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/2309.02105v1
Improving Query-Focused Meeting Summarization with Query-Relevant Knowledge
Tiezheng Yu, Ziwei Ji, Pascale Fung
cs.CL, cs.AI
Query-Focused Meeting Summarization (QFMS) aims to generate a summary of a given meeting transcript conditioned upon a query. The main challenges for QFMS are the long input text length and sparse query-relevant information in the meeting transcript. In this paper, we propose a knowledge-enhanced two-stage framework ca...
http://arxiv.org/pdf/2309.02105v1
2023-09-05T10:26:02
2023-09-05T10:26:02
{ "title": [ 0.0069351196, 0.0045394897, -0.05987549, -0.11669922, 0.017196655, -0.0040245056, -0.06896973, 0.018844604, 0.016433716, 0.059020996, -0.012321472, 0.010017395, 0.02330017, 0.0031585693, -0.07208252, 0.0076904297, 0.013298035, -0.011...
http://arxiv.org/abs/2309.02092v3
Where are We in Event-centric Emotion Analysis? Bridging Emotion Role Labeling and Appraisal-based Approaches
Roman Klinger
cs.CL
The term emotion analysis in text subsumes various natural language processing tasks which have in common the goal to enable computers to understand emotions. Most popular is emotion classification in which one or multiple emotions are assigned to a predefined textual unit. While such setting is appropriate for identif...
http://arxiv.org/pdf/2309.02092v3
2023-10-12T11:46:23
2023-09-05T09:56:29
{ "title": [ -0.0028152466, -0.002281189, -0.020385742, -0.05355835, 0.017944336, -0.0061035156, -0.008682251, -0.0107803345, -0.0067596436, 0.031951904, 0.011688232, 0.0069618225, 0.0033721924, -0.03768921, -0.040008545, 0.01802063, -0.012077332, ...
http://arxiv.org/abs/2309.02077v1
An Automatic Evaluation Framework for Multi-turn Medical Consultations Capabilities of Large Language Models
Yusheng Liao, Yutong Meng, Hongcheng Liu, Yanfeng Wang, Yu Wang
cs.CL
Large language models (LLMs) have achieved significant success in interacting with human. However, recent studies have revealed that these models often suffer from hallucinations, leading to overly confident but incorrect judgments. This limits their application in the medical domain, where tasks require the utmost acc...
http://arxiv.org/pdf/2309.02077v1
2023-09-05T09:24:48
2023-09-05T09:24:48
{ "title": [ 0.00031805038, 0.0053482056, -0.028900146, -0.08996582, 0.0025730133, 0.0109939575, -0.07989502, -0.006828308, 0.005317688, 0.047210693, -0.02154541, -0.0027542114, -0.0018072128, -0.012359619, -0.058563232, -0.0065612793, 0.0032596588, ...
http://arxiv.org/abs/2309.02045v1
Enhance Multi-domain Sentiment Analysis of Review Texts through Prompting Strategies
Yajing Wang, Zongwei Luo
cs.CL, cs.AI, cs.LG
Large Language Models (LLMs) have made significant strides in both scientific research and practical applications. Existing studies have demonstrated the state-of-the-art (SOTA) performance of LLMs in various natural language processing tasks. However, the question of how to further enhance LLMs' performance in specifi...
http://arxiv.org/pdf/2309.02045v1
2023-09-05T08:44:23
2023-09-05T08:44:23
{ "title": [ -0.0042266846, -0.011436462, -0.008407593, -0.030090332, -0.009109497, -0.013366699, -0.008483887, -0.016647339, -0.0030155182, 0.027160645, -0.04425049, 0.0340271, -0.0028076172, 0.0034942627, -0.046447754, -0.004962921, -0.0135650635, ...
http://arxiv.org/abs/2309.01953v1
Bilevel Scheduled Sampling for Dialogue Generation
Jiawen Liu, Kan Li
cs.CL
Exposure bias poses a common challenge in numerous natural language processing tasks, particularly in the dialog generation. In response to this issue, researchers have devised various techniques, among which scheduled sampling has proven to be an effective method for mitigating exposure bias. However, the existing sta...
http://arxiv.org/pdf/2309.01953v1
2023-09-05T05:05:06
2023-09-05T05:05:06
{ "title": [ -0.00762558, -0.014190674, -0.04220581, -0.08721924, -0.008285522, -0.002986908, -0.0541687, 0.017532349, -0.018569946, 0.032440186, -0.017456055, 0.02041626, -0.012611389, -0.021331787, -0.034088135, -0.0002644062, -0.016357422, 0.0...
http://arxiv.org/abs/2309.01947v1
TODM: Train Once Deploy Many Efficient Supernet-Based RNN-T Compression For On-device ASR Models
Yuan Shangguan, Haichuan Yang, Danni Li, Chunyang Wu, Yassir Fathullah, Dilin Wang, Ayushi Dalmia, Raghuraman Krishnamoorthi, Ozlem Kalinli, Junteng Jia, Jay Mahadeokar, Xin Lei, Mike Seltzer, Vikas Chandra
cs.CL, cs.LG, cs.SD, eess.AS
Automatic Speech Recognition (ASR) models need to be optimized for specific hardware before they can be deployed on devices. This can be done by tuning the model's hyperparameters or exploring variations in its architecture. Re-training and re-validating models after making these changes can be a resource-intensive tas...
http://arxiv.org/pdf/2309.01947v1
2023-09-05T04:47:55
2023-09-05T04:47:55
{ "title": [ -0.0008301735, -0.020599365, -0.016464233, -0.0501709, -0.04547119, 0.0046463013, -0.06262207, 0.013908386, -0.039794922, 0.03845215, 0.040374756, 0.0060424805, 0.02746582, -0.061950684, 0.02279663, 0.007183075, 0.0073776245, -0.0156...
http://arxiv.org/abs/2309.01940v3
CodeApex: A Bilingual Programming Evaluation Benchmark for Large Language Models
Lingyue Fu, Huacan Chai, Shuang Luo, Kounianhua Du, Weiming Zhang, Longteng Fan, Jiayi Lei, Renting Rui, Jianghao Lin, Yuchen Fang, Yifan Liu, Jingkuan Wang, Siyuan Qi, Kangning Zhang, Weinan Zhang, Yong Yu
cs.CL, cs.AI
With the emergence of Large Language Models (LLMs), there has been a significant improvement in the programming capabilities of models, attracting growing attention from researchers. We propose CodeApex, a bilingual benchmark dataset focusing on the programming comprehension and code generation abilities of LLMs. CodeA...
http://arxiv.org/pdf/2309.01940v3
2023-09-10T13:32:38
2023-09-05T04:12:01
{ "title": [ -0.06573486, -0.02810669, -0.026489258, -0.054840088, -0.014427185, 0.016464233, -0.070007324, 0.013412476, -0.0046920776, 0.037872314, 0.004878998, 0.0066184998, -0.0051193237, -0.025741577, -0.050750732, 0.0070533752, -0.02217102, ...
http://arxiv.org/abs/2309.01885v1
QuantEase: Optimization-based Quantization for Language Models -- An Efficient and Intuitive Algorithm
Kayhan Behdin, Ayan Acharya, Aman Gupta, Sathiya Keerthi, Rahul Mazumder
stat.ML, cs.CL, cs.LG
With the rising popularity of Large Language Models (LLMs), there has been an increasing interest in compression techniques that enable their efficient deployment. This study focuses on the Post-Training Quantization (PTQ) of LLMs. Drawing from recent advances, our work introduces QuantEase, a layer-wise quantization f...
http://arxiv.org/pdf/2309.01885v1
2023-09-05T01:39:09
2023-09-05T01:39:09
{ "title": [ -0.049438477, -0.0035915375, -0.051361084, -0.04586792, 0.010353088, -0.0026454926, -0.059570312, -0.017745972, -0.0074806213, 0.05758667, 0.011474609, -0.0056533813, 0.016784668, -0.017364502, -0.06274414, 0.018585205, -0.005958557, ...
http://arxiv.org/abs/2309.12361v1
ChatGPT Assisting Diagnosis of Neuro-ophthalmology Diseases Based on Case Reports
Yeganeh Madadi, Mohammad Delsoz, Priscilla A. Lao, Joseph W. Fong, TJ Hollingsworth, Malik Y. Kahook, Siamak Yousefi
cs.CY, cs.AI, cs.CL
Objective: To evaluate the efficiency of large language models (LLMs) such as ChatGPT to assist in diagnosing neuro-ophthalmic diseases based on detailed case descriptions. Methods: We selected 22 different case reports of neuro-ophthalmic diseases from a publicly available online database. These cases included a wide ...
http://arxiv.org/pdf/2309.12361v1
2023-09-05T00:44:23
2023-09-05T00:44:23
{ "title": [ 0.013122559, -0.03503418, -0.030822754, -0.03955078, -0.030822754, -0.007358551, -0.035705566, -0.0007791519, -0.016143799, 0.038391113, -0.027511597, 0.011909485, 0.0082092285, -0.03781128, -0.012710571, -0.0037326813, -0.00013387203, ...
http://arxiv.org/abs/2309.01868v1
On the Planning, Search, and Memorization Capabilities of Large Language Models
Yunhao Yang, Anshul Tomar
cs.CL, cs.AI
The rapid advancement of large language models, such as the Generative Pre-trained Transformer (GPT) series, has had significant implications across various disciplines. In this study, we investigate the potential of the state-of-the-art large language model (GPT-4) for planning tasks. We explore its effectiveness in m...
http://arxiv.org/pdf/2309.01868v1
2023-09-05T00:19:31
2023-09-05T00:19:31
{ "title": [ -0.02330017, -0.013633728, -0.016937256, -0.06463623, -0.011390686, 0.0064849854, -0.04815674, 0.010353088, 0.0021762848, 0.033355713, -0.023910522, -0.0039520264, -0.0031909943, -0.008285522, -0.042144775, -0.0011730194, -0.0030975342, ...
http://arxiv.org/abs/2309.01860v2
Attention-Driven Multi-Modal Fusion: Enhancing Sign Language Recognition and Translation
Zaber Ibn Abdul Hakim, Rasman Mubtasim Swargo, Muhammad Abdullah Adnan
cs.CV, cs.CL, cs.LG
In this paper, we devise a mechanism for the addition of multi-modal information with an existing pipeline for continuous sign language recognition and translation. In our procedure, we have incorporated optical flow information with RGB images to enrich the features with movement-related information. This work studies...
http://arxiv.org/pdf/2309.01860v2
2023-09-06T13:06:45
2023-09-04T23:31:29
{ "title": [ -0.015731812, 0.014434814, -0.024841309, -0.033843994, -0.015129089, 0.00092697144, -0.047729492, -0.019454956, 0.011711121, 0.03717041, -0.04748535, 0.0019607544, 0.027526855, -0.008895874, -0.04345703, -0.02017212, -0.011558533, 0....
http://arxiv.org/abs/2309.02466v1
Minimal Effective Theory for Phonotactic Memory: Capturing Local Correlations due to Errors in Speech
Paul Myles Eugenio
eess.AS, cs.CL, cs.SD
Spoken language evolves constrained by the economy of speech, which depends on factors such as the structure of the human mouth. This gives rise to local phonetic correlations in spoken words. Here we demonstrate that these local correlations facilitate the learning of spoken words by reducing their information content...
http://arxiv.org/pdf/2309.02466v1
2023-09-04T22:11:26
2023-09-04T22:11:26
{ "title": [ 0.003944397, 0.006351471, -0.034973145, -0.06402588, 0.010543823, -0.0046691895, -0.02017212, 0.0060768127, -0.006855011, 0.05230713, 0.006038666, 0.008117676, 0.0040779114, -0.023910522, -0.061584473, 0.006011963, -0.019134521, 0.05...
http://arxiv.org/abs/2309.01826v2
One Wide Feedforward is All You Need
Telmo Pessoa Pires, António V. Lopes, Yannick Assogba, Hendra Setiawan
cs.CL, cs.AI
The Transformer architecture has two main non-embedding components: Attention and the Feed Forward Network (FFN). Attention captures interdependencies between words regardless of their position, while the FFN non-linearly transforms each input token independently. In this work we explore the role of the FFN, and find t...
http://arxiv.org/pdf/2309.01826v2
2023-10-21T08:33:44
2023-09-04T21:30:21
{ "title": [ 0.028549194, -0.038116455, -0.029327393, -0.07171631, -0.035308838, -0.01576233, -0.034851074, 0.052581787, 0.015899658, 0.043121338, -0.010009766, -0.005542755, -0.0072631836, -0.025131226, -0.02810669, -0.017044067, 0.02468872, -0....
http://arxiv.org/abs/2309.02465v1
Towards Foundational AI Models for Additive Manufacturing: Language Models for G-Code Debugging, Manipulation, and Comprehension
Anushrut Jignasu, Kelly Marshall, Baskar Ganapathysubramanian, Aditya Balu, Chinmay Hegde, Adarsh Krishnamurthy
cs.SE, cs.AI, cs.CL, cs.LG
3D printing or additive manufacturing is a revolutionary technology that enables the creation of physical objects from digital models. However, the quality and accuracy of 3D printing depend on the correctness and efficiency of the G-code, a low-level numerical control programming language that instructs 3D printers ho...
http://arxiv.org/pdf/2309.02465v1
2023-09-04T21:22:28
2023-09-04T21:22:28
{ "title": [ -0.057159424, -0.010169983, -0.05429077, -0.0317688, -0.02168274, 0.0085372925, -0.064208984, 0.006790161, -0.013252258, 0.023803711, 0.006175995, -0.031707764, 0.028121948, -0.049041748, -0.015197754, -0.0030345917, -0.0006599426, 0...
http://arxiv.org/abs/2309.01812v1
Into the Single Cell Multiverse: an End-to-End Dataset for Procedural Knowledge Extraction in Biomedical Texts
Ruth Dannenfelser, Jeffrey Zhong, Ran Zhang, Vicky Yao
cs.CL
Many of the most commonly explored natural language processing (NLP) information extraction tasks can be thought of as evaluations of declarative knowledge, or fact-based information extraction. Procedural knowledge extraction, i.e., breaking down a described process into a series of steps, has received much less atten...
http://arxiv.org/pdf/2309.01812v1
2023-09-04T21:02:36
2023-09-04T21:02:36
{ "title": [ -0.024887085, -0.022720337, -0.047943115, -0.074157715, -0.020584106, 0.011627197, -0.05819702, 0.023971558, 0.014923096, 0.036224365, -0.0041770935, 0.013061523, 0.0057296753, -0.016220093, -0.044647217, 0.018127441, -0.0000028014183, ...
http://arxiv.org/abs/2309.01809v1
Are Emergent Abilities in Large Language Models just In-Context Learning?
Sheng Lu, Irina Bigoulaeva, Rachneet Sachdeva, Harish Tayyar Madabushi, Iryna Gurevych
cs.CL
Large language models have exhibited emergent abilities, demonstrating exceptional performance across diverse tasks for which they were not explicitly trained, including those that require complex reasoning abilities. The emergence of such abilities carries profound implications for the future direction of research in ...
http://arxiv.org/pdf/2309.01809v1
2023-09-04T20:54:11
2023-09-04T20:54:11
{ "title": [ -0.006149292, -0.004558563, -0.029327393, -0.032043457, -0.03189087, -0.022598267, 0.0033168793, 0.023910522, -0.009864807, 0.027114868, -0.0491333, -0.004299164, 0.009292603, -0.017303467, -0.017059326, -0.030899048, -0.010131836, 0...
http://arxiv.org/abs/2309.01740v2
An Empirical Analysis for Zero-Shot Multi-Label Classification on COVID-19 CT Scans and Uncurated Reports
Ethan Dack, Lorenzo Brigato, Matthew McMurray, Matthias Fontanellaz, Thomas Frauenfelder, Hanno Hoppe, Aristomenis Exadaktylos, Thomas Geiser, Manuela Funke-Chambour, Andreas Christe, Lukas Ebner, Stavroula Mougiakakou
eess.IV, cs.CL, cs.CV, cs.LG
The pandemic resulted in vast repositories of unstructured data, including radiology reports, due to increased medical examinations. Previous research on automated diagnosis of COVID-19 primarily focuses on X-ray images, despite their lower precision compared to computed tomography (CT) scans. In this work, we leverage...
http://arxiv.org/pdf/2309.01740v2
2023-09-06T09:34:53
2023-09-04T17:58:01
{ "title": [ 0.013244629, -0.010345459, -0.0435791, -0.04284668, -0.026473999, -0.031051636, -0.008399963, 0.031311035, -0.035064697, 0.033966064, -0.03414917, 0.029373169, -0.012916565, -0.026428223, 0.0006055832, -0.016571045, 0.00932312, 0.025...
http://arxiv.org/abs/2309.01717v2
Interdisciplinary Fairness in Imbalanced Research Proposal Topic Inference: A Hierarchical Transformer-based Method with Selective Interpolation
Meng Xiao, Min Wu, Ziyue Qiao, Yanjie Fu, Zhiyuan Ning, Yi Du, Yuanchun Zhou
cs.CL, cs.AI
The objective of topic inference in research proposals aims to obtain the most suitable disciplinary division from the discipline system defined by a funding agency. The agency will subsequently find appropriate peer review experts from their database based on this division. Automated topic inference can reduce human e...
http://arxiv.org/pdf/2309.01717v2
2023-09-11T01:33:47
2023-09-04T16:54:49
{ "title": [ -0.009796143, 0.029159546, -0.05166626, -0.020553589, 0.013618469, -0.015731812, -0.031799316, -0.059326172, -0.018310547, 0.010971069, -0.05117798, 0.043029785, -0.0013847351, -0.015327454, -0.0007557869, -0.0236969, -0.042175293, 0...
http://arxiv.org/abs/2309.01715v1
Prompting or Fine-tuning? A Comparative Study of Large Language Models for Taxonomy Construction
Boqi Chen, Fandi Yi, Dániel Varró
cs.CL, cs.LG
Taxonomies represent hierarchical relations between entities, frequently applied in various software modeling and natural language processing (NLP) activities. They are typically subject to a set of structural constraints restricting their content. However, manual taxonomy construction can be time-consuming, incomplete...
http://arxiv.org/pdf/2309.01715v1
2023-09-04T16:53:17
2023-09-04T16:53:17
{ "title": [ -0.020614624, -0.015182495, -0.03475952, -0.06573486, -0.002357483, 0.0005850792, -0.06262207, -0.007888794, -0.0062294006, 0.0048294067, -0.027786255, 0.01272583, -0.016921997, -0.020980835, -0.019546509, -0.008369446, -0.014701843, ...
http://arxiv.org/abs/2309.01686v1
MathAttack: Attacking Large Language Models Towards Math Solving Ability
Zihao Zhou, Qiufeng Wang, Mingyu Jin, Jie Yao, Jianan Ye, Wei Liu, Wei Wang, Xiaowei Huang, Kaizhu Huang
cs.CL
With the boom of Large Language Models (LLMs), the research of solving Math Word Problem (MWP) has recently made great progress. However, there are few studies to examine the security of LLMs in math solving ability. Instead of attacking prompts in the use of LLMs, we propose a MathAttack model to attack MWP samples wh...
http://arxiv.org/pdf/2309.01686v1
2023-09-04T16:02:23
2023-09-04T16:02:23
{ "title": [ -0.008270264, -0.022216797, -0.027755737, -0.020751953, -0.007675171, -0.023742676, 0.010025024, -0.005420685, -0.025772095, 0.02532959, -0.0011730194, 0.01638794, 0.0039749146, -0.021575928, -0.014190674, -0.007209778, -0.016830444, ...
http://arxiv.org/abs/2309.01684v1
CRUISE-Screening: Living Literature Reviews Toolbox
Wojciech Kusa, Petr Knoth, Allan Hanbury
cs.IR, cs.CL, cs.DL
Keeping up with research and finding related work is still a time-consuming task for academics. Researchers sift through thousands of studies to identify a few relevant ones. Automation techniques can help by increasing the efficiency and effectiveness of this task. To this end, we developed CRUISE-Screening, a web-bas...
http://arxiv.org/pdf/2309.01684v1
2023-09-04T15:58:43
2023-09-04T15:58:43
{ "title": [ -0.0050468445, 0.019210815, -0.023406982, 0.0029144287, -0.014770508, -0.020080566, -0.028060913, -0.040863037, -0.021560669, 0.04006958, -0.03326416, 0.02394104, 0.006462097, -0.032684326, -0.020004272, -0.014854431, 0.019165039, -0...
http://arxiv.org/abs/2309.01669v1
Donkii: Can Annotation Error Detection Methods Find Errors in Instruction-Tuning Datasets?
Leon Weber-Genzel, Robert Litschko, Ekaterina Artemova, Barbara Plank
cs.CL
Instruction-tuning has become an integral part of training pipelines for Large Language Models (LLMs) and has been shown to yield strong performance gains. In an orthogonal line of research, Annotation Error Detection (AED) has emerged as a tool for detecting quality issues of gold-standard labels. But so far, the appl...
http://arxiv.org/pdf/2309.01669v1
2023-09-04T15:34:02
2023-09-04T15:34:02
{ "title": [ -0.057128906, -0.011672974, -0.023498535, -0.0024871826, -0.030395508, -0.009124756, -0.030273438, -0.011741638, 0.03062439, 0.011299133, -0.045410156, -0.0050239563, -0.016738892, -0.025985718, -0.038238525, -0.012458801, -0.009674072, ...
http://arxiv.org/abs/2309.01664v1
Fine-grained Affective Processing Capabilities Emerging from Large Language Models
Joost Broekens, Bernhard Hilpert, Suzan Verberne, Kim Baraka, Patrick Gebhard, Aske Plaat
cs.CL, cs.AI, cs.HC
Large language models, in particular generative pre-trained transformers (GPTs), show impressive results on a wide variety of language-related tasks. In this paper, we explore ChatGPT's zero-shot ability to perform affective computing tasks using prompting alone. We show that ChatGPT a) performs meaningful sentiment an...
http://arxiv.org/pdf/2309.01664v1
2023-09-04T15:32:47
2023-09-04T15:32:47
{ "title": [ -0.0066604614, -0.01576233, -0.026260376, -0.063964844, -0.0009756088, -0.008636475, -0.025863647, 0.013702393, 0.03366089, 0.025726318, -0.022064209, -0.02357483, 0.016937256, -0.02243042, -0.061279297, 0.017547607, -0.0034656525, 0...
http://arxiv.org/abs/2309.01660v1
Unveiling Theory of Mind in Large Language Models: A Parallel to Single Neurons in the Human Brain
Mohsen Jamali, Ziv M. Williams, Jing Cai
cs.CL, cs.AI
With their recent development, large language models (LLMs) have been found to exhibit a certain level of Theory of Mind (ToM), a complex cognitive capacity that is related to our conscious mind and that allows us to infer another's beliefs and perspective. While human ToM capabilities are believed to derive from the n...
http://arxiv.org/pdf/2309.01660v1
2023-09-04T15:26:15
2023-09-04T15:26:15
{ "title": [ -0.019012451, -0.0116119385, -0.011276245, -0.01991272, -0.028305054, -0.018920898, -0.012268066, -0.019042969, -0.030166626, 0.036743164, 0.008026123, -0.010604858, 0.033233643, -0.04309082, -0.030471802, 0.0062217712, -0.004676819, ...
http://arxiv.org/abs/2309.01659v1
Evolving linguistic divergence on polarizing social media
Andres Karjus, Christine Cuskley
cs.SI, cs.CL
Language change is influenced by many factors, but often starts from synchronic variation, where multiple linguistic patterns or forms coexist, or where different speech communities use language in increasingly different ways. Besides regional or economic reasons, communities may form and segregate based on political a...
http://arxiv.org/pdf/2309.01659v1
2023-09-04T15:21:55
2023-09-04T15:21:55
{ "title": [ 0.006729126, 0.0158844, -0.023757935, -0.021072388, -0.003408432, -0.017044067, 0.0030212402, -0.019561768, -0.053619385, 0.04269409, -0.022872925, 0.03265381, -0.008598328, -0.01600647, -0.017959595, -0.008895874, 0.00623703, 0.0550...
http://arxiv.org/abs/2309.01645v1
Exploring the effectiveness of ChatGPT-based feedback compared with teacher feedback and self-feedback: Evidence from Chinese to English translation
Siyi Cao, Linping Zhong
cs.CL
ChatGPT,a cutting-edge AI-powered Chatbot,can quickly generate responses on given commands. While it was reported that ChatGPT had the capacity to deliver useful feedback, it is still unclear about its effectiveness compared with conventional feedback approaches,such as teacher feedback (TF) and self-feedback (SF). To ...
http://arxiv.org/pdf/2309.01645v1
2023-09-04T14:54:39
2023-09-04T14:54:39
{ "title": [ -0.033721924, -0.011856079, 0.009925842, 0.024856567, -0.038146973, -0.0044021606, -0.016418457, -0.037963867, -0.021072388, 0.03062439, -0.0059394836, 0.029205322, -0.0035209656, -0.019317627, -0.03967285, -0.0003478527, -0.028762817, ...
http://arxiv.org/abs/2309.01618v1
Critical Behavioral Traits Foster Peer Engagement in Online Mental Health Communities
Aseem Srivastava, Tanya Gupta, Alison Cerezo, Sarah Peregrine, Lord, Md Shad Akhtar, Tanmoy Chakraborty
cs.CL
Online Mental Health Communities (OMHCs), such as Reddit, have witnessed a surge in popularity as go-to platforms for seeking information and support in managing mental health needs. Platforms like Reddit offer immediate interactions with peers, granting users a vital space for seeking mental health assistance. However...
http://arxiv.org/pdf/2309.01618v1
2023-09-04T14:00:12
2023-09-04T14:00:12
{ "title": [ 0.029159546, -0.0017061234, -0.015571594, -0.029220581, 0.011871338, -0.023651123, 0.014099121, -0.007610321, -0.037506104, 0.050964355, 0.013793945, 0.024719238, 0.03930664, -0.055755615, 0.004211426, 0.02279663, -0.0013828278, 0.03...
http://arxiv.org/abs/2309.01606v1
Geo-Encoder: A Chunk-Argument Bi-Encoder Framework for Chinese Geographic Re-Ranking
Yong Cao, Ruixue Ding, Boli Chen, Xianzhi Li, Min Chen, Daniel Hershcovich, Pengjun Xie, Fei Huang
cs.CL
Chinese geographic re-ranking task aims to find the most relevant addresses among retrieved candidates, which is crucial for location-related services such as navigation maps. Unlike the general sentences, geographic contexts are closely intertwined with geographical concepts, from general spans (e.g., province) to spe...
http://arxiv.org/pdf/2309.01606v1
2023-09-04T13:44:50
2023-09-04T13:44:50
{ "title": [ 0.020065308, 0.0016078949, 0.019515991, -0.031280518, 0.0019483566, 0.010314941, -0.0046653748, -0.035461426, 0.024337769, 0.053649902, -0.03161621, 0.0018539429, 0.023864746, -0.011054993, -0.043060303, 0.009101868, -0.00869751, 0.0...
http://arxiv.org/abs/2309.12360v1
Efficient Social Choice via NLP and Sampling
Lior Ashkenazy, Nimrod Talmon
cs.CY, cs.AI, cs.CL, cs.LG
Attention-Aware Social Choice tackles the fundamental conflict faced by some agent communities between their desire to include all members in the decision making processes and the limited time and attention that are at the disposal of the community members. Here, we investigate a combination of two techniques for atten...
http://arxiv.org/pdf/2309.12360v1
2023-09-04T13:30:31
2023-09-04T13:30:31
{ "title": [ 0.018997192, -0.002216339, -0.007926941, -0.042877197, 0.013549805, -0.012741089, 0.015144348, -0.026916504, 0.004875183, 0.00018906593, -0.062408447, 0.02772522, -0.035339355, 0.009483337, -0.02204895, -0.013267517, -0.029220581, 0....
http://arxiv.org/abs/2309.01576v1
A Comparative Analysis of Pretrained Language Models for Text-to-Speech
Marcel Granero-Moya, Penny Karanasou, Sri Karlapati, Bastian Schnell, Nicole Peinelt, Alexis Moinet, Thomas Drugman
cs.CL, cs.SD, eess.AS
State-of-the-art text-to-speech (TTS) systems have utilized pretrained language models (PLMs) to enhance prosody and create more natural-sounding speech. However, while PLMs have been extensively researched for natural language understanding (NLU), their impact on TTS has been overlooked. In this study, we aim to addre...
http://arxiv.org/pdf/2309.01576v1
2023-09-04T13:02:27
2023-09-04T13:02:27
{ "title": [ -0.0014028549, -0.0047950745, -0.022491455, -0.068115234, 0.0051727295, -0.00026869774, -0.059326172, 0.012214661, -0.014587402, 0.05545044, -0.01260376, 0.011505127, 0.021408081, -0.010749817, -0.050811768, 0.011688232, -0.0061073303, ...
http://arxiv.org/abs/2309.01538v2
ChatRule: Mining Logical Rules with Large Language Models for Knowledge Graph Reasoning
Linhao Luo, Jiaxin Ju, Bo Xiong, Yuan-Fang Li, Gholamreza Haffari, Shirui Pan
cs.AI, cs.CL
Logical rules are essential for uncovering the logical connections between relations, which could improve the reasoning performance and provide interpretable results on knowledge graphs (KGs). Although there have been many efforts to mine meaningful logical rules over KGs, existing methods suffer from the computational...
http://arxiv.org/pdf/2309.01538v2
2023-09-13T13:00:23
2023-09-04T11:38:02
{ "title": [ -0.053100586, -0.04144287, -0.07183838, -0.07489014, -0.020019531, 0.021499634, -0.09063721, 0.0357666, -0.006263733, 0.027740479, -0.025894165, 0.012916565, -0.00920105, -0.009498596, -0.031982422, -0.012954712, -0.019607544, 0.0548...
http://arxiv.org/abs/2309.01522v2
What are Public Concerns about ChatGPT? A Novel Self-Supervised Neural Topic Model Tells You
Rui Wang, Xing Liu, Yanan Wang, Haiping Huang
cs.CL
The recently released artificial intelligence conversational agent, ChatGPT, has gained significant attention in academia and real life. A multitude of early ChatGPT users eagerly explore its capabilities and share their opinions on it via social media. Both user queries and social media posts express public concerns r...
http://arxiv.org/pdf/2309.01522v2
2023-10-26T08:15:30
2023-09-04T11:05:10
{ "title": [ 0.004096985, 0.0037345886, 0.010299683, -0.0048675537, -0.026351929, -0.0015211105, -0.028457642, -0.04675293, -0.027236938, 0.026885986, -0.002729416, 0.017990112, -0.0256958, -0.018951416, -0.027145386, -0.0101623535, -0.008079529, ...
http://arxiv.org/abs/2309.01456v2
LLM and Infrastructure as a Code use case
Thibault Chanus, Michael Aubertin
cs.CL
Cloud computing and the evolution of management methodologies such as Lean Management or Agile entail a profound transformation in both system construction and maintenance approaches. These practices are encompassed within the term "DevOps." This descriptive approach to an information system or application, alongside t...
http://arxiv.org/pdf/2309.01456v2
2023-11-02T09:00:49
2023-09-04T09:05:17
{ "title": [ -0.022857666, -0.041412354, -0.03173828, -0.038604736, -0.07019043, -0.0014467239, -0.048675537, 0.040618896, -0.014022827, 0.009529114, -0.036315918, -0.015335083, -0.003993988, -0.035369873, 0.03125, -0.02407837, -0.0078048706, 0.0...
http://arxiv.org/abs/2309.01455v1
NumHG: A Dataset for Number-Focused Headline Generation
Jian-Tao Huang, Chung-Chi Chen, Hen-Hsen Huang, Hsin-Hsi Chen
cs.CL
Headline generation, a key task in abstractive summarization, strives to condense a full-length article into a succinct, single line of text. Notably, while contemporary encoder-decoder models excel based on the ROUGE metric, they often falter when it comes to the precise generation of numerals in headlines. We identif...
http://arxiv.org/pdf/2309.01455v1
2023-09-04T09:03:53
2023-09-04T09:03:53
{ "title": [ -0.031188965, -0.0032634735, -0.04675293, -0.052978516, 0.00047659874, -0.00014817715, -0.034973145, -0.009094238, -0.011070251, 0.02583313, -0.009490967, 0.02748108, -0.051727295, -0.008201599, -0.04055786, 0.004436493, -0.019226074, ...
http://arxiv.org/abs/2309.01446v2
Open Sesame! Universal Black Box Jailbreaking of Large Language Models
Raz Lapid, Ron Langberg, Moshe Sipper
cs.CL, cs.CV, cs.NE
Large language models (LLMs), designed to provide helpful and safe responses, often rely on alignment techniques to align with user intent and social guidelines. Unfortunately, this alignment can be exploited by malicious actors seeking to manipulate an LLM's outputs for unintended purposes. In this paper we introduce ...
http://arxiv.org/pdf/2309.01446v2
2023-09-17T13:19:11
2023-09-04T08:54:20
{ "title": [ -0.025161743, -0.014915466, -0.053771973, -0.062683105, -0.0049438477, -0.007118225, -0.05645752, 0.003227234, -0.019042969, 0.01966858, -0.008857727, 0.0058250427, 0.018997192, -0.03302002, -0.011672974, 0.004512787, -0.006252289, 0...
http://arxiv.org/abs/2309.02459v2
Text-Only Domain Adaptation for End-to-End Speech Recognition through Down-Sampling Acoustic Representation
Jiaxu Zhu, Weinan Tong, Yaoxun Xu, Changhe Song, Zhiyong Wu, Zhao You, Dan Su, Dong Yu, Helen Meng
cs.SD, cs.CL, eess.AS
Mapping two modalities, speech and text, into a shared representation space, is a research topic of using text-only data to improve end-to-end automatic speech recognition (ASR) performance in new domains. However, the length of speech representation and text representation is inconsistent. Although the previous method...
http://arxiv.org/pdf/2309.02459v2
2023-10-07T04:23:48
2023-09-04T08:52:59
{ "title": [ -0.01335144, 0.0021686554, -0.04837036, -0.06976318, 0.007789612, -0.009719849, -0.03125, 0.02545166, 0.008102417, 0.043029785, -0.007255554, 0.0066986084, 0.02331543, -0.021255493, -0.038391113, 0.02041626, -0.011878967, 0.039978027...
http://arxiv.org/abs/2309.01437v2
SememeASR: Boosting Performance of End-to-End Speech Recognition against Domain and Long-Tailed Data Shift with Sememe Semantic Knowledge
Jiaxu Zhu, Changhe Song, Zhiyong Wu, Helen Meng
cs.SD, cs.CL, eess.AS
Recently, excellent progress has been made in speech recognition. However, pure data-driven approaches have struggled to solve the problem in domain-mismatch and long-tailed data. Considering that knowledge-driven approaches can help data-driven approaches alleviate their flaws, we introduce sememe-based semantic knowl...
http://arxiv.org/pdf/2309.01437v2
2023-10-07T04:30:26
2023-09-04T08:35:05
{ "title": [ -0.0055236816, 0.0073509216, -0.10046387, -0.07330322, 0.025177002, -0.030197144, -0.029449463, 0.02709961, -0.034179688, 0.030319214, -0.0055236816, 0.021621704, -0.0061187744, -0.040985107, -0.0035686493, -0.014823914, -0.007583618, ...
http://arxiv.org/abs/2309.01431v1
Benchmarking Large Language Models in Retrieval-Augmented Generation
Jiawei Chen, Hongyu Lin, Xianpei Han, Le Sun
cs.CL
Retrieval-Augmented Generation (RAG) is a promising approach for mitigating the hallucination of large language models (LLMs). However, existing research lacks rigorous evaluation of the impact of retrieval-augmented generation on different large language models, which make it challenging to identify the potential bott...
http://arxiv.org/pdf/2309.01431v1
2023-09-04T08:28:44
2023-09-04T08:28:44
{ "title": [ -0.021377563, -0.006843567, -0.029037476, -0.06713867, -0.0002323389, 0.01448822, -0.06713867, 0.0004312992, 0.017868042, 0.05239868, 0.0067749023, 0.009635925, 0.017684937, -0.017822266, -0.051361084, 0.031143188, -0.01171875, 0.019...
http://arxiv.org/abs/2309.01413v1
Hateful Messages: A Conversational Data Set of Hate Speech produced by Adolescents on Discord
Jan Fillies, Silvio Peikert, Adrian Paschke
cs.CL
With the rise of social media, a rise of hateful content can be observed. Even though the understanding and definitions of hate speech varies, platforms, communities, and legislature all acknowledge the problem. Therefore, adolescents are a new and active group of social media users. The majority of adolescents experie...
http://arxiv.org/pdf/2309.01413v1
2023-09-04T07:48:52
2023-09-04T07:48:52
{ "title": [ 0.036712646, 0.0012922287, -0.040771484, -0.03677368, 0.012969971, -0.035308838, 0.002576828, -0.0058898926, 0.021820068, 0.044311523, -0.007270813, -0.0042152405, 0.045776367, -0.034088135, -0.029464722, 0.005809784, -0.035247803, 0...
http://arxiv.org/abs/2309.01398v2
Zero-shot information extraction from radiological reports using ChatGPT
Danqing Hu, Bing Liu, Xiaofeng Zhu, Xudong Lu, Nan Wu
cs.CL
Electronic health records contain an enormous amount of valuable information, but many are recorded in free text. Information extraction is the strategy to transform the sequence of characters into structured data, which can be employed for secondary analysis. However, the traditional information extraction components,...
http://arxiv.org/pdf/2309.01398v2
2023-09-07T01:36:08
2023-09-04T07:00:26
{ "title": [ -0.010292053, -0.010673523, -0.0390625, -0.056152344, -0.01473999, 0.005268097, -0.034362793, 0.014472961, -0.0071258545, 0.03527832, 0.004310608, 0.012191772, -0.032592773, -0.02381897, -0.049621582, 0.016998291, -0.024169922, 0.043...
http://arxiv.org/abs/2309.01370v1
ReOnto: A Neuro-Symbolic Approach for Biomedical Relation Extraction
Monika Jain, Kuldeep Singh, Raghava Mutharaju
cs.CL, cs.AI, cs.IR, cs.LG
Relation Extraction (RE) is the task of extracting semantic relationships between entities in a sentence and aligning them to relations defined in a vocabulary, which is generally in the form of a Knowledge Graph (KG) or an ontology. Various approaches have been proposed so far to address this task. However, applying t...
http://arxiv.org/pdf/2309.01370v1
2023-09-04T05:36:58
2023-09-04T05:36:58
{ "title": [ -0.008682251, -0.0072898865, -0.01600647, -0.06378174, -0.00026845932, 0.026046753, -0.03894043, -0.019897461, 0.0335083, 0.030853271, 0.010307312, -0.019577026, -0.0030097961, -0.027877808, -0.066467285, 0.014320374, -0.012039185, 0...
http://arxiv.org/abs/2309.01352v1
Self-driven Grounding: Large Language Model Agents with Automatical Language-aligned Skill Learning
Shaohui Peng, Xing Hu, Qi Yi, Rui Zhang, Jiaming Guo, Di Huang, Zikang Tian, Ruizhi Chen, Zidong Du, Qi Guo, Yunji Chen, Ling Li
cs.CL, cs.AI
Large language models (LLMs) show their powerful automatic reasoning and planning capability with a wealth of semantic knowledge about the human world. However, the grounding problem still hinders the applications of LLMs in the real-world environment. Existing studies try to fine-tune the LLM or utilize pre-defined be...
http://arxiv.org/pdf/2309.01352v1
2023-09-04T04:31:24
2023-09-04T04:31:24
{ "title": [ -0.023239136, 0.008270264, -0.04345703, -0.043548584, 0.0006213188, -0.012336731, -0.026565552, 0.005268097, -0.014854431, 0.042510986, -0.013908386, 0.006713867, 0.017211914, -0.010253906, -0.03475952, 0.0033855438, -0.013305664, 0....
http://arxiv.org/abs/2309.01339v1
UniSA: Unified Generative Framework for Sentiment Analysis
Zaijing Li, Ting-En Lin, Yuchuan Wu, Meng Liu, Fengxiao Tang, Ming Zhao, Yongbin Li
cs.CL, cs.AI, cs.CV, cs.MM
Sentiment analysis is a crucial task that aims to understand people's emotional states and predict emotional categories based on multimodal information. It consists of several subtasks, such as emotion recognition in conversation (ERC), aspect-based sentiment analysis (ABSA), and multimodal sentiment analysis (MSA). Ho...
http://arxiv.org/pdf/2309.01339v1
2023-09-04T03:49:30
2023-09-04T03:49:30
{ "title": [ -0.00096321106, -0.014030457, -0.02911377, -0.06512451, 0.0055618286, 0.00087594986, -0.03869629, -0.011444092, 0.033325195, 0.0368042, -0.025253296, 0.0027275085, -0.0068740845, -0.0045394897, -0.07183838, 0.010444641, -0.012039185, ...
http://arxiv.org/abs/2309.03227v1
Learning a Patent-Informed Biomedical Knowledge Graph Reveals Technological Potential of Drug Repositioning Candidates
Yongseung Jegal, Jaewoong Choi, Jiho Lee, Ki-Su Park, Seyoung Lee, Janghyeok Yoon
cs.AI, cs.CL, cs.LG, q-bio.QM
Drug repositioning-a promising strategy for discovering new therapeutic uses for existing drugs-has been increasingly explored in the computational science literature using biomedical databases. However, the technological potential of drug repositioning candidates has often been overlooked. This study presents a novel ...
http://arxiv.org/pdf/2309.03227v1
2023-09-04T02:30:19
2023-09-04T02:30:19
{ "title": [ -0.008560181, 0.00038409233, -0.032348633, -0.018325806, -0.01727295, -0.022964478, -0.021011353, -0.014472961, -0.011451721, 0.0034866333, -0.027145386, -0.000269413, -0.030471802, -0.023986816, -0.027328491, -0.030548096, 0.01210022, ...
http://arxiv.org/abs/2309.09980v1
Code Representation Pre-training with Complements from Program Executions
Jiabo Huang, Jianyu Zhao, Yuyang Rong, Yiwen Guo, Yifeng He, Hao Chen
cs.SE, cs.AI, cs.CL
Large language models (LLMs) for natural language processing have been grafted onto programming language modeling for advancing code intelligence. Although it can be represented in the text format, code is syntactically more rigorous in order to be properly compiled or interpreted to perform a desired set of behaviors ...
http://arxiv.org/pdf/2309.09980v1
2023-09-04T01:57:22
2023-09-04T01:57:22
{ "title": [ -0.0030269623, -0.0037155151, -0.031066895, -0.054595947, -0.013336182, -0.011444092, -0.035705566, 0.0029563904, -0.017486572, 0.03805542, -0.016098022, 0.018600464, 0.0006098747, -0.032104492, 0.017837524, 0.0019607544, 0.02330017, ...
http://arxiv.org/abs/2309.01256v1
BDC-Adapter: Brownian Distance Covariance for Better Vision-Language Reasoning
Yi Zhang, Ce Zhang, Zihan Liao, Yushun Tang, Zhihai He
cs.CV, cs.CL
Large-scale pre-trained Vision-Language Models (VLMs), such as CLIP and ALIGN, have introduced a new paradigm for learning transferable visual representations. Recently, there has been a surge of interest among researchers in developing lightweight fine-tuning techniques to adapt these models to downstream visual tasks...
http://arxiv.org/pdf/2309.01256v1
2023-09-03T19:45:02
2023-09-03T19:45:02
{ "title": [ -0.017318726, -0.011558533, -0.047821045, -0.05114746, -0.0013017654, -0.02520752, -0.023742676, 0.006298065, 0.032226562, 0.010025024, -0.032196045, -0.0042648315, 0.009689331, -0.01676941, -0.037017822, -0.004459381, 0.017501831, 0...
http://arxiv.org/abs/2309.01249v1
Large AI Model Empowered Multimodal Semantic Communications
Feibo Jiang, Yubo Peng, Li Dong, Kezhi Wang, Kun Yang, Cunhua Pan, Xiaohu You
cs.AI, cs.CL, cs.LG
Multimodal signals, including text, audio, image and video, can be integrated into Semantic Communication (SC) for providing an immersive experience with low latency and high quality at the semantic level. However, the multimodal SC has several challenges, including data heterogeneity, semantic ambiguity, and signal fa...
http://arxiv.org/pdf/2309.01249v1
2023-09-03T19:24:34
2023-09-03T19:24:34
{ "title": [ -0.038330078, 0.0054740906, -0.070739746, -0.05505371, -0.006839752, -0.025543213, -0.040924072, 0.0075416565, -0.006385803, 0.035980225, -0.005569458, -0.012542725, 0.033416748, -0.044769287, -0.006996155, -0.0018463135, 0.0017967224, ...
http://arxiv.org/abs/2309.01245v1
Representations Matter: Embedding Modes of Large Language Models using Dynamic Mode Decomposition
Mohamed Akrout
cs.CL, cs.AI
Existing large language models (LLMs) are known for generating "hallucinated" content, namely a fabricated text of plausibly looking, yet unfounded, facts. To identify when these hallucination scenarios occur, we examine the properties of the generated text in the embedding space. Specifically, we draw inspiration from...
http://arxiv.org/pdf/2309.01245v1
2023-09-03T19:10:18
2023-09-03T19:10:18
{ "title": [ -0.016693115, -0.006008148, -0.0345459, -0.07446289, 0.011276245, 0.009254456, -0.05227661, 0.010528564, 0.007972717, 0.039886475, 0.014602661, -0.003452301, 0.026733398, -0.037597656, -0.025360107, 0.023757935, -0.0041236877, 0.0559...
http://arxiv.org/abs/2309.01219v2
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
Yue Zhang, Yafu Li, Leyang Cui, Deng Cai, Lemao Liu, Tingchen Fu, Xinting Huang, Enbo Zhao, Yu Zhang, Yulong Chen, Longyue Wang, Anh Tuan Luu, Wei Bi, Freda Shi, Shuming Shi
cs.CL, cs.AI, cs.CY, cs.LG
While large language models (LLMs) have demonstrated remarkable capabilities across a range of downstream tasks, a significant concern revolves around their propensity to exhibit hallucinations: LLMs occasionally generate content that diverges from the user input, contradicts previously generated context, or misaligns ...
http://arxiv.org/pdf/2309.01219v2
2023-09-24T16:03:24
2023-09-03T16:56:48
{ "title": [ -0.021102905, 0.0006351471, -0.014823914, -0.027160645, -0.017868042, -0.004058838, -0.024902344, -0.021636963, 0.019714355, 0.03805542, -0.029373169, 0.013084412, 0.013977051, -0.0041923523, -0.0552063, 0.016799927, -0.027435303, 0....
http://arxiv.org/abs/2309.01196v1
A Visual Interpretation-Based Self-Improved Classification System Using Virtual Adversarial Training
Shuai Jiang, Sayaka Kamei, Chen Li, Shengzhe Hou, Yasuhiko Morimoto
cs.CL, cs.AI
The successful application of large pre-trained models such as BERT in natural language processing has attracted more attention from researchers. Since the BERT typically acts as an end-to-end black box, classification systems based on it usually have difficulty in interpretation and low robustness. This paper proposes...
http://arxiv.org/pdf/2309.01196v1
2023-09-03T15:07:24
2023-09-03T15:07:24
{ "title": [ -0.000341177, 0.03173828, -0.0390625, -0.04425049, 0.020187378, -0.014053345, -0.015144348, -0.024017334, -0.023330688, 0.02633667, -0.012214661, -0.0056610107, -0.044311523, -0.022445679, -0.06384277, -0.039276123, 0.010437012, 0.03...
http://arxiv.org/abs/2309.01157v1
Large Language Models for Generative Recommendation: A Survey and Visionary Discussions
Lei Li, Yongfeng Zhang, Dugang Liu, Li Chen
cs.IR, cs.AI, cs.CL
Recent years have witnessed the wide adoption of large language models (LLM) in different fields, especially natural language processing and computer vision. Such a trend can also be observed in recommender systems (RS). However, most of related work treat LLM as a component of the conventional recommendation pipeline ...
http://arxiv.org/pdf/2309.01157v1
2023-09-03T12:33:47
2023-09-03T12:33:47
{ "title": [ -0.009971619, -0.015060425, -0.03564453, -0.05407715, -0.0023765564, -0.0004415512, -0.035186768, 0.0075645447, -0.00028133392, 0.03567505, -0.02166748, 0.017028809, 0.0040245056, -0.009017944, -0.036712646, 0.006313324, -0.011482239, ...
http://arxiv.org/abs/2309.01131v1
Attention Where It Matters: Rethinking Visual Document Understanding with Selective Region Concentration
Haoyu Cao, Changcun Bao, Chaohu Liu, Huang Chen, Kun Yin, Hao Liu, Yinsong Liu, Deqiang Jiang, Xing Sun
cs.CV, cs.CL
We propose a novel end-to-end document understanding model called SeRum (SElective Region Understanding Model) for extracting meaningful information from document images, including document analysis, retrieval, and office automation. Unlike state-of-the-art approaches that rely on multi-stage technical schemes and ar...
http://arxiv.org/pdf/2309.01131v1
2023-09-03T10:14:34
2023-09-03T10:14:34
{ "title": [ -0.016433716, 0.0026550293, -0.025222778, -0.05065918, 0.0062294006, -0.015304565, -0.021224976, -0.011154175, 0.02923584, 0.015083313, -0.05343628, -0.0045204163, -0.013458252, 0.0049209595, -0.06048584, -0.012886047, 0.0026454926, ...
http://arxiv.org/abs/2309.01114v1
MedChatZH: a Better Medical Adviser Learns from Better Instructions
Yang Tan, Mingchen Li, Zijie Huang, Huiqun Yu, Guisheng Fan
cs.CL, cs.AI
Generative large language models (LLMs) have shown great success in various applications, including question-answering (QA) and dialogue systems. However, in specialized domains like traditional Chinese medical QA, these models may perform unsatisfactorily without fine-tuning on domain-specific datasets. To address thi...
http://arxiv.org/pdf/2309.01114v1
2023-09-03T08:08:15
2023-09-03T08:08:15
{ "title": [ 0.050628662, -0.011833191, -0.026641846, -0.06530762, -0.006767273, -0.01008606, -0.032104492, -0.023971558, -0.057037354, 0.033111572, 0.0033016205, 0.02507019, 0.0077171326, -0.051757812, 0.029525757, -0.01448822, 0.030380249, -0.0...
http://arxiv.org/abs/2309.01105v2
A Study on the Implementation of Generative AI Services Using an Enterprise Data-Based LLM Application Architecture
Cheonsu Jeong
cs.AI, cs.CL
This study presents a method for implementing generative AI services by utilizing the Large Language Models (LLM) application architecture. With recent advancements in generative AI technology, LLMs have gained prominence across various domains. In this context, the research addresses the challenge of information scarc...
http://arxiv.org/pdf/2309.01105v2
2023-09-18T11:36:50
2023-09-03T07:03:17
{ "title": [ -0.046875, -0.0107803345, -0.06774902, -0.02142334, -0.012519836, -0.030166626, -0.038330078, -0.0067481995, 0.016220093, 0.02986145, -0.04663086, 0.023208618, 0.0033874512, -0.023162842, -0.022720337, -0.014152527, -0.011077881, 0.0...
http://arxiv.org/abs/2309.01071v1
Business Process Text Sketch Automation Generation Using Large Language Model
Rui Zhu, Quanzhou Hu, Wenxin Li, Honghao Xiao, Chaogang Wang, Zixin Zhou
cs.CL
Business Process Management (BPM) is gaining increasing attention as it has the potential to cut costs while boosting output and quality. Business process document generation is a crucial stage in BPM. However, due to a shortage of datasets, data-driven deep learning techniques struggle to deliver the expected results....
http://arxiv.org/pdf/2309.01071v1
2023-09-03T04:19:02
2023-09-03T04:19:02
{ "title": [ -0.058441162, -0.022705078, -0.062347412, -0.03869629, -0.035125732, 0.009391785, -0.09106445, 0.015960693, -0.0046157837, 0.05630493, -0.027038574, -0.007255554, 0.043121338, -0.014755249, -0.036193848, -0.014587402, -0.0005340576, ...
http://arxiv.org/abs/2309.06358v1
Generative Data Augmentation using LLMs improves Distributional Robustness in Question Answering
Arijit Ghosh Chowdhury, Aman Chadha
cs.CL, cs.AI
Robustness in Natural Language Processing continues to be a pertinent issue, where state of the art models under-perform under naturally shifted distributions. In the context of Question Answering, work on domain adaptation methods continues to be a growing body of research. However, very little attention has been give...
http://arxiv.org/pdf/2309.06358v1
2023-09-03T03:27:06
2023-09-03T03:27:06
{ "title": [ 0.0045318604, 0.003616333, -0.054260254, -0.04840088, -0.004169464, -0.023803711, -0.046813965, 0.0004105568, 0.020980835, 0.017364502, -0.033111572, 0.028137207, -0.009002686, -0.011222839, -0.048583984, 0.007083893, -0.0042266846, ...
http://arxiv.org/abs/2309.01029v2
Explainability for Large Language Models: A Survey
Haiyan Zhao, Hanjie Chen, Fan Yang, Ninghao Liu, Huiqi Deng, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, Mengnan Du
cs.CL, cs.AI, cs.LG
Large language models (LLMs) have demonstrated impressive capabilities in natural language processing. However, their internal mechanisms are still unclear and this lack of transparency poses unwanted risks for downstream applications. Therefore, understanding and explaining these models is crucial for elucidating thei...
http://arxiv.org/pdf/2309.01029v2
2023-09-17T02:24:30
2023-09-02T22:14:26
{ "title": [ -0.037200928, -0.0072669983, -0.015365601, -0.092041016, 0.002664566, 0.041534424, -0.070129395, 0.022766113, 0.0021572113, -0.0016489029, -0.014228821, -0.008148193, 0.0019683838, -0.018310547, -0.022384644, 0.00052690506, -0.028411865,...
http://arxiv.org/abs/2309.01026v2
Zero-Shot Recommendations with Pre-Trained Large Language Models for Multimodal Nudging
Rachel M. Harrison, Anton Dereventsov, Anton Bibin
cs.AI, cs.CL, cs.IR, cs.LG, cs.MM
We present a method for zero-shot recommendation of multimodal non-stationary content that leverages recent advancements in the field of generative AI. We propose rendering inputs of different modalities as textual descriptions and to utilize pre-trained LLMs to obtain their numerical representations by computing seman...
http://arxiv.org/pdf/2309.01026v2
2023-10-01T02:57:42
2023-09-02T21:29:53
{ "title": [ 0.004447937, 0.0047454834, -0.029647827, -0.060516357, -0.0078048706, -0.0093688965, -0.035827637, -0.0000795722, -0.009017944, 0.03555298, -0.010017395, 0.01600647, 0.008224487, -0.020187378, -0.018539429, -0.00017297268, -0.009338379, ...
http://arxiv.org/abs/2309.00986v1
ModelScope-Agent: Building Your Customizable Agent System with Open-source Large Language Models
Chenliang Li, Hehong Chen, Ming Yan, Weizhou Shen, Haiyang Xu, Zhikai Wu, Zhicheng Zhang, Wenmeng Zhou, Yingda Chen, Chen Cheng, Hongzhu Shi, Ji Zhang, Fei Huang, Jingren Zhou
cs.CL
Large language models (LLMs) have recently demonstrated remarkable capabilities to comprehend human intentions, engage in reasoning, and design planning-like behavior. To further unleash the power of LLMs to accomplish complex tasks, there is a growing trend to build agent framework that equips LLMs, such as ChatGPT, w...
http://arxiv.org/pdf/2309.00986v1
2023-09-02T16:50:30
2023-09-02T16:50:30
{ "title": [ -0.04458618, -0.011146545, -0.052124023, -0.06506348, -0.004924774, 0.018844604, -0.06903076, -0.0070762634, -0.01626587, 0.03427124, -0.027557373, -0.0015468597, 0.03817749, -0.023147583, -0.018417358, -0.0029888153, -0.00016319752, ...
http://arxiv.org/abs/2309.00952v1
Bridge Diffusion Model: bridge non-English language-native text-to-image diffusion model with English communities
Shanyuan Liu, Dawei Leng, Yuhui Yin
cs.CL, cs.AI
Text-to-Image generation (TTI) technologies are advancing rapidly, especially in the English language communities. However, English-native TTI models inherently carry biases from English world centric training data, which creates a dilemma for development of other language-native TTI models. One common choice is fine-t...
http://arxiv.org/pdf/2309.00952v1
2023-09-02T14:30:56
2023-09-02T14:30:56
{ "title": [ -0.0027694702, 0.026031494, -0.008361816, 0.039489746, -0.011482239, -0.028686523, 0.0064849854, -0.041900635, -0.013641357, 0.05831909, -0.028152466, 0.024047852, 0.044830322, -0.029663086, 0.01058197, 0.0055732727, -0.014709473, 0....
http://arxiv.org/abs/2309.00949v1
Multilingual Text Representation
Fahim Faisal
cs.CL
Modern NLP breakthrough includes large multilingual models capable of performing tasks across more than 100 languages. State-of-the-art language models came a long way, starting from the simple one-hot representation of words capable of performing tasks like natural language understanding, common-sense reasoning, or qu...
http://arxiv.org/pdf/2309.00949v1
2023-09-02T14:21:22
2023-09-02T14:21:22
{ "title": [ -0.0013542175, -0.009033203, -0.00010687113, -0.052856445, -0.033203125, 0.004371643, -0.071777344, -0.0067863464, 0.020553589, 0.06878662, -0.011383057, -0.007091522, 0.03955078, -0.024383545, -0.015319824, 0.03062439, 0.020263672, ...
http://arxiv.org/abs/2309.05668v1
Studying the impacts of pre-training using ChatGPT-generated text on downstream tasks
Sarthak Anand
cs.CL, cs.AI
In recent times, significant advancements have been witnessed in the field of language models, particularly with the emergence of Large Language Models (LLMs) that are trained on vast amounts of data extracted from internet archives. These LLMs, such as ChatGPT, have become widely accessible, allowing users to generate...
http://arxiv.org/pdf/2309.05668v1
2023-09-02T12:56:15
2023-09-02T12:56:15
{ "title": [ 0.0018262863, -0.007537842, -0.018310547, -0.040130615, -0.01737976, -0.0075569153, -0.030929565, -0.019302368, -0.02961731, 0.029876709, 0.0014457703, 0.041137695, -0.016601562, -0.020233154, -0.0058174133, 0.0034656525, -0.014533997, ...
http://arxiv.org/abs/2309.00917v2
Knowledge Graph Embeddings for Multi-Lingual Structured Representations of Radiology Reports
Tom van Sonsbeek, Xiantong Zhen, Marcel Worring
cs.CL, cs.AI, 68T07
The way we analyse clinical texts has undergone major changes over the last years. The introduction of language models such as BERT led to adaptations for the (bio)medical domain like PubMedBERT and ClinicalBERT. These models rely on large databases of archived medical documents. While performing well in terms of accur...
http://arxiv.org/pdf/2309.00917v2
2023-09-14T14:25:37
2023-09-02T11:46:41
{ "title": [ -0.02609253, -0.019607544, -0.054626465, -0.08843994, -0.014923096, 0.0020999908, -0.079956055, 0.029647827, 0.00655365, 0.02720642, -0.016159058, -0.0038833618, -0.01499176, -0.01727295, -0.029403687, -0.004009247, 0.010627747, 0.01...
http://arxiv.org/abs/2309.00916v1
BLSP: Bootstrapping Language-Speech Pre-training via Behavior Alignment of Continuation Writing
Chen Wang, Minpeng Liao, Zhongqiang Huang, Jinliang Lu, Junhong Wu, Yuchen Liu, Chengqing Zong, Jiajun Zhang
cs.CL, cs.SD, eess.AS
The emergence of large language models (LLMs) has sparked significant interest in extending their remarkable language capabilities to speech. However, modality alignment between speech and text still remains an open problem. Current solutions can be categorized into two strategies. One is a cascaded approach where outp...
http://arxiv.org/pdf/2309.00916v1
2023-09-02T11:46:05
2023-09-02T11:46:05
{ "title": [ -0.00945282, -0.013824463, -0.0357666, -0.08618164, -0.019012451, 0.009407043, -0.0859375, 0.010688782, -0.04345703, 0.024459839, -0.015274048, 0.037841797, 0.014328003, -0.02494812, 0.0008196831, 0.0063972473, -0.006137848, 0.049224...
http://arxiv.org/abs/2309.00857v2
Evaluating Transformer's Ability to Learn Mildly Context-Sensitive Languages
Shunjie Wang, Shane Steinert-Threlkeld
cs.CL
Despite the fact that Transformers perform well in NLP tasks, recent studies suggest that self-attention is theoretically limited in learning even some regular and context-free languages. These findings motivated us to think about their implications in modeling natural language, which is hypothesized to be mildly conte...
http://arxiv.org/pdf/2309.00857v2
2023-10-19T11:41:58
2023-09-02T08:17:29
{ "title": [ 0.02456665, 0.023544312, -0.026733398, 0.015258789, -0.011245728, -0.028030396, 0.00032544136, -0.036132812, -0.0335083, 0.035949707, -0.040893555, -0.013946533, 0.016845703, -0.026290894, 0.0031929016, -0.027694702, -0.009605408, 0....
http://arxiv.org/abs/2309.00841v1
LeanContext: Cost-Efficient Domain-Specific Question Answering Using LLMs
Md Adnan Arefeen, Biplob Debnath, Srimat Chakradhar
cs.CL, cs.AI, cs.IR
Question-answering (QA) is a significant application of Large Language Models (LLMs), shaping chatbot capabilities across healthcare, education, and customer service. However, widespread LLM integration presents a challenge for small businesses due to the high expenses of LLM API usage. Costs rise rapidly when domain-s...
http://arxiv.org/pdf/2309.00841v1
2023-09-02T06:33:18
2023-09-02T06:33:18
{ "title": [ -0.020614624, -0.029083252, -0.04711914, -0.072753906, -0.023849487, -0.0029201508, -0.06402588, 0.01878357, -0.0071983337, 0.0131073, -0.03878784, 0.029830933, -0.004886627, -0.010269165, -0.018981934, -0.0102005005, -0.00945282, 0....
http://arxiv.org/abs/2309.00789v1
LinkTransformer: A Unified Package for Record Linkage with Transformer Language Models
Abhishek Arora, Melissa Dell
cs.CL
Linking information across sources is fundamental to a variety of analyses in social science, business, and government. While large language models (LLMs) offer enormous promise for improving record linkage in noisy datasets, in many domains approximate string matching packages in popular softwares such as R and Stata ...
http://arxiv.org/pdf/2309.00789v1
2023-09-02T01:45:27
2023-09-02T01:45:27
{ "title": [ -0.020141602, -0.01625061, -0.062286377, -0.101989746, 0.0074043274, 0.004627228, -0.093444824, 0.0065078735, 0.011062622, 0.024551392, -0.021606445, -0.0026550293, 0.03363037, -0.0345459, -0.01713562, 0.020477295, -0.005794525, 0.05...
http://arxiv.org/abs/2309.00779v1
Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties
Taylor Sorensen, Liwei Jiang, Jena Hwang, Sydney Levine, Valentina Pyatkin, Peter West, Nouha Dziri, Ximing Lu, Kavel Rao, Chandra Bhagavatula, Maarten Sap, John Tasioulas, Yejin Choi
cs.CL, cs.AI
Human values are crucial to human decision-making. Value pluralism is the view that multiple correct values may be held in tension with one another (e.g., when considering lying to a friend to protect their feelings, how does one balance honesty with friendship?). As statistical learners, AI systems fit to averages by ...
http://arxiv.org/pdf/2309.00779v1
2023-09-02T01:24:59
2023-09-02T01:24:59
{ "title": [ -0.014564514, -0.005672455, -0.04537964, -0.019226074, 0.0010681152, -0.032684326, 0.015007019, 0.0027999878, -0.0154953, 0.026062012, -0.021331787, 0.019561768, -0.009407043, -0.02494812, -0.034362793, -0.015625, 0.0007958412, 0.033...
http://arxiv.org/abs/2309.00770v1
Bias and Fairness in Large Language Models: A Survey
Isabel O. Gallegos, Ryan A. Rossi, Joe Barrow, Md Mehrab Tanjim, Sungchul Kim, Franck Dernoncourt, Tong Yu, Ruiyi Zhang, Nesreen K. Ahmed
cs.CL, cs.AI, cs.CY, cs.LG
Rapid advancements of large language models (LLMs) have enabled the processing, understanding, and generation of human-like text, with increasing integration into systems that touch our social sphere. Despite this success, these models can learn, perpetuate, and amplify harmful social biases. In this paper, we present ...
http://arxiv.org/pdf/2309.00770v1
2023-09-02T00:32:55
2023-09-02T00:32:55
{ "title": [ -0.04043579, -0.01071167, -0.06262207, -0.05834961, 0.0030555725, -0.004131317, -0.054504395, 0.014144897, 0.010002136, 0.030395508, -0.04031372, -0.0062408447, 0.025283813, -0.010360718, -0.037902832, 0.0029277802, -0.036102295, 0.0...
http://arxiv.org/abs/2309.00754v1
Efficient RLHF: Reducing the Memory Usage of PPO
Michael Santacroce, Yadong Lu, Han Yu, Yuanzhi Li, Yelong Shen
cs.LG, cs.AI, cs.CL
Reinforcement Learning with Human Feedback (RLHF) has revolutionized language modeling by aligning models with human preferences. However, the RL stage, Proximal Policy Optimization (PPO), requires over 3x the memory of Supervised Fine-Tuning (SFT), making it infeasible to use for most practitioners. To address this is...
http://arxiv.org/pdf/2309.00754v1
2023-09-01T22:57:20
2023-09-01T22:57:20
{ "title": [ -0.003332138, -0.01789856, -0.056030273, -0.097595215, -0.036468506, -0.0107040405, -0.066101074, 0.02986145, -0.052642822, 0.04647827, 0.037322998, 0.0011816025, 0.02494812, -0.06451416, 0.022720337, 0.009246826, 0.004081726, 0.0003...
http://arxiv.org/abs/2309.00751v1
Let the Models Respond: Interpreting Language Model Detoxification Through the Lens of Prompt Dependence
Daniel Scalena, Gabriele Sarti, Malvina Nissim, Elisabetta Fersini
cs.CL
Due to language models' propensity to generate toxic or hateful responses, several techniques were developed to align model generations with users' preferences. Despite the effectiveness of such methods in improving the safety of model interactions, their impact on models' internal processes is still poorly understood....
http://arxiv.org/pdf/2309.00751v1
2023-09-01T22:26:06
2023-09-01T22:26:06
{ "title": [ -0.01499939, -0.0023899078, -0.039215088, -0.06915283, 0.0017366409, -0.003168106, -0.048736572, 0.000579834, -0.009208679, 0.009124756, -0.003627777, -0.0017614365, -0.030380249, -0.041870117, -0.04623413, -0.007972717, 0.0016126633, ...
http://arxiv.org/abs/2309.00743v1
Language-Conditioned Change-point Detection to Identify Sub-Tasks in Robotics Domains
Divyanshu Raj, Chitta Baral, Nakul Gopalan
cs.RO, cs.AI, cs.CL
In this work, we present an approach to identify sub-tasks within a demonstrated robot trajectory using language instructions. We identify these sub-tasks using language provided during demonstrations as guidance to identify sub-segments of a longer robot trajectory. Given a sequence of natural language instructions an...
http://arxiv.org/pdf/2309.00743v1
2023-09-01T21:40:34
2023-09-01T21:40:34
{ "title": [ -0.03149414, 0.0030212402, -0.03741455, -0.01361084, -0.0013961792, -0.027435303, -0.021270752, -0.030639648, 0.004638672, 0.009490967, -0.02746582, 0.018981934, -0.0012168884, -0.020446777, -0.0064926147, -0.0025081635, 0.0039711, 0...
http://arxiv.org/abs/2309.00723v2
Contextual Biasing of Named-Entities with Large Language Models
Chuanneng Sun, Zeeshan Ahmed, Yingyi Ma, Zhe Liu, Lucas Kabela, Yutong Pang, Ozlem Kalinli
cs.CL, cs.AI, cs.LG, cs.SD, eess.AS, 68T10, I.2.7
This paper studies contextual biasing with Large Language Models (LLMs), where during second-pass rescoring additional contextual information is provided to a LLM to boost Automatic Speech Recognition (ASR) performance. We propose to leverage prompts for a LLM without fine tuning during rescoring which incorporate a bi...
http://arxiv.org/pdf/2309.00723v2
2023-09-22T02:06:10
2023-09-01T20:15:48
{ "title": [ 0.013206482, 0.0061454773, -0.045074463, -0.070617676, 0.026901245, -0.014083862, -0.028823853, -0.013931274, 0.026168823, 0.03729248, -0.058044434, 0.006061554, 0.037353516, 0.003742218, -0.025466919, 0.006881714, -0.027618408, 0.05...
http://arxiv.org/abs/2309.00615v1
Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following
Ziyu Guo, Renrui Zhang, Xiangyang Zhu, Yiwen Tang, Xianzheng Ma, Jiaming Han, Kexin Chen, Peng Gao, Xianzhi Li, Hongsheng Li, Pheng-Ann Heng
cs.CV, cs.AI, cs.CL, cs.LG, cs.MM
We introduce Point-Bind, a 3D multi-modality model aligning point clouds with 2D image, language, audio, and video. Guided by ImageBind, we construct a joint embedding space between 3D and multi-modalities, enabling many promising applications, e.g., any-to-3D generation, 3D embedding arithmetic, and 3D open-world unde...
http://arxiv.org/pdf/2309.00615v1
2023-09-01T17:59:47
2023-09-01T17:59:47
{ "title": [ -0.046447754, -0.00166893, -0.061187744, -0.013748169, -0.027648926, -0.01876831, -0.0395813, -0.025268555, -0.037017822, 0.025665283, -0.023330688, 0.02633667, -0.0058403015, -0.024810791, 0.006072998, -0.013824463, 0.018112183, -0....
http://arxiv.org/abs/2309.00614v2
Baseline Defenses for Adversarial Attacks Against Aligned Language Models
Neel Jain, Avi Schwarzschild, Yuxin Wen, Gowthami Somepalli, John Kirchenbauer, Ping-yeh Chiang, Micah Goldblum, Aniruddha Saha, Jonas Geiping, Tom Goldstein
cs.LG, cs.CL, cs.CR
As Large Language Models quickly become ubiquitous, it becomes critical to understand their security vulnerabilities. Recent work shows that text optimizers can produce jailbreaking prompts that bypass moderation and alignment. Drawing from the rich body of work on adversarial machine learning, we approach these attack...
http://arxiv.org/pdf/2309.00614v2
2023-09-04T17:47:36
2023-09-01T17:59:44
{ "title": [ -0.030700684, -0.01134491, -0.034301758, -0.05117798, 0.009536743, 0.0043258667, -0.032409668, 0.0127334595, -0.036224365, 0.04031372, 0.017288208, 0.02041626, -0.013626099, -0.020309448, -0.03717041, 0.004257202, -0.010917664, 0.033...
http://arxiv.org/abs/2309.00667v1
Taken out of context: On measuring situational awareness in LLMs
Lukas Berglund, Asa Cooper Stickland, Mikita Balesni, Max Kaufmann, Meg Tong, Tomasz Korbak, Daniel Kokotajlo, Owain Evans
cs.CL, cs.LG
We aim to better understand the emergence of `situational awareness' in large language models (LLMs). A model is situationally aware if it's aware that it's a model and can recognize whether it's currently in testing or deployment. Today's LLMs are tested for safety and alignment before they are deployed. An LLM could ...
http://arxiv.org/pdf/2309.00667v1
2023-09-01T17:27:37
2023-09-01T17:27:37
{ "title": [ -0.007888794, 0.016540527, -0.036895752, -0.0043258667, -0.007472992, -0.02949524, -0.007888794, -0.047821045, 0.015960693, 0.011070251, -0.07507324, 0.011978149, -0.033447266, -0.011749268, -0.03756714, -0.039642334, -0.008384705, -...
http://arxiv.org/abs/2309.00424v4
Learning Speech Representation From Contrastive Token-Acoustic Pretraining
Chunyu Qiang, Hao Li, Yixin Tian, Ruibo Fu, Tao Wang, Longbiao Wang, Jianwu Dang
eess.AS, cs.AI, cs.CL, cs.SD
For fine-grained generation and recognition tasks such as minimally-supervised text-to-speech (TTS), voice conversion (VC), and automatic speech recognition (ASR), the intermediate representations extracted from speech should serve as a "bridge" between text and acoustic information, containing information from both mo...
http://arxiv.org/pdf/2309.00424v4
2023-09-27T08:41:16
2023-09-01T12:35:43
{ "title": [ 0.025314331, 0.0014829636, -0.020614624, -0.07104492, 0.0019721985, -0.0087509155, -0.027053833, 0.008644104, -0.0026416779, 0.03793335, -0.023773193, 0.0211792, -0.00869751, -0.0178833, -0.04046631, -0.009490967, -0.008331299, 0.051...
http://arxiv.org/abs/2309.00386v1
Satisfiability Checking of Multi-Variable TPTL with Unilateral Intervals Is PSPACE-Complete
Shankara Narayanan Krishna, Khushraj Nanik Madnani, Rupak Majumdar, Paritosh K. Pandya
cs.LO, cs.CL, cs.FL, F.4; F.4.3; F.1.1
We investigate the decidability of the ${0,\infty}$ fragment of Timed Propositional Temporal Logic (TPTL). We show that the satisfiability checking of TPTL$^{0,\infty}$ is PSPACE-complete. Moreover, even its 1-variable fragment (1-TPTL$^{0,\infty}$) is strictly more expressive than Metric Interval Temporal Logic (MITL)...
http://arxiv.org/pdf/2309.00386v1
2023-09-01T10:49:19
2023-09-01T10:49:19
{ "title": [ 0.00856781, 0.0039978027, -0.0011444092, -0.07122803, -0.02784729, 0.0099487305, -0.099243164, -0.013702393, -0.008506775, 0.0012493134, -0.0385437, 0.0075950623, -0.005466461, -0.030166626, 0.03515625, -0.017990112, -0.0009570122, -...
http://arxiv.org/abs/2309.00384v2
BatchPrompt: Accomplish more with less
Jianzhe Lin, Maurice Diesendruck, Liang Du, Robin Abraham
cs.CL
As the ever-increasing token limits of large language models (LLMs) have enabled long context as input, prompting with single data samples might no longer an efficient way. A straightforward strategy improving efficiency is to batch data within the token limit (e.g., 8k for gpt-3.5-turbo; 32k for GPT-4), which we call ...
http://arxiv.org/pdf/2309.00384v2
2023-09-05T23:03:12
2023-09-01T10:44:36
{ "title": [ 0.03668213, -0.008583069, -0.07458496, -0.078125, -0.010726929, -0.031707764, -0.033996582, 0.028244019, -0.046447754, 0.029693604, -0.022735596, -0.015716553, 0.013214111, -0.046417236, 0.017288208, -0.03152466, 0.01890564, 0.027694...
http://arxiv.org/abs/2309.00378v1
Long-Term Memorability On Advertisements
Harini S I, Somesh Singh, Yaman K Singla, Aanisha Bhattacharyya, Veeky Baths, Changyou Chen, Rajiv Ratn Shah, Balaji Krishnamurthy
cs.CL, cs.CV, cs.HC
Marketers spend billions of dollars on advertisements but to what end? At the purchase time, if customers cannot recognize a brand for which they saw an ad, the money spent on the ad is essentially wasted. Despite its importance in marketing, until now, there has been no study on the memorability of ads in the ML liter...
http://arxiv.org/pdf/2309.00378v1
2023-09-01T10:27:04
2023-09-01T10:27:04
{ "title": [ 0.06561279, 0.011062622, -0.0021152496, -0.034057617, -0.019760132, -0.040374756, 0.012138367, -0.0079422, 0.014572144, 0.0579834, -0.022613525, -0.0040016174, -0.0071029663, -0.031829834, -0.021560669, 0.005203247, 0.019805908, 0.01...
http://arxiv.org/abs/2309.03223v1
Examining the Effectiveness of Chatbots in Gathering Family History Information in Comparison to the Standard In-Person Interview-Based Approach
Kieron Drumm, Vincent Tran
cs.HC, cs.CL, cs.LG
One of the most common things that a genealogist is tasked with is the gathering of a person's initial family history, normally via in-person interviews or with the use of a platform such as ancestry.com, as this can provide a strong foundation upon which a genealogist may build. However, the ability to conduct these i...
http://arxiv.org/pdf/2309.03223v1
2023-09-01T10:09:09
2023-09-01T10:09:09
{ "title": [ 0.056610107, -0.021530151, 0.0034332275, -0.017211914, -0.020263672, -0.030014038, 0.006259918, -0.018417358, 0.0025024414, 0.052734375, -0.03277588, 0.02468872, 0.019744873, -0.0098724365, -0.010238647, -0.018569946, 0.010932922, 0....
http://arxiv.org/abs/2309.00368v1
When Do Discourse Markers Affect Computational Sentence Understanding?
Ruiqi Li, Liesbeth Allein, Damien Sileo, Marie-Francine Moens
cs.CL
The capabilities and use cases of automatic natural language processing (NLP) have grown significantly over the last few years. While much work has been devoted to understanding how humans deal with discourse connectives, this phenomenon is understudied in computational systems. Therefore, it is important to put NLP mo...
http://arxiv.org/pdf/2309.00368v1
2023-09-01T09:54:28
2023-09-01T09:54:28
{ "title": [ -0.0029354095, 0.0062675476, 0.038726807, -0.021057129, -0.04031372, 0.003900528, -0.0256958, -0.03753662, -0.008850098, 0.011489868, -0.010276794, -0.005783081, -0.0057907104, -0.030334473, -0.020248413, -0.009162903, -0.016723633, ...
http://arxiv.org/abs/2309.00359v3
Large Content And Behavior Models To Understand, Simulate, And Optimize Content And Behavior
Ashmit Khandelwal, Aditya Agrawal, Aanisha Bhattacharyya, Yaman K Singla, Somesh Singh, Uttaran Bhattacharya, Ishita Dasgupta, Stefano Petrangeli, Rajiv Ratn Shah, Changyou Chen, Balaji Krishnamurthy
cs.CL, cs.CV
Shannon, in his seminal paper introducing information theory, divided the communication into three levels: technical, semantic, and effectivenss. While the technical level is concerned with accurate reconstruction of transmitted symbols, the semantic and effectiveness levels deal with the inferred meaning and its effec...
http://arxiv.org/pdf/2309.00359v3
2023-10-26T12:18:51
2023-09-01T09:34:49
{ "title": [ 0.0022792816, 0.002922058, -0.028518677, -0.050567627, 0.0009560585, -0.022323608, -0.04220581, -0.0069389343, -0.025360107, 0.01146698, -0.03186035, 0.008842468, 0.013519287, -0.034057617, 0.045898438, -0.015136719, 0.020462036, -0....
http://arxiv.org/abs/2309.00312v3
Insights Into the Nutritional Prevention of Macular Degeneration based on a Comparative Topic Modeling Approach
Lucas Cassiel Jacaruso
cs.CL
Topic modeling and text mining are subsets of Natural Language Processing (NLP) with relevance for conducting meta-analysis (MA) and systematic review (SR). For evidence synthesis, the above NLP methods are conventionally used for topic-specific literature searches or extracting values from reports to automate essentia...
http://arxiv.org/pdf/2309.00312v3
2023-11-12T03:53:50
2023-09-01T07:53:28
{ "title": [ -0.015045166, -0.008888245, -0.033416748, -0.04888916, 0.0057907104, 0.002746582, -0.02696228, -0.009925842, -0.028533936, 0.03225708, 0.017227173, 0.03515625, -0.016448975, -0.02835083, -0.015419006, 0.00012469292, 0.010635376, -0.0...
http://arxiv.org/abs/2309.00284v1
Enhancing the vocal range of single-speaker singing voice synthesis with melody-unsupervised pre-training
Shaohuan Zhou, Xu Li, Zhiyong Wu, Ying Shan, Helen Meng
cs.SD, cs.CL, eess.AS
The single-speaker singing voice synthesis (SVS) usually underperforms at pitch values that are out of the singer's vocal range or associated with limited training samples. Based on our previous work, this work proposes a melody-unsupervised multi-speaker pre-training method conducted on a multi-singer dataset to enhan...
http://arxiv.org/pdf/2309.00284v1
2023-09-01T06:40:41
2023-09-01T06:40:41
{ "title": [ 0.012046814, -0.015403748, -0.054016113, -0.10015869, -0.015083313, -0.009590149, -0.053497314, 0.047943115, -0.033721924, 0.04763794, -0.012710571, 0.013496399, 0.02848816, -0.04159546, -0.012229919, 0.00008624792, -0.017669678, 0.0...
http://arxiv.org/abs/2309.00267v1
RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback
Harrison Lee, Samrat Phatale, Hassan Mansoor, Kellie Lu, Thomas Mesnard, Colton Bishop, Victor Carbune, Abhinav Rastogi
cs.CL, cs.AI, cs.LG
Reinforcement learning from human feedback (RLHF) is effective at aligning large language models (LLMs) to human preferences, but gathering high quality human preference labels is a key bottleneck. We conduct a head-to-head comparison of RLHF vs. RL from AI Feedback (RLAIF) - a technique where preferences are labeled b...
http://arxiv.org/pdf/2309.00267v1
2023-09-01T05:53:33
2023-09-01T05:53:33
{ "title": [ -0.02456665, -0.0041160583, -0.024810791, -0.018310547, 0.007537842, -0.012252808, 0.008705139, -0.031021118, -0.027648926, 0.009864807, -0.041229248, 0.020690918, -0.040527344, -0.0031166077, -0.022338867, -0.029388428, -0.0009994507, ...
http://arxiv.org/abs/2309.00254v1
Why do universal adversarial attacks work on large language models?: Geometry might be the answer
Varshini Subhash, Anna Bialas, Weiwei Pan, Finale Doshi-Velez
cs.LG, cs.CL, cs.CR
Transformer based large language models with emergent capabilities are becoming increasingly ubiquitous in society. However, the task of understanding and interpreting their internal workings, in the context of adversarial attacks, remains largely unsolved. Gradient-based universal adversarial attacks have been shown t...
http://arxiv.org/pdf/2309.00254v1
2023-09-01T05:09:49
2023-09-01T05:09:49
{ "title": [ -0.03677368, -0.017532349, -0.036224365, -0.041625977, -0.021347046, -0.013702393, -0.031341553, 0.0046463013, -0.033355713, 0.03552246, 0.0038204193, 0.020996094, 0.010620117, -0.025787354, -0.003238678, 0.0055618286, 0.0056533813, ...
http://arxiv.org/abs/2309.00246v1
Detecting Suicidality in Arabic Tweets Using Machine Learning and Deep Learning Techniques
Asma Abdulsalam, Areej Alhothali, Saleh Al-Ghamdi
cs.CL
Social media platforms have revolutionized traditional communication techniques by enabling people globally to connect instantaneously, openly, and frequently. People use social media to share personal stories and express their opinion. Negative emotions such as thoughts of death, self-harm, and hardship are commonly e...
http://arxiv.org/pdf/2309.00246v1
2023-09-01T04:30:59
2023-09-01T04:30:59
{ "title": [ 0.018981934, 0.008987427, 0.002532959, -0.0574646, 0.021987915, -0.00737381, -0.00856781, -0.018203735, -0.004142761, 0.03942871, -0.019989014, 0.01751709, 0.011550903, -0.024383545, -0.038116455, 0.0019664764, -0.011070251, 0.022430...
http://arxiv.org/abs/2309.00244v1
NeuroSurgeon: A Toolkit for Subnetwork Analysis
Michael A. Lepori, Ellie Pavlick, Thomas Serre
cs.LG, cs.CL
Despite recent advances in the field of explainability, much remains unknown about the algorithms that neural networks learn to represent. Recent work has attempted to understand trained models by decomposing them into functional circuits (Csord\'as et al., 2020; Lepori et al., 2023). To advance this research, we devel...
http://arxiv.org/pdf/2309.00244v1
2023-09-01T04:26:55
2023-09-01T04:26:55
{ "title": [ -0.032348633, 0.0013103485, -0.049957275, -0.035247803, 0.01675415, -0.010505676, -0.026397705, -0.01713562, -0.029342651, 0.03378296, -0.0014019012, 0.000096440315, -0.0038585663, -0.033355713, -0.02268982, -0.0026664734, 0.017547607, ...
http://arxiv.org/abs/2309.00240v1
FactLLaMA: Optimizing Instruction-Following Language Models with External Knowledge for Automated Fact-Checking
Tsun-Hin Cheung, Kin-Man Lam
cs.CL, cs.AI
Automatic fact-checking plays a crucial role in combating the spread of misinformation. Large Language Models (LLMs) and Instruction-Following variants, such as InstructGPT and Alpaca, have shown remarkable performance in various natural language processing tasks. However, their knowledge may not always be up-to-date o...
http://arxiv.org/pdf/2309.00240v1
2023-09-01T04:14:39
2023-09-01T04:14:39
{ "title": [ -0.031051636, -0.018814087, -0.024002075, -0.078430176, -0.014892578, 0.016494751, -0.07305908, -0.0014638901, -0.025802612, 0.030303955, -0.02130127, 0.017791748, -0.0016593933, -0.0060310364, -0.036590576, -0.015701294, -0.040802002, ...
http://arxiv.org/abs/2309.00238v1
ALJP: An Arabic Legal Judgment Prediction in Personal Status Cases Using Machine Learning Models
Salwa Abbara, Mona Hafez, Aya Kazzaz, Areej Alhothali, Alhanouf Alsolami
cs.AI, cs.CL
Legal Judgment Prediction (LJP) aims to predict judgment outcomes based on case description. Several researchers have developed techniques to assist potential clients by predicting the outcome in the legal profession. However, none of the proposed techniques were implemented in Arabic, and only a few attempts were impl...
http://arxiv.org/pdf/2309.00238v1
2023-09-01T04:08:45
2023-09-01T04:08:45
{ "title": [ -0.017486572, -0.035308838, -0.038970947, -0.072753906, -0.016693115, -0.0028839111, -0.042510986, 0.026229858, -0.00011378527, 0.032318115, -0.014785767, 0.0015563965, 0.0051460266, -0.031066895, -0.038970947, -0.0031414032, -0.02207946...
http://arxiv.org/abs/2309.00237v2
Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes
Sunjun Kweon, Junu Kim, Jiyoun Kim, Sujeong Im, Eunbyeol Cho, Seongsu Bae, Jungwoo Oh, Gyubok Lee, Jong Hak Moon, Seng Chan You, Seungjin Baek, Chang Hoon Han, Yoon Bin Jung, Yohan Jo, Edward Choi
cs.CL, cs.AI
The development of large language models tailored for handling patients' clinical notes is often hindered by the limited accessibility and usability of these notes due to strict privacy regulations. To address these challenges, we first create synthetic large-scale clinical notes using publicly available case reports e...
http://arxiv.org/pdf/2309.00237v2
2023-09-06T18:11:15
2023-09-01T04:01:20
{ "title": [ -0.03250122, -0.023117065, -0.03253174, -0.103881836, -0.027511597, 0.03503418, -0.108947754, 0.0234375, -0.015434265, 0.025848389, -0.019470215, 0.0018959045, -0.011604309, -0.015655518, -0.018966675, -0.0104599, 0.009124756, 0.0276...
http://arxiv.org/abs/2309.00236v2
Image Hijacks: Adversarial Images can Control Generative Models at Runtime
Luke Bailey, Euan Ong, Stuart Russell, Scott Emmons
cs.LG, cs.CL, cs.CR
Are foundation models secure from malicious actors? In this work, we focus on the image input to a vision-language model (VLM). We discover image hijacks, adversarial images that control generative models at runtime. We introduce Behaviour Matching, a general method for creating image hijacks, and we use it to explore ...
http://arxiv.org/pdf/2309.00236v2
2023-09-18T17:59:23
2023-09-01T03:53:40
{ "title": [ 0.016418457, -0.014167786, -0.007709503, -0.023605347, -0.007160187, -0.020828247, 0.015113831, -0.036346436, -0.01374054, 0.01802063, -0.011207581, 0.035583496, -0.01966858, -0.019088745, -0.017608643, -0.00076532364, 0.0037822723, ...
http://arxiv.org/abs/2309.00230v1
JoTR: A Joint Transformer and Reinforcement Learning Framework for Dialog Policy Learning
Wai-Chung Kwan, Huimin Wang, Hongru Wang, Zezhong Wang, Xian Wu, Yefeng Zheng, Kam-Fai Wong
cs.CL
Dialogue policy learning (DPL) is a crucial component of dialogue modelling. Its primary role is to determine the appropriate abstract response, commonly referred to as the "dialogue action". Traditional DPL methodologies have treated this as a sequential decision problem, using pre-defined action candidates extracted ...
http://arxiv.org/pdf/2309.00230v1
2023-09-01T03:19:53
2023-09-01T03:19:53
{ "title": [ 0.01802063, 0.0045166016, -0.04067993, -0.074645996, 0.02128601, -0.019683838, -0.03050232, 5.9604645e-7, -0.0362854, 0.015426636, -0.0039634705, 0.022949219, -0.037750244, -0.028060913, -0.0079193115, -0.018859863, -0.027145386, 0.0...
http://arxiv.org/abs/2309.00223v1
The FruitShell French synthesis system at the Blizzard 2023 Challenge
Xin Qi, Xiaopeng Wang, Zhiyong Wang, Wang Liu, Mingming Ding, Shuchen Shi
eess.AS, cs.CL, cs.SD
This paper presents a French text-to-speech synthesis system for the Blizzard Challenge 2023. The challenge consists of two tasks: generating high-quality speech from female speakers and generating speech that closely resembles specific individuals. Regarding the competition data, we conducted a screening process to re...
http://arxiv.org/pdf/2309.00223v1
2023-09-01T02:56:20
2023-09-01T02:56:20
{ "title": [ -0.04269409, -0.016860962, -0.056488037, -0.046844482, -0.03137207, 0.007698059, -0.08483887, 0.025787354, 0.0044898987, 0.034179688, -0.042236328, -0.0043525696, 0.04446411, -0.03112793, 0.020828247, -0.0090408325, 0.015350342, 0.01...
http://arxiv.org/abs/2309.00215v1
Towards Addressing the Misalignment of Object Proposal Evaluation for Vision-Language Tasks via Semantic Grounding
Joshua Feinglass, Yezhou Yang
cs.CV, cs.CL
Object proposal generation serves as a standard pre-processing step in Vision-Language (VL) tasks (image captioning, visual question answering, etc.). The performance of object proposals generated for VL tasks is currently evaluated across all available annotations, a protocol that we show is misaligned - higher scores...
http://arxiv.org/pdf/2309.00215v1
2023-09-01T02:19:41
2023-09-01T02:19:41
{ "title": [ 0.0049705505, 0.002811432, -0.027389526, -0.044189453, 0.0047035217, -0.025604248, -0.019180298, -0.009887695, 0.018920898, 0.0140686035, -0.034210205, 0.0062332153, -0.0029735565, -0.012084961, -0.032836914, -0.020309448, -0.0018663406,...
http://arxiv.org/abs/2309.00208v1
Large Language Models for Semantic Monitoring of Corporate Disclosures: A Case Study on Korea's Top 50 KOSPI Companies
Junwon Sung, Woojin Heo, Yunkyung Byun, Youngsam Kim
cs.CL, cs.AI
In the rapidly advancing domain of artificial intelligence, state-of-the-art language models such as OpenAI's GPT-3.5-turbo and GPT-4 offer unprecedented opportunities for automating complex tasks. This research paper delves into the capabilities of these models for semantically analyzing corporate disclosures in the K...
http://arxiv.org/pdf/2309.00208v1
2023-09-01T01:51:28
2023-09-01T01:51:28
{ "title": [ -0.04925537, -0.018844604, -0.081848145, -0.060394287, -0.022994995, 0.00762558, -0.08795166, 0.015777588, -0.014305115, 0.03111267, -0.048187256, -0.003255844, 0.013038635, -0.0017251968, -0.027954102, -0.027679443, -0.0082473755, 0...
http://arxiv.org/abs/2309.00180v1
Exploring the law of text geographic information
Zhenhua Wang, Daiyu Zhang, Ming Ren, Guang Xu
cs.CL
Textual geographic information is indispensable and heavily relied upon in practical applications. The absence of clear distribution poses challenges in effectively harnessing geographic information, thereby driving our quest for exploration. We contend that geographic information is influenced by human behavior, cogni...
http://arxiv.org/pdf/2309.00180v1
2023-09-01T00:14:51
2023-09-01T00:14:51
{ "title": [ -0.015625, -0.03225708, -0.021499634, 0.009819031, -0.041107178, -0.019607544, -0.005973816, 0.015533447, -0.015403748, 0.050598145, -0.077697754, -0.0020008087, 0.026382446, -0.0037288666, -0.01763916, -0.043762207, 0.008796692, 0.0...
http://arxiv.org/abs/2309.00178v1
Will Sentiment Analysis Need Subculture? A New Data Augmentation Approach
Zhenhua Wang, Simin He, Guang Xu, Ming Ren
cs.CL
The renowned proverb that "The pen is mightier than the sword" underscores the formidable influence wielded by text expressions in shaping sentiments. Indeed, well-crafted written can deeply resonate within cultures, conveying profound sentiments. Nowadays, the omnipresence of the Internet has fostered a subculture tha...
http://arxiv.org/pdf/2309.00178v1
2023-09-01T00:11:56
2023-09-01T00:11:56
{ "title": [ 0.0071144104, -0.005584717, -0.033325195, -0.04840088, 0.024093628, -0.01689148, 0.009170532, -0.006504059, 0.014389038, 0.033477783, -0.045135498, -0.001411438, -0.0057144165, -0.003250122, -0.064331055, -0.0058288574, -0.007789612, ...
http://arxiv.org/abs/2309.00155v1
LLM in the Shell: Generative Honeypots
Muris Sladić, Veronica Valeros, Carlos Catania, Sebastian Garcia
cs.CR, cs.AI, cs.CL
Honeypots are essential tools in cybersecurity. However, most of them (even the high-interaction ones) lack the required realism to engage and fool human attackers. This limitation makes them easily discernible, hindering their effectiveness. This work introduces a novel method to create dynamic and realistic software ...
http://arxiv.org/pdf/2309.00155v1
2023-08-31T22:05:46
2023-08-31T22:05:46
{ "title": [ 0.002456665, -0.010818481, -0.017944336, -0.025665283, -0.04864502, -0.018814087, -0.022277832, -0.012237549, -0.04522705, 0.024353027, -0.02947998, 0.034088135, -0.0037403107, -0.033081055, 0.011779785, -0.027252197, -0.009170532, 0...