708, 0.0009045601, -0.036895752, -0.038330078, 0.011039734, -0.0029735565, -0.026535034, -0.013519287, 0.008430481, 0.03086853, -0.03543091, 0.010009766, -0.0016469955, -0.004432678, -0.05407715, -0.015075684, -0.03881836, 0...
http://arxiv.org/abs/2304.02746v1
Sejarah dan Perkembangan Teknik Natural Language Processing (NLP) Bahasa Indonesia: Tinjauan tentang sejarah, perkembangan teknologi, dan aplikasi NLP dalam bahasa Indonesia
Mukhlis Amien
cs.CL, 68T50, I.2.7
This study provides an overview of the history of the development of Natural Language Processing (NLP) in the context of the Indonesian language, with a focus on the basic technologies, methods, and practical applications that have been developed. This review covers developments in basic NLP technologies such as stemmi...
http://arxiv.org/pdf/2304.02746v1
2023-03-28T02:31:47
2023-03-28T02:31:47
{ "title": [ -0.016555786, -0.0211792, -0.017471313, 0.012107849, -0.042999268, -0.010040283, -0.017990112, 0.021881104, -0.03994751, 0.01133728, -0.045898438, 0.040618896, -0.027267456, -0.014328003, 0.024429321, -0.044677734, -0.021102905, 0.02...
http://arxiv.org/abs/2303.15682v1
Pre-training Transformers for Knowledge Graph Completion
Sanxing Chen, Hao Cheng, Xiaodong Liu, Jian Jiao, Yangfeng Ji, Jianfeng Gao
cs.CL, cs.LG
Learning transferable representation of knowledge graphs (KGs) is challenging due to the heterogeneous, multi-relational nature of graph structures. Inspired by Transformer-based pretrained language models' success on learning transferable representation for texts, we introduce a novel inductive KG representation model...
http://arxiv.org/pdf/2303.15682v1
2023-03-28T02:10:37
2023-03-28T02:10:37
{ "title": [ -0.015609741, -0.0051193237, -0.0501709, -0.10650635, 0.002506256, 0.017150879, -0.0748291, 0.0184021, -0.00440979, 0.031585693, -0.04058838, 0.026275635, 0.0038013458, 0.0056877136, -0.03024292, -0.0010328293, 0.0015611649, 0.052856...
http://arxiv.org/abs/2303.15662v2
ChatGPT4PCG Competition: Character-like Level Generation for Science Birds
Pittawat Taveekitworachai, Febri Abdullah, Mury F. Dewantoro, Ruck Thawonmas, Julian Togelius, Jochen Renz
cs.AI, cs.CL, I.2.7; I.2.8
This paper presents the first ChatGPT4PCG Competition at the 2023 IEEE Conference on Games. The objective of this competition is for participants to create effective prompts for ChatGPT--enabling it to generate Science Birds levels with high stability and character-like qualities--fully using their creativity as well a...
http://arxiv.org/pdf/2303.15662v2
2023-05-29T05:32:04
2023-03-28T01:07:38
{ "title": [ -0.030258179, -0.04043579, 0.002029419, -0.0014886856, -0.037902832, -0.0038642883, -0.003452301, 0.000115573406, -0.015991211, 0.035339355, 0.015281677, 0.00869751, -0.011474609, -0.046813965, -0.011985779, 0.0032405853, -0.026931763, ...
http://arxiv.org/abs/2303.15655v1
Joint embedding in Hierarchical distance and semantic representation learning for link prediction
Jin Liu, Jianye Chen, Chongfeng Fan, Fengyu Zhou
cs.CL
The link prediction task aims to predict missing entities or relations in the knowledge graph and is essential for the downstream application. Existing well-known models deal with this task by mainly focusing on representing knowledge graph triplets in the distance space or semantic space. However, they can not fully c...
http://arxiv.org/pdf/2303.15655v1
2023-03-28T00:42:29
2023-03-28T00:42:29
{ "title": [ 0.015327454, 0.008529663, -0.04827881, -0.07244873, 0.019088745, -0.017059326, -0.020477295, -0.005115509, -0.0011577606, 0.037872314, -0.0005931854, 0.016937256, 0.021270752, -0.02999878, -0.0072021484, 0.016235352, -0.008781433, 0....
http://arxiv.org/abs/2303.15647v1
Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning
Vladislav Lialin, Vijeta Deshpande, Anna Rumshisky
cs.CL
This paper presents a systematic overview and comparison of parameter-efficient fine-tuning methods covering over 40 papers published between February 2019 and February 2023. These methods aim to resolve the infeasibility and impracticality of fine-tuning large language models by only training a small set of parameters...
http://arxiv.org/pdf/2303.15647v1
2023-03-28T00:06:38
2023-03-28T00:06:38
{ "title": [ -0.023071289, -0.028198242, -0.032073975, -0.084350586, 0.019882202, 0.013259888, -0.03616333, 0.0037956238, -0.022628784, 0.034454346, -0.0020618439, 0.009147644, -0.00920105, -0.02923584, -0.039489746, 0.002500534, 0.0032024384, 0....
http://arxiv.org/abs/2303.15621v2
ChatGPT as a Factual Inconsistency Evaluator for Text Summarization
Zheheng Luo, Qianqian Xie, Sophia Ananiadou
cs.CL
The performance of text summarization has been greatly boosted by pre-trained language models. A main concern of existing methods is that most generated summaries are not factually inconsistent with their source documents. To alleviate the problem, many efforts have focused on developing effective factuality evaluation...
http://arxiv.org/pdf/2303.15621v2
2023-04-13T10:59:39
2023-03-27T22:30:39
{ "title": [ 0.0018806458, 0.00623703, -0.016143799, -0.04724121, -0.02104187, 0.007587433, -0.028884888, -0.0152282715, -0.031585693, 0.047943115, -0.025634766, 0.041412354, 0.020584106, 0.0039749146, -0.039886475, 0.004272461, -0.0385437, 0.062...
http://arxiv.org/abs/2303.15619v1
Typhoon: Towards an Effective Task-Specific Masking Strategy for Pre-trained Language Models
Muhammed Shahir Abdurrahman, Hashem Elezabi, Bruce Changlong Xu
cs.CL, cs.AI
Through exploiting a high level of parallelism enabled by graphics processing units, transformer architectures have enabled tremendous strides forward in the field of natural language processing. In a traditional masked language model, special MASK tokens are used to prompt our model to gather contextual information fr...
http://arxiv.org/pdf/2303.15619v1
2023-03-27T22:27:23
2023-03-27T22:27:23
{ "title": [ -0.029037476, -0.0030841827, -0.05706787, -0.06732178, 0.0055618286, -0.0028839111, -0.06896973, 0.012107849, 0.0079422, 0.028289795, 0.0028591156, 0.0032157898, 0.026565552, -0.033996582, -0.030548096, 0.02357483, -0.011985779, 0.03...
http://arxiv.org/abs/2303.16777v1
Not cool, calm or collected: Using emotional language to detect COVID-19 misinformation
Gabriel Asher, Phil Bohlman, Karsten Kleyensteuber
cs.CL, cs.LG, cs.SI
COVID-19 misinformation on social media platforms such as twitter is a threat to effective pandemic management. Prior works on tweet COVID-19 misinformation negates the role of semantic features common to twitter such as charged emotions. Thus, we present a novel COVID-19 misinformation model, which uses both a tweet e...
http://arxiv.org/pdf/2303.16777v1
2023-03-27T22:24:05
2023-03-27T22:24:05
{ "title": [ -0.0043258667, 0.0024967194, -0.06384277, -0.033447266, 0.03274536, -0.03427124, -0.0057754517, -0.010185242, -0.038726807, 0.027359009, -0.01878357, 0.02279663, 0.0069351196, -0.020187378, -0.008346558, -0.019546509, 0.008575439, 0....
http://arxiv.org/abs/2304.02017v6
Unlocking the Potential of ChatGPT: A Comprehensive Exploration of its Applications, Advantages, Limitations, and Future Directions in Natural Language Processing
Walid Hariri
cs.CL
Large language models have revolutionized the field of artificial intelligence and have been used in various applications. Among these models, ChatGPT (Chat Generative Pre-trained Transformer) has been developed by OpenAI, it stands out as a powerful tool that has been widely adopted. ChatGPT has been successfully appl...
http://arxiv.org/pdf/2304.02017v6
2023-11-15T18:34:35
2023-03-27T21:27:58
{ "title": [ 0.0075912476, -0.009513855, -0.004940033, -0.0206604, -0.01625061, -0.00497818, -0.021240234, -0.0073432922, 0.0018730164, 0.05621338, -0.019439697, 0.035614014, -0.0057411194, -0.0010137558, -0.035308838, 0.037078857, -0.019104004, ...
http://arxiv.org/abs/2303.15587v1
Linguistically Informed ChatGPT Prompts to Enhance Japanese-Chinese Machine Translation: A Case Study on Attributive Clauses
Wenshi Gu
cs.CL
In the field of Japanese-Chinese translation linguistics, the issue of correctly translating attributive clauses has persistently proven to be challenging. Present-day machine translation tools often fail to accurately translate attributive clauses from Japanese to Chinese. In light of this, this paper investigates the...
http://arxiv.org/pdf/2303.15587v1
2023-03-27T20:33:40
2023-03-27T20:33:40
{ "title": [ -0.016342163, -0.020355225, 0.021377563, 0.009803772, -0.047088623, 0.016159058, -0.044830322, -0.03152466, -0.018539429, 0.041168213, -0.010276794, 0.01159668, 0.04623413, -0.025299072, -0.01574707, 0.0102005005, -0.02671814, 0.0468...
http://arxiv.org/abs/2304.00008v3
On the Creativity of Large Language Models
Giorgio Franceschelli, Mirco Musolesi
cs.AI, cs.CL, cs.CY
Large Language Models (LLMs) are revolutionizing several areas of Artificial Intelligence. One of the most remarkable applications is creative writing, e.g., poetry or storytelling: the generated outputs are often of astonishing quality. However, a natural question arises: can LLMs be really considered creative? In thi...
http://arxiv.org/pdf/2304.00008v3
2023-07-09T18:00:02
2023-03-27T18:00:01
{ "title": [ -0.029922485, -0.011520386, -0.02330017, -0.059295654, 0.003610611, 0.014305115, -0.062438965, -0.0035171509, 0.008735657, 0.033691406, -0.010498047, -0.010147095, 0.0046920776, -0.0154953, -0.04232788, 0.01838684, -0.021484375, 0.05...
http://arxiv.org/abs/2303.15445v1
IRFL: Image Recognition of Figurative Language
Ron Yosef, Yonatan Bitton, Dafna Shahaf
cs.CL, cs.AI, cs.CV
Figures of speech such as metaphors, similes, and idioms allow language to be expressive, invoke emotion, and communicate abstract ideas that might otherwise be difficult to visualize. These figurative forms are often conveyed through multiple modes, such as text and images, and frequently appear in advertising, news, ...
http://arxiv.org/pdf/2303.15445v1
2023-03-27T17:59:55
2023-03-27T17:59:55
{ "title": [ 0.00031518936, 0.007255554, -0.012687683, -0.03024292, 0.009559631, -0.007331848, -0.01625061, -0.036376953, 0.023376465, 0.03375244, -0.02897644, -0.00016498566, 0.006175995, -0.009300232, -0.048461914, 0.004573822, -0.0132369995, 0...
http://arxiv.org/abs/2303.15441v1
Zero-shot Model Diagnosis
Jinqi Luo, Zhaoning Wang, Chen Henry Wu, Dong Huang, Fernando De la Torre
cs.CV, cs.AI, cs.CL, cs.LG
When it comes to deploying deep vision models, the behavior of these systems must be explicable to ensure confidence in their reliability and fairness. A common approach to evaluate deep learning models is to build a labeled test set with attributes of interest and assess how well it performs. However, creating a balan...
http://arxiv.org/pdf/2303.15441v1
2023-03-27T17:59:33
2023-03-27T17:59:33
{ "title": [ 0.03225708, 0.0017757416, -0.044189453, -0.04626465, -0.040039062, -0.03463745, -0.06536865, 0.0002682209, 0.031829834, 0.040283203, -0.015556335, -0.03555298, 0.021713257, -0.052642822, -0.0011005402, 0.023544312, 0.039794922, 0.009...
http://arxiv.org/abs/2303.15430v2
TextMI: Textualize Multimodal Information for Integrating Non-verbal Cues in Pre-trained Language Models
Md Kamrul Hasan, Md Saiful Islam, Sangwu Lee, Wasifur Rahman, Iftekhar Naim, Mohammed Ibrahim Khan, Ehsan Hoque
cs.CL, cs.LG
Pre-trained large language models have recently achieved ground-breaking performance in a wide variety of language understanding tasks. However, the same model can not be applied to multimodal behavior understanding tasks (e.g., video sentiment/humor detection) unless non-verbal features (e.g., acoustic and visual) can...
http://arxiv.org/pdf/2303.15430v2
2023-03-29T04:49:46
2023-03-27T17:54:32
{ "title": [ -0.036865234, -0.011184692, -0.03744507, -0.045166016, -0.03869629, 0.00042033195, -0.08880615, -0.017440796, -0.013435364, 0.027511597, -0.0206604, -0.005317688, 0.046295166, -0.023757935, -0.0054130554, -0.009231567, -0.006916046, ...
http://arxiv.org/abs/2303.15422v2
KPEval: Towards Fine-grained Semantic-based Evaluation of Keyphrase Extraction and Generation Systems
Di Wu, Da Yin, Kai-Wei Chang
cs.CL
Despite the significant advancements in keyphrase extraction and keyphrase generation methods, the predominant approach for evaluation only relies on exact matching with human references and disregards reference-free attributes. This scheme fails to recognize systems that generate keyphrases semantically equivalent to ...
http://arxiv.org/pdf/2303.15422v2
2023-11-01T05:00:38
2023-03-27T17:45:38
{ "title": [ -0.03768921, -0.01737976, -0.035583496, -0.03527832, -0.0007843971, 0.005458832, -0.03540039, -0.0009288788, 0.018997192, 0.048461914, -0.020584106, 0.01675415, 0.00756073, -0.00504303, -0.062805176, 0.0040016174, -0.03390503, 0.0401...
http://arxiv.org/abs/2303.15413v4
Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation
Susung Hong, Donghoon Ahn, Seungryong Kim
cs.CV, cs.CL, cs.GR, cs.LG
Existing score-distilling text-to-3D generation techniques, despite their considerable promise, often encounter the view inconsistency problem. One of the most notable issues is the Janus problem, where the most canonical view of an object (\textit{e.g}., face or head) appears in other views. In this work, we explore e...
http://arxiv.org/pdf/2303.15413v4
2023-11-08T03:15:47
2023-03-27T17:31:13
{ "title": [ -0.0368042, 0.016571045, -0.051727295, -0.0079956055, 0.002462387, -0.0027103424, -0.023452759, -0.03164673, 0.012992859, 0.026031494, -0.021759033, 0.0041542053, -0.0020065308, -0.025482178, -0.026367188, -0.0012693405, -0.008041382, ...
http://arxiv.org/abs/2303.15381v1
Causal schema induction for knowledge discovery
Michael Regan, Jena D. Hwang, Keisuke Sakaguchi, James Pustejovsky
cs.CL
Making sense of familiar yet new situations typically involves making generalizations about causal schemas, stories that help humans reason about event sequences. Reasoning about events includes identifying cause and effect relations shared across event instances, a process we refer to as causal schema induction. Stati...
http://arxiv.org/pdf/2303.15381v1
2023-03-27T16:55:49
2023-03-27T16:55:49
{ "title": [ -0.026184082, -0.01474762, -0.060302734, -0.06866455, 0.009887695, -0.006099701, -0.037322998, 0.011734009, -0.006515503, -0.018005371, -0.028900146, -0.010978699, -0.031921387, -0.046447754, 0.0023212433, -0.023788452, -0.008453369, ...
http://arxiv.org/abs/2303.15350v1
Improving Neural Topic Models with Wasserstein Knowledge Distillation
Suman Adhya, Debarshi Kumar Sanyal
cs.CL, cs.IR, cs.LG
Topic modeling is a dominant method for exploring document collections on the web and in digital libraries. Recent approaches to topic modeling use pretrained contextualized language models and variational autoencoders. However, large neural topic models have a considerable memory footprint. In this paper, we propose a...
http://arxiv.org/pdf/2303.15350v1
2023-03-27T16:07:44
2023-03-27T16:07:44
{ "title": [ -0.011566162, 0.013725281, -0.054992676, -0.042541504, 0.0116119385, -0.016525269, -0.059143066, -0.025924683, 0.033569336, 0.028259277, -0.015686035, -0.003206253, -0.014724731, -0.008117676, -0.041412354, 0.010818481, 0.008148193, ...
http://arxiv.org/abs/2303.15265v1
Bilex Rx: Lexical Data Augmentation for Massively Multilingual Machine Translation
Alex Jones, Isaac Caswell, Ishank Saxena, Orhan Firat
cs.CL, cs.AI, cs.LG, I.2.7
Neural machine translation (NMT) has progressed rapidly over the past several years, and modern models are able to achieve relatively high quality using only monolingual text data, an approach dubbed Unsupervised Machine Translation (UNMT). However, these models still struggle in a variety of ways, including aspects of...
http://arxiv.org/pdf/2303.15265v1
2023-03-27T14:54:43
2023-03-27T14:54:43
{ "title": [ -0.023086548, -0.017715454, -0.031280518, -0.07904053, -0.005531311, 0.023757935, -0.05606079, 0.014839172, 0.00793457, 0.054870605, -0.031585693, 0.023742676, 0.03253174, 0.0027503967, -0.04083252, 0.029373169, -0.018127441, 0.05050...
http://arxiv.org/abs/2303.15247v2
Zero-Shot Composed Image Retrieval with Textual Inversion
Alberto Baldrati, Lorenzo Agnolucci, Marco Bertini, Alberto Del Bimbo
cs.CV, cs.CL, cs.IR
Composed Image Retrieval (CIR) aims to retrieve a target image based on a query composed of a reference image and a relative caption that describes the difference between the two images. The high effort and cost required for labeling datasets for CIR hamper the widespread usage of existing methods, as they rely on supe...
http://arxiv.org/pdf/2303.15247v2
2023-08-19T14:04:41
2023-03-27T14:31:25
{ "title": [ 0.016845703, 0.0017776489, -0.0019245148, -0.029953003, 0.007659912, -0.009925842, 0.031311035, -0.018341064, 0.054595947, 0.053985596, -0.036590576, -0.0019779205, -0.0024414062, 0.0140686035, -0.08355713, 0.006958008, 0.0013990402, ...
http://arxiv.org/abs/2303.15230v1
Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning
Siteng Huang, Biao Gong, Yutong Feng, Yiliang Lv, Donglin Wang
cs.CV, cs.CL, cs.LG
Recent compositional zero-shot learning (CZSL) methods adapt pre-trained vision-language models (VLMs) by constructing trainable prompts only for composed state-object pairs. Relying on learning the joint representation of seen compositions, these methods ignore the explicit modeling of the state and object, thus limit...
http://arxiv.org/pdf/2303.15230v1
2023-03-27T14:10:26
2023-03-27T14:10:26
{ "title": [ -0.00038170815, -0.014472961, -0.02885437, -0.07470703, -0.010070801, 0.00050354004, -0.035736084, 0.018325806, 0.014389038, 0.01398468, -0.011886597, 0.018157959, -0.021759033, -0.03552246, -0.014450073, 0.0043678284, -0.018218994, ...
http://arxiv.org/abs/2303.15193v1
CoCon: A Data Set on Combined Contextualized Research Artifact Use
Tarek Saier, Youxiang Dong, Michael Färber
cs.DL, cs.CL
In the wake of information overload in academia, methodologies and systems for search, recommendation, and prediction to aid researchers in identifying relevant research are actively studied and developed. Existing work, however, is limited in terms of granularity, focusing only on the level of papers or a single type ...
http://arxiv.org/pdf/2303.15193v1
2023-03-27T13:29:09
2023-03-27T13:29:09
{ "title": [ -0.0051612854, 0.0063056946, -0.066223145, -0.056854248, 0.021362305, -0.01751709, -0.03475952, -0.031341553, -0.023345947, 0.042175293, -0.017929077, 0.027694702, -0.0107803345, -0.015823364, -0.035583496, 0.015159607, -0.011665344, ...
http://arxiv.org/abs/2303.15190v1
Evaluating self-attention interpretability through human-grounded experimental protocol
Milan Bhan, Nina Achache, Victor Legrand, Annabelle Blangero, Nicolas Chesneau
cs.LG, cs.CL
Attention mechanisms have played a crucial role in the development of complex architectures such as Transformers in natural language processing. However, Transformers remain hard to interpret and are considered as black-boxes. This paper aims to assess how attention coefficients from Transformers can help in providing ...
http://arxiv.org/pdf/2303.15190v1
2023-03-27T13:26:02
2023-03-27T13:26:02
{ "title": [ -0.012702942, 0.017181396, -0.037750244, -0.026977539, 0.011726379, -0.022247314, -0.02331543, -0.026733398, -0.017089844, 0.026016235, 0.007789612, -0.013580322, -0.018432617, -0.021987915, -0.066833496, -0.011505127, -0.0028591156, ...
http://arxiv.org/abs/2303.15132v1
Cross-utterance ASR Rescoring with Graph-based Label Propagation
Srinath Tankasala, Long Chen, Andreas Stolcke, Anirudh Raju, Qianli Deng, Chander Chandak, Aparna Khare, Roland Maas, Venkatesh Ravichandran
eess.AS, cs.CL, cs.LG, cs.SD
We propose a novel approach for ASR N-best hypothesis rescoring with graph-based label propagation by leveraging cross-utterance acoustic similarity. In contrast to conventional neural language model (LM) based ASR rescoring/reranking models, our approach focuses on acoustic information and conducts the rescoring colla...
http://arxiv.org/pdf/2303.15132v1
2023-03-27T12:08:05
2023-03-27T12:08:05
{ "title": [ -0.021057129, -0.017288208, -0.028640747, -0.07598877, -0.014823914, 0.0155181885, -0.05883789, 0.027557373, -0.017868042, 0.036468506, -0.00045490265, 0.017532349, 0.0087890625, -0.02709961, -0.017288208, 0.024642944, -0.018218994, ...
http://arxiv.org/abs/2303.15125v1
LMCanvas: Object-Oriented Interaction to Personalize Large Language Model-Powered Writing Environments
Tae Soo Kim, Arghya Sarkar, Yoonjoo Lee, Minsuk Chang, Juho Kim
cs.HC, cs.CL
Large language models (LLMs) can enhance writing by automating or supporting specific tasks in writers' workflows (e.g., paraphrasing, creating analogies). Leveraging this capability, a collection of interfaces have been developed that provide LLM-powered tools for specific writing tasks. However, these interfaces prov...
http://arxiv.org/pdf/2303.15125v1
2023-03-27T11:56:26
2023-03-27T11:56:26
{ "title": [ -0.023956299, -0.019699097, -0.042266846, -0.049316406, -0.030288696, 0.0013895035, -0.051849365, 0.008552551, -0.02305603, 0.03515625, -0.01789856, 0.005622864, 0.020370483, -0.02268982, -0.01651001, -0.011932373, -0.01637268, 0.032...
http://arxiv.org/abs/2303.16760v1
ACO-tagger: A Novel Method for Part-of-Speech Tagging using Ant Colony Optimization
Amirhossein Mohammadi, Sara Hajiaghajani, Mohammad Bahrani
cs.CL, cs.AI, cs.LG, cs.NE
Swarm Intelligence algorithms have gained significant attention in recent years as a means of solving complex and non-deterministic problems. These algorithms are inspired by the collective behavior of natural creatures, and they simulate this behavior to develop intelligent agents for computational tasks. One such alg...
http://arxiv.org/pdf/2303.16760v1
2023-03-27T11:48:40
2023-03-27T11:48:40
{ "title": [ -0.021331787, 0.004825592, -0.027648926, -0.054107666, 0.013771057, 0.009353638, -0.046020508, -0.010978699, -0.04257202, 0.062561035, 0.0009469986, 0.047698975, 0.004360199, -0.018066406, -0.025238037, 0.013473511, -0.02192688, 0.04...
http://arxiv.org/abs/2303.15116v1
An ontology-aided, natural language-based approach for multi-constraint BIM model querying
Mengtian Yin, Llewellyn Tang, Chris Webster, Shen Xu, Xiongyi Li, Huaquan Ying
cs.CL
Being able to efficiently retrieve the required building information is critical for construction project stakeholders to carry out their engineering and management activities. Natural language interface (NLI) systems are emerging as a time and cost-effective way to query Building Information Models (BIMs). However, th...
http://arxiv.org/pdf/2303.15116v1
2023-03-27T11:35:40
2023-03-27T11:35:40
{ "title": [ -0.06384277, -0.027954102, -0.05999756, -0.062408447, -0.008850098, 0.023895264, -0.06933594, 0.030929565, 0.0058898926, 0.015411377, -0.025802612, -0.01423645, 0.015510559, -0.03805542, -0.0110321045, -0.0054397583, -0.030319214, 0....
http://arxiv.org/abs/2303.15110v1
Beyond Toxic: Toxicity Detection Datasets are Not Enough for Brand Safety
Elizaveta Korotkova, Isaac Kwan Yin Chung
cs.CL, cs.AI
The rapid growth in user generated content on social media has resulted in a significant rise in demand for automated content moderation. Various methods and frameworks have been proposed for the tasks of hate speech detection and toxic comment classification. In this work, we combine common datasets to extend these ta...
http://arxiv.org/pdf/2303.15110v1
2023-03-27T11:29:09
2023-03-27T11:29:09
{ "title": [ 0.007446289, 0.013404846, -0.03970337, -0.050872803, 0.028060913, -0.03479004, -0.013168335, -0.01927185, -0.023620605, 0.00440979, -0.055664062, 0.02947998, -0.039093018, -0.008117676, -0.020721436, -0.033081055, 0.0019388199, 0.030...
http://arxiv.org/abs/2303.15100v2
An Information Extraction Study: Take In Mind the Tokenization!
Christos Theodoropoulos, Marie-Francine Moens
cs.CL
Current research on the advantages and trade-offs of using characters, instead of tokenized text, as input for deep learning models, has evolved substantially. New token-free models remove the traditional tokenization step; however, their efficiency remains unclear. Moreover, the effect of tokenization is relatively un...
http://arxiv.org/pdf/2303.15100v2
2023-04-01T19:04:58
2023-03-27T11:08:35
{ "title": [ -0.037109375, 0.010879517, -0.05114746, -0.037109375, 0.007686615, -0.016342163, -0.019195557, 0.002872467, -0.029937744, 0.01424408, -0.058258057, 0.03390503, -0.016601562, -0.003414154, -0.00869751, -0.033813477, -0.026031494, 0.04...
http://arxiv.org/abs/2303.15078v3
Large Language Models are Diverse Role-Players for Summarization Evaluation
Ning Wu, Ming Gong, Linjun Shou, Shining Liang, Daxin Jiang
cs.CL
Text summarization has a wide range of applications in many scenarios. The evaluation of the quality of the generated text is a complex problem. A big challenge to language evaluation is that there is a clear divergence between existing metrics and human evaluation. A document summary's quality can be assessed by human...
http://arxiv.org/pdf/2303.15078v3
2023-09-19T10:07:55
2023-03-27T10:40:59
{ "title": [ -0.0023765564, -0.001250267, -0.03414917, -0.08654785, -0.030532837, -0.010353088, -0.08898926, 0.003648758, -0.019577026, 0.03881836, -0.009063721, 0.009780884, -0.0116119385, -0.020339966, -0.034851074, 0.0018186569, -0.005493164, ...
http://arxiv.org/abs/2303.15056v2
ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks
Fabrizio Gilardi, Meysam Alizadeh, Maël Kubli
cs.CL, cs.CY
Many NLP applications require manual data annotations for a variety of tasks, notably to train classifiers or evaluate the performance of unsupervised models. Depending on the size and degree of complexity, the tasks may be conducted by crowd-workers on platforms such as MTurk as well as trained annotators, such as res...
http://arxiv.org/pdf/2303.15056v2
2023-07-19T14:10:55
2023-03-27T09:59:48
{ "title": [ 0.030563354, -0.00019705296, -0.0035114288, -0.02709961, -0.031097412, -0.013290405, -0.020568848, -0.020889282, -0.026382446, 0.04498291, -0.029541016, 0.01852417, -0.004573822, -0.004135132, -0.026443481, 0.0032310486, -0.0128479, ...
http://arxiv.org/abs/2303.15049v3
InterviewBot: Real-Time End-to-End Dialogue System to Interview Students for College Admission
Zihao Wang, Nathan Keyes, Terry Crawford, Jinho D. Choi
cs.CL, cs.HC
We present the InterviewBot that dynamically integrates conversation history and customized topics into a coherent embedding space to conduct 10 mins hybrid-domain (open and closed) conversations with foreign students applying to U.S. colleges for assessing their academic and cultural readiness. To build a neural-based...
http://arxiv.org/pdf/2303.15049v3
2023-09-05T15:33:49
2023-03-27T09:46:56
{ "title": [ 0.013671875, -0.011436462, -0.0004146099, -0.0074691772, 0.0061416626, -0.016479492, 0.045166016, -0.036224365, -0.0033550262, 0.02861023, -0.01109314, 0.007041931, -0.0020446777, -0.023864746, -0.036376953, -0.016418457, -0.027816772, ...
http://arxiv.org/abs/2303.15016v1
Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification
Chunpu Xu, Jing Li
cs.CL, cs.AI, cs.IR, cs.MM
Social media is daily creating massive multimedia content with paired image and text, presenting the pressing need to automate the vision and language understanding for various multimodal classification tasks. Compared to the commonly researched visual-lingual data, social media posts tend to exhibit more implicit imag...
http://arxiv.org/pdf/2303.15016v1
2023-03-27T08:59:55
2023-03-27T08:59:55
{ "title": [ 0.01272583, 0.018798828, -0.012084961, -0.03253174, 0.0031528473, -0.018310547, -0.008293152, -0.016448975, -0.02017212, 0.040527344, -0.01852417, 0.033203125, -0.02130127, -0.0060577393, -0.039886475, -0.01676941, 0.0020885468, 0.05...
http://arxiv.org/abs/2303.15006v1
Curriculum Learning for Compositional Visual Reasoning
Wafa Aissa, Marin Ferecatu, Michel Crucianu
cs.CL, cs.CV, cs.LG
Visual Question Answering (VQA) is a complex task requiring large datasets and expensive training. Neural Module Networks (NMN) first translate the question to a reasoning path, then follow that path to analyze the image and provide an answer. We propose an NMN method that relies on predefined cross-modal embeddings to...
http://arxiv.org/pdf/2303.15006v1
2023-03-27T08:47:18
2023-03-27T08:47:18
{ "title": [ -0.015129089, -0.015609741, -0.044128418, -0.05657959, -0.010368347, -0.01612854, -0.03604126, 0.0072784424, 0.01914978, 0.0019273758, -0.030883789, 0.010955811, -0.04244995, -0.024536133, -0.019241333, -0.02558899, -0.002500534, 0.0...
http://arxiv.org/abs/2303.14963v1
Variation and Instability in Dialect-Based Embedding Spaces
Jonathan Dunn
cs.CL
This paper measures variation in embedding spaces which have been trained on different regional varieties of English while controlling for instability in the embeddings. While previous work has shown that it is possible to distinguish between similar varieties of a language, this paper experiments with two follow-up qu...
http://arxiv.org/pdf/2303.14963v1
2023-03-27T07:53:23
2023-03-27T07:53:23
{ "title": [ 0.005493164, 0.0004785061, -0.02848816, -0.026947021, 0.006919861, -0.014122009, -0.02420044, -0.01890564, 0.0064735413, 0.036010742, -0.022644043, 0.002752304, 0.0013494492, -0.020217896, -0.039001465, 0.00062179565, 0.0046043396, 0...
http://arxiv.org/abs/2303.16778v1
Assorted, Archetypal and Annotated Two Million (3A2M) Cooking Recipes Dataset based on Active Learning
Nazmus Sakib, G. M. Shahariar, Md. Mohsinul Kabir, Md. Kamrul Hasan, Hasan Mahmud
cs.CL, cs.AI, cs.LG
Cooking recipes allow individuals to exchange culinary ideas and provide food preparation instructions. Due to a lack of adequate labeled data, categorizing raw recipes found online to the appropriate food genres is a challenging task in this domain. Utilizing the knowledge of domain experts to categorize recipes could...
http://arxiv.org/pdf/2303.16778v1
2023-03-27T07:53:18
2023-03-27T07:53:18
{ "title": [ 0.013435364, 0.018844604, -0.024414062, -0.028915405, 0.01524353, -0.03125, -0.020706177, -0.01687622, -0.015235901, 0.056732178, 0.0309906, 0.01033783, 0.008285522, -0.05709839, 0.0013914108, 0.022277832, 0.02508545, 0.007713318, ...
http://arxiv.org/abs/2303.14957v1
unarXive 2022: All arXiv Publications Pre-Processed for NLP, Including Structured Full-Text and Citation Network
Tarek Saier, Johan Krause, Michael Färber
cs.DL, cs.CL
Large-scale data sets on scholarly publications are the basis for a variety of bibliometric analyses and natural language processing (NLP) applications. Especially data sets derived from publication's full-text have recently gained attention. While several such data sets already exist, we see key shortcomings in terms ...
http://arxiv.org/pdf/2303.14957v1
2023-03-27T07:40:59
2023-03-27T07:40:59
{ "title": [ -0.029556274, -0.014152527, -0.0043411255, -0.026885986, -0.04650879, -0.002445221, -0.07647705, 0.001660347, -0.008216858, -0.0053138733, 0.04458618, -0.00052547455, 0.013282776, -0.07537842, 0.04611206, 0.010345459, 0.023452759, 0....
http://arxiv.org/abs/2303.14956v2
Unified Text Structuralization with Instruction-tuned Language Models
Xuanfan Ni, Piji Li, Huayang Li
cs.CL
Text structuralization is one of the important fields of natural language processing (NLP) consists of information extraction (IE) and structure formalization. However, current studies of text structuralization suffer from a shortage of manually annotated high-quality datasets from different domains and languages, whic...
http://arxiv.org/pdf/2303.14956v2
2023-03-30T13:41:54
2023-03-27T07:39:05
{ "title": [ -0.05731201, -0.017166138, -0.012672424, -0.07312012, -0.02305603, 0.035949707, -0.09674072, -0.005054474, -0.015274048, 0.0440979, -0.009651184, 0.02420044, -0.004753113, 0.001657486, -0.042419434, 0.0126571655, -0.012237549, 0.0332...
http://arxiv.org/abs/2303.14951v1
Improving Contextualized Topic Models with Negative Sampling
Suman Adhya, Avishek Lahiri, Debarshi Kumar Sanyal, Partha Pratim Das
cs.CL, cs.LG
Topic modeling has emerged as a dominant method for exploring large document collections. Recent approaches to topic modeling use large contextualized language models and variational autoencoders. In this paper, we propose a negative sampling mechanism for a contextualized topic model to improve the quality of the gene...
http://arxiv.org/pdf/2303.14951v1
2023-03-27T07:28:46
2023-03-27T07:28:46
{ "title": [ 0.024398804, 0.022018433, -0.019729614, -0.05154419, 0.013198853, -0.0079574585, -0.02822876, -0.057159424, 0.03567505, 0.03036499, -0.037902832, 0.044128418, -0.005428314, -0.005054474, -0.039123535, 0.022125244, -0.005794525, 0.005...
http://arxiv.org/abs/2303.14935v1
TabIQA: Table Questions Answering on Business Document Images
Phuc Nguyen, Nam Tuan Ly, Hideaki Takeda, Atsuhiro Takasu
cs.CV, cs.CL
Table answering questions from business documents has many challenges that require understanding tabular structures, cross-document referencing, and additional numeric computations beyond simple search queries. This paper introduces a novel pipeline, named TabIQA, to answer questions about business document images. Tab...
http://arxiv.org/pdf/2303.14935v1
2023-03-27T06:31:21
2023-03-27T06:31:21
{ "title": [ -0.008354187, -0.027923584, -0.046783447, -0.033172607, -0.026367188, -0.023590088, -0.030532837, -0.004421234, 0.0158844, 0.04714966, -0.050079346, 0.0082473755, 0.035095215, -0.0090408325, -0.043548584, -0.025405884, 0.0184021, -0....
http://arxiv.org/abs/2303.14920v1
Adapting Pretrained Language Models for Solving Tabular Prediction Problems in the Electronic Health Record
Christopher McMaster, David FL Liew, Douglas EV Pires
cs.CL, cs.LG
We propose an approach for adapting the DeBERTa model for electronic health record (EHR) tasks using domain adaptation. We pretrain a small DeBERTa model on a dataset consisting of MIMIC-III discharge summaries, clinical notes, radiology reports, and PubMed abstracts. We compare this model's performance with a DeBERTa ...
http://arxiv.org/pdf/2303.14920v1
2023-03-27T05:34:19
2023-03-27T05:34:19
{ "title": [ -0.010421753, -0.021820068, -0.033355713, -0.089416504, -0.008522034, 0.011260986, -0.08258057, 0.0057907104, -0.013450623, 0.045776367, -0.01335907, 0.0037899017, 0.019866943, -0.017349243, -0.0423584, -0.0126953125, 0.010063171, 0....
http://arxiv.org/abs/2303.16763v1
Meeting Action Item Detection with Regularized Context Modeling
Jiaqing Liu, Chong Deng, Qinglin Zhang, Qian Chen, Wen Wang
cs.CL
Meetings are increasingly important for collaborations. Action items in meeting transcripts are crucial for managing post-meeting to-do tasks, which usually are summarized laboriously. The Action Item Detection task aims to automatically detect meeting content associated with action items. However, datasets manually an...
http://arxiv.org/pdf/2303.16763v1
2023-03-27T03:27:08
2023-03-27T03:27:08
{ "title": [ 0.0052108765, 0.03503418, -0.06304932, -0.06726074, 0.035095215, -0.02255249, -0.03857422, -0.0357666, 0.045318604, 0.03942871, -0.043884277, 0.010948181, 0.02494812, -0.009414673, -0.04534912, 0.004070282, -0.0044059753, 0.03125, ...
http://arxiv.org/abs/2303.14898v1
Mutually-paced Knowledge Distillation for Cross-lingual Temporal Knowledge Graph Reasoning
Ruijie Wang, Zheng Li, Jingfeng Yang, Tianyu Cao, Chao Zhang, Bing Yin, Tarek Abdelzaher
cs.LG, cs.AI, cs.CL, cs.SI
This paper investigates cross-lingual temporal knowledge graph reasoning problem, which aims to facilitate reasoning on Temporal Knowledge Graphs (TKGs) in low-resource languages by transfering knowledge from TKGs in high-resource ones. The cross-lingual distillation ability across TKGs becomes increasingly crucial, in...
http://arxiv.org/pdf/2303.14898v1
2023-03-27T03:15:27
2023-03-27T03:15:27
{ "title": [ -0.011856079, -0.014129639, -0.07696533, -0.11645508, 0.0102005005, -0.011245728, -0.07849121, 0.044769287, 0.02748108, 0.028015137, -0.017333984, -0.005970001, 0.006916046, -0.018432617, -0.040771484, -0.00868988, 0.0049819946, 0.01...
http://arxiv.org/abs/2303.16753v2
Scaling Pre-trained Language Models to Deeper via Parameter-efficient Architecture
Peiyu Liu, Ze-Feng Gao, Yushuo Chen, Wayne Xin Zhao, Ji-Rong Wen
cs.CL
In this paper, we propose a highly parameter-efficient approach to scaling pre-trained language models (PLMs) to a deeper model depth. Unlike prior work that shares all parameters or uses extra blocks, we design a more capable parameter-sharing architecture based on matrix product operator (MPO). MPO decomposition can ...
http://arxiv.org/pdf/2303.16753v2
2023-04-11T02:45:10
2023-03-27T02:34:09
{ "title": [ -0.0050735474, -0.012290955, -0.02432251, -0.09051514, -0.0015583038, 0.0067367554, -0.06121826, 0.0135650635, 0.0020065308, 0.033325195, -0.017730713, 0.006111145, 0.01574707, -0.011375427, -0.0340271, 0.018508911, -0.00023293495, 0...
http://arxiv.org/abs/2303.14871v1
Coupling Artificial Neurons in BERT and Biological Neurons in the Human Brain
Xu Liu, Mengyue Zhou, Gaosheng Shi, Yu Du, Lin Zhao, Zihao Wu, David Liu, Tianming Liu, Xintao Hu
cs.CL
Linking computational natural language processing (NLP) models and neural responses to language in the human brain on the one hand facilitates the effort towards disentangling the neural representations underpinning language perception, on the other hand provides neurolinguistics evidence to evaluate and improve NLP mo...
http://arxiv.org/pdf/2303.14871v1
2023-03-27T01:41:48
2023-03-27T01:41:48
{ "title": [ 0.05343628, -0.0008044243, 0.02192688, -0.01928711, -0.03387451, -0.025375366, 0.016067505, -0.022720337, -0.040740967, 0.057403564, 0.0021915436, -0.02583313, 0.05795288, -0.055633545, 0.0057792664, 0.006412506, 0.008903503, 0.00968...
http://arxiv.org/abs/2303.14814v1
WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation
Jongheon Jeong, Yang Zou, Taewan Kim, Dongqing Zhang, Avinash Ravichandran, Onkar Dabeer
cs.CV, cs.AI, cs.CL
Visual anomaly classification and segmentation are vital for automating industrial quality inspection. The focus of prior research in the field has been on training custom models for each quality inspection task, which requires task-specific images and annotation. In this paper we move away from this regime, addressing...
http://arxiv.org/pdf/2303.14814v1
2023-03-26T20:41:21
2023-03-26T20:41:21
{ "title": [ 0.0129776, -0.005367279, -0.03717041, -0.041137695, 0.0085372925, -0.013885498, -0.0019302368, -0.01826477, -0.00894165, 0.011207581, 0.0024776459, 0.013679504, -0.026947021, -0.033325195, -0.031082153, 0.0013370514, -0.005302429, 0....
http://arxiv.org/abs/2303.14770v1
Koala: An Index for Quantifying Overlaps with Pre-training Corpora
Thuy-Trang Vu, Xuanli He, Gholamreza Haffari, Ehsan Shareghi
cs.CL, cs.LG
In very recent years more attention has been placed on probing the role of pre-training data in Large Language Models (LLMs) downstream behaviour. Despite the importance, there is no public tool that supports such analysis of pre-training corpora at large scale. To help research in this space, we launch Koala, a search...
http://arxiv.org/pdf/2303.14770v1
2023-03-26T16:29:18
2023-03-26T16:29:18
{ "title": [ -0.01979065, -0.0024108887, -0.03717041, -0.058166504, -0.006095886, 0.00046563148, -0.04751587, -0.016738892, -0.009109497, 0.027359009, -0.026046753, 0.032989502, 0.0030765533, -0.02519226, -0.004623413, -0.0043144226, -0.020309448, ...
http://arxiv.org/abs/2303.14742v1
Exploring the Impact of Instruction Data Scaling on Large Language Models: An Empirical Study on Real-World Use Cases
Yunjie Ji, Yong Deng, Yan Gong, Yiping Peng, Qiang Niu, Lei Zhang, Baochang Ma, Xiangang Li
cs.CL
The success of ChatGPT has recently attracted numerous efforts to replicate it, with instruction-tuning strategies being a key factor in achieving remarkable results. Instruction-tuning not only significantly enhances the model's performance and generalization but also makes the model's generated results more consisten...
http://arxiv.org/pdf/2303.14742v1
2023-03-26T14:49:37
2023-03-26T14:49:37
{ "title": [ -0.03753662, -0.027557373, -0.06744385, -0.05682373, -0.010566711, -0.020324707, -0.049591064, 0.024810791, -0.00818634, 0.02999878, -0.028518677, 0.012161255, -0.011116028, -0.006084442, -0.03817749, -0.019500732, -0.011695862, 0.03...
http://arxiv.org/abs/2303.14725v2
Natural Language Reasoning, A Survey
Fei Yu, Hongbo Zhang, Prayag Tiwari, Benyou Wang
cs.CL
This survey paper proposes a clearer view of natural language reasoning in the field of Natural Language Processing (NLP), both conceptually and practically. Conceptually, we provide a distinct definition for natural language reasoning in NLP, based on both philosophy and NLP scenarios, discuss what types of tasks requ...
http://arxiv.org/pdf/2303.14725v2
2023-05-13T15:56:44
2023-03-26T13:44:18
{ "title": [ -0.026367188, -0.021270752, -0.022750854, -0.083496094, 0.014526367, 0.028900146, -0.04425049, 0.01486969, 0.005241394, 0.012199402, -0.028915405, 0.0016641617, -0.0068244934, -0.015701294, -0.026519775, 0.0025558472, -0.016067505, 0...
http://arxiv.org/abs/2303.14704v2
Task-oriented Memory-efficient Pruning-Adapter
Guorun Wang, Jun Yang, Yaoru Sun
cs.CL, cs.LG
The Outstanding performance and growing size of Large Language Models has led to increased attention in parameter efficient learning. The two predominant approaches are Adapters and Pruning. Adapters are to freeze the model and give it a new weight matrix on the side, which can significantly reduce the time and memory ...
http://arxiv.org/pdf/2303.14704v2
2023-04-06T03:44:38
2023-03-26T12:18:00
{ "title": [ 0.008705139, -0.0013771057, -0.046661377, -0.07318115, -0.00642395, -0.013908386, -0.024612427, 0.032989502, 0.020690918, 0.04550171, 0.014533997, -0.033325195, 0.043182373, -0.038482666, -0.034973145, 0.028335571, 0.024093628, 0.004...
http://arxiv.org/abs/2303.14655v2
GOAL: A Challenging Knowledge-grounded Video Captioning Benchmark for Real-time Soccer Commentary Generation
Ji Qi, Jifan Yu, Teng Tu, Kunyu Gao, Yifan Xu, Xinyu Guan, Xiaozhi Wang, Yuxiao Dong, Bin Xu, Lei Hou, Juanzi Li, Jie Tang, Weidong Guo, Hui Liu, Yu Xu
cs.CV, cs.CL, cs.LG
Despite the recent emergence of video captioning models, how to generate vivid, fine-grained video descriptions based on the background knowledge (i.e., long and informative commentary about the domain-specific scenes with appropriate reasoning) is still far from being solved, which however has great applications such ...
http://arxiv.org/pdf/2303.14655v2
2023-10-05T06:55:13
2023-03-26T08:43:36
{ "title": [ -0.0075798035, -0.017745972, -0.0151901245, -0.04534912, -0.011116028, -0.011787415, -0.022109985, 0.0072135925, 0.004337311, 0.05770874, -0.025665283, 0.01184845, 0.027618408, -0.0038490295, -0.061309814, 0.011627197, -0.005748749, ...
http://arxiv.org/abs/2303.14647v1
Farspredict: A benchmark dataset for link prediction
Najmeh Torabian, Behrouz Minaei-Bidgoli, Mohsen Jahanshahi
cs.AI, cs.CL, 00Axx, E.1; H.3
Link prediction with knowledge graph embedding (KGE) is a popular method for knowledge graph completion. Furthermore, training KGEs on non-English knowledge graph promote knowledge extraction and knowledge graph reasoning in the context of these languages. However, many challenges in non-English KGEs pose to learning a...
http://arxiv.org/pdf/2303.14647v1
2023-03-26T07:41:26
2023-03-26T07:41:26
{ "title": [ 0.0040283203, -0.012184143, -0.032684326, -0.06628418, 0.010261536, -0.00217247, -0.039276123, -0.017837524, 0.0067367554, 0.026290894, 0.014389038, 0.014427185, -0.010749817, -0.029296875, -0.022583008, 0.023498535, -0.013122559, 0....
http://arxiv.org/abs/2303.16764v1
Boosting Few-Shot Text Classification via Distribution Estimation
Han Liu, Feng Zhang, Xiaotong Zhang, Siyang Zhao, Fenglong Ma, Xiao-Ming Wu, Hongyang Chen, Hong Yu, Xianchao Zhang
cs.CL
Distribution estimation has been demonstrated as one of the most effective approaches in dealing with few-shot image classification, as the low-level patterns and underlying representations can be easily transferred across different tasks in computer vision domain. However, directly applying this approach to few-shot t...
http://arxiv.org/pdf/2303.16764v1
2023-03-26T05:58:39
2023-03-26T05:58:39
{ "title": [ -0.011489868, -0.0021743774, -0.026260376, -0.025131226, 0.00793457, -0.0016698837, -0.0009279251, -0.008277893, 0.0070877075, 0.019104004, -0.04232788, 0.020568848, -0.024337769, 0.005844116, -0.052124023, -0.034606934, -0.011474609, ...
http://arxiv.org/abs/2303.14589v1
SASS: Data and Methods for Subject Aware Sentence Simplification
Brad Windsor, Luke Martin, Anand Tyagi
cs.CL, cs.AI
Sentence simplification tends to focus on the generic simplification of sentences by making them more readable and easier to understand. This paper provides a dataset aimed at training models that perform subject aware sentence simplifications rather than simplifying sentences as a whole. We also test models on that da...
http://arxiv.org/pdf/2303.14589v1
2023-03-26T00:02:25
2023-03-26T00:02:25
{ "title": [ -0.027908325, -0.007911682, -0.051605225, -0.07495117, 0.0102005005, 0.014335632, -0.055847168, -0.0021800995, 0.007827759, 0.03479004, 0.008651733, 0.024291992, 0.004852295, -0.019821167, -0.050933838, 0.017181396, -0.00014173985, 0...
http://arxiv.org/abs/2303.14588v1
Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization
Bashar Al-Rfooh, Gheith Abandah, Rami Al-Rfou
cs.CL
Most of previous work on learning diacritization of the Arabic language relied on training models from scratch. In this paper, we investigate how to leverage pre-trained language models to learn diacritization. We finetune token-free pre-trained multilingual models (ByT5) to learn to predict and insert missing diacriti...
http://arxiv.org/pdf/2303.14588v1
2023-03-25T23:41:33
2023-03-25T23:41:33
{ "title": [ -0.016311646, -0.006542206, -0.008239746, -0.078308105, 0.004535675, 0.01651001, -0.04675293, 0.0050964355, -0.008125305, 0.05432129, -0.0033092499, 0.0061569214, 0.022903442, -0.015304565, -0.05847168, 0.0035533905, 0.0036354065, 0....
http://arxiv.org/abs/2303.14582v1
Identification of Negative Transfers in Multitask Learning Using Surrogate Models
Dongyue Li, Huy L. Nguyen, Hongyang R. Zhang
cs.LG, cs.AI, cs.CL, stat.ML
Multitask learning is widely used in practice to train a low-resource target task by augmenting it with multiple related source tasks. Yet, naively combining all the source tasks with a target task does not always improve the prediction performance for the target task due to negative transfers. Thus, a critical problem...
http://arxiv.org/pdf/2303.14582v1
2023-03-25T23:16:11
2023-03-25T23:16:11
{ "title": [ 0.023590088, 0.02633667, -0.0088272095, 0.011405945, 0.003583908, -0.034179688, 0.008956909, -0.06304932, -0.033081055, 0.025558472, 0.0024662018, 0.02645874, -0.025939941, -0.039886475, -0.017349243, -0.01927185, -0.010444641, 0.028...
http://arxiv.org/abs/2303.14576v1
Automatic Generation of Multiple-Choice Questions
Cheng Zhang
cs.CL
Creating multiple-choice questions to assess reading comprehension of a given article involves generating question-answer pairs (QAPs) and adequate distractors. We present two methods to tackle the challenge of QAP generations: (1) A deep-learning-based end-to-end question generation system based on T5 Transformer with...
http://arxiv.org/pdf/2303.14576v1
2023-03-25T22:45:54
2023-03-25T22:45:54
{ "title": [ -0.026748657, 0.004901886, -0.033447266, -0.026031494, -0.0017538071, -0.0002901554, 0.004211426, -0.015045166, 0.027297974, 0.017410278, -0.017730713, 0.008644104, -0.03515625, -0.017303467, -0.040618896, 0.0041046143, -0.010482788, ...
http://arxiv.org/abs/2303.14537v1
Deep Augmentation: Enhancing Self-Supervised Learning through Transformations in Higher Activation Space
Rickard Brüel-Gabrielsson, Tongzhou Wang, Manel Baradad, Justin Solomon
cs.LG, cs.CL, cs.CV
We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentatio...
http://arxiv.org/pdf/2303.14537v1
2023-03-25T19:03:57
2023-03-25T19:03:57
{ "title": [ 0.010116577, 0.01838684, -0.060668945, -0.058776855, 0.029953003, -0.022323608, -0.019805908, -0.019302368, -0.008811951, 0.016433716, -0.01852417, 0.027862549, -0.024765015, -0.021209717, -0.017852783, 0.0077171326, 0.0037269592, 0....
http://arxiv.org/abs/2303.14524v2
Chat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender System
Yunfan Gao, Tao Sheng, Youlin Xiang, Yun Xiong, Haofen Wang, Jiawei Zhang
cs.IR, cs.CL, cs.LG
Large language models (LLMs) have demonstrated their significant potential to be applied for addressing various application tasks. However, traditional recommender systems continue to face great challenges such as poor interactivity and explainability, which actually also hinder their broad deployment in real-world sys...
http://arxiv.org/pdf/2303.14524v2
2023-04-04T03:51:27
2023-03-25T17:37:43
{ "title": [ -0.027816772, -0.016967773, -0.0340271, -0.03942871, -0.019180298, 0.012413025, -0.037963867, 0.0033912659, -0.015991211, 0.011405945, -0.03253174, 0.03475952, -0.037475586, -0.012458801, -0.025131226, -0.02029419, -0.027999878, 0.02...
http://arxiv.org/abs/2303.14480v1
GANTEE: Generative Adversatial Network for Taxonomy Entering Evaluation
Zhouhong Gu, Sihang Jiang, Jingping Liu, Yanghua Xiao, Hongwei Feng, Zhixu Li, Jiaqing Liang, Jian Zhong
cs.AI, cs.CL
Taxonomy is formulated as directed acyclic concepts graphs or trees that support many downstream tasks. Many new coming concepts need to be added to an existing taxonomy. The traditional taxonomy expansion task aims only at finding the best position for new coming concepts in the existing taxonomy. However, they have t...
http://arxiv.org/pdf/2303.14480v1
2023-03-25T14:24:50
2023-03-25T14:24:50
{ "title": [ -0.0036773682, -0.0056419373, -0.033325195, -0.046447754, 0.015327454, -0.01386261, -0.0033283234, -0.010002136, -0.0046195984, -0.0027656555, -0.01638794, 0.016525269, -0.050567627, -0.022262573, -0.034088135, -0.022537231, -0.021896362...
http://arxiv.org/abs/2303.14475v1
Informed Machine Learning, Centrality, CNN, Relevant Document Detection, Repatriation of Indigenous Human Remains
Md Abul Bashar, Richi Nayak, Gareth Knapman, Paul Turnbull, Cressida Fforde
cs.CL, cs.AI, cs.CY, cs.IR, cs.LG, cs.SI
Among the pressing issues facing Australian and other First Nations peoples is the repatriation of the bodily remains of their ancestors, which are currently held in Western scientific institutions. The success of securing the return of these remains to their communities for reburial depends largely on locating informa...
http://arxiv.org/pdf/2303.14475v1
2023-03-25T14:08:21
2023-03-25T14:08:21
{ "title": [ 0.020111084, 0.0040016174, -0.019180298, -0.010658264, -0.025848389, -0.042114258, 0.0102005005, -0.0022945404, 0.00010973215, 0.039611816, -0.031707764, 0.017913818, 0.045928955, -0.033843994, 0.013694763, -0.01335907, 0.009628296, ...
http://arxiv.org/abs/2303.14461v1
Indian Language Summarization using Pretrained Sequence-to-Sequence Models
Ashok Urlana, Sahil Manoj Bhatt, Nirmal Surange, Manish Shrivastava
cs.CL
The ILSUM shared task focuses on text summarization for two major Indian languages- Hindi and Gujarati, along with English. In this task, we experiment with various pretrained sequence-to-sequence models to find out the best model for each of the languages. We present a detailed overview of the models and our approache...
http://arxiv.org/pdf/2303.14461v1
2023-03-25T13:05:54
2023-03-25T13:05:54
{ "title": [ -0.0032043457, 0.010139465, -0.013977051, -0.08746338, -0.005683899, 0.021728516, -0.081970215, 0.0026683807, -0.0071640015, 0.046447754, -0.000027656555, 0.014335632, 0.013336182, -0.0104904175, -0.04333496, 0.014732361, 0.0031471252, ...
http://arxiv.org/abs/2303.14452v1
COFFEE: A Contrastive Oracle-Free Framework for Event Extraction
Meiru Zhang, Yixuan Su, Zaiqiao Meng, Zihao Fu, Nigel Collier
cs.CL
Event extraction is a complex information extraction task that involves extracting events from unstructured text. Prior classification-based methods require comprehensive entity annotations for joint training, while newer generation-based methods rely on heuristic templates containing oracle information such as event t...
http://arxiv.org/pdf/2303.14452v1
2023-03-25T12:12:30
2023-03-25T12:12:30
{ "title": [ -0.0019893646, -0.0064201355, -0.04864502, -0.11425781, 0.028167725, 0.014183044, -0.06781006, -0.011054993, 0.024932861, 0.02973938, 0.007896423, -0.004425049, 0.010360718, -0.029632568, -0.0473938, 0.024795532, -0.008628845, 0.0163...