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/2308.14951v1
Robust Open-Set Spoken Language Identification and the CU MultiLang Dataset
Mustafa Eyceoz, Justin Lee, Siddharth Pittie, Homayoon Beigi
cs.CL, cs.AI, cs.LG, eess.AS
Most state-of-the-art spoken language identification models are closed-set; in other words, they can only output a language label from the set of classes they were trained on. Open-set spoken language identification systems, however, gain the ability to detect when an input exhibits none of the original languages. In t...
http://arxiv.org/pdf/2308.14951v1
2023-08-29T00:44:27
2023-08-29T00:44:27
{ "title": [ -0.022857666, 0.022247314, -0.061279297, -0.052337646, 0.022827148, -0.0014152527, -0.07324219, -0.015975952, -0.013450623, 0.041107178, -0.01763916, 0.00818634, 0.0043144226, -0.019958496, -0.04034424, -0.0048599243, -0.008293152, 0...
http://arxiv.org/abs/2308.14921v1
Gender bias and stereotypes in Large Language Models
Hadas Kotek, Rikker Dockum, David Q. Sun
cs.CL, cs.CY, cs.LG
Large Language Models (LLMs) have made substantial progress in the past several months, shattering state-of-the-art benchmarks in many domains. This paper investigates LLMs' behavior with respect to gender stereotypes, a known issue for prior models. We use a simple paradigm to test the presence of gender bias, buildin...
http://arxiv.org/pdf/2308.14921v1
2023-08-28T22:32:05
2023-08-28T22:32:05
{ "title": [ -0.04901123, -0.0026435852, -0.06506348, -0.043701172, -0.007217407, -0.021362305, -0.043914795, 0.011329651, 0.0061302185, 0.042388916, -0.037719727, -0.014755249, 0.024429321, -0.0020751953, -0.044708252, -0.0028514862, -0.032958984, ...
http://arxiv.org/abs/2308.14905v1
Neural approaches to spoken content embedding
Shane Settle
cs.CL, cs.SD, eess.AS
Comparing spoken segments is a central operation to speech processing. Traditional approaches in this area have favored frame-level dynamic programming algorithms, such as dynamic time warping, because they require no supervision, but they are limited in performance and efficiency. As an alternative, acoustic word embe...
http://arxiv.org/pdf/2308.14905v1
2023-08-28T21:16:08
2023-08-28T21:16:08
{ "title": [ -0.008270264, 0.00856781, -0.030822754, -0.050048828, 0.012817383, -0.008354187, -0.025772095, 0.0070991516, 0.0051078796, 0.03955078, -0.012870789, -0.0087509155, 0.002073288, -0.00945282, -0.048065186, 0.0048561096, -0.012023926, 0...
http://arxiv.org/abs/2308.14903v1
MEMORY-VQ: Compression for Tractable Internet-Scale Memory
Yury Zemlyanskiy, Michiel de Jong, Luke Vilnis, Santiago Ontañón, William W. Cohen, Sumit Sanghai, Joshua Ainslie
cs.CL
Retrieval augmentation is a powerful but expensive method to make language models more knowledgeable about the world. Memory-based methods like LUMEN pre-compute token representations for retrieved passages to drastically speed up inference. However, memory also leads to much greater storage requirements from storing p...
http://arxiv.org/pdf/2308.14903v1
2023-08-28T21:11:18
2023-08-28T21:11:18
{ "title": [ -0.042877197, -0.022964478, -0.054534912, -0.06945801, 0.008377075, 0.0062942505, -0.01725769, 0.035614014, 0.011222839, 0.04647827, 0.0020656586, -0.013755798, 0.004814148, -0.015060425, -0.057525635, 0.019104004, 0.004169464, 0.008...
http://arxiv.org/abs/2308.14894v1
Multiscale Contextual Learning for Speech Emotion Recognition in Emergency Call Center Conversations
Théo Deschamps-Berger, Lori Lamel, Laurence Devillers
cs.CL, cs.SD, eess.AS
Emotion recognition in conversations is essential for ensuring advanced human-machine interactions. However, creating robust and accurate emotion recognition systems in real life is challenging, mainly due to the scarcity of emotion datasets collected in the wild and the inability to take into account the dialogue cont...
http://arxiv.org/pdf/2308.14894v1
2023-08-28T20:31:45
2023-08-28T20:31:45
{ "title": [ 0.029434204, 0.019485474, -0.057922363, -0.06555176, 0.023925781, -0.0231781, -0.028305054, 0.0060577393, 0.005683899, 0.052642822, -0.03652954, -0.0025405884, 0.03982544, -0.028808594, -0.040161133, -0.012130737, -0.011146545, 0.013...
http://arxiv.org/abs/2308.14873v1
CommunityFish: A Poisson-based Document Scaling With Hierarchical Clustering
Sami Diaf
cs.CL, cs.LG, cs.SI
Document scaling has been a key component in text-as-data applications for social scientists and a major field of interest for political researchers, who aim at uncovering differences between speakers or parties with the help of different probabilistic and non-probabilistic approaches. Yet, most of these techniques are...
http://arxiv.org/pdf/2308.14873v1
2023-08-28T19:52:18
2023-08-28T19:52:18
{ "title": [ -0.031799316, -0.0043296814, -0.055847168, -0.051574707, 0.017196655, -0.005519867, -0.029403687, -0.016403198, 0.009628296, 0.04522705, -0.02470398, 0.014915466, 0.035614014, 0.0032405853, -0.045288086, 0.018035889, 0.0024547577, 0....
http://arxiv.org/abs/2308.14850v1
Attention Visualizer Package: Revealing Word Importance for Deeper Insight into Encoder-Only Transformer Models
Ala Alam Falaki, Robin Gras
cs.CL, cs.AI
This report introduces the Attention Visualizer package, which is crafted to visually illustrate the significance of individual words in encoder-only transformer-based models. In contrast to other methods that center on tokens and self-attention scores, our approach will examine the words and their impact on the final ...
http://arxiv.org/pdf/2308.14850v1
2023-08-28T19:11:52
2023-08-28T19:11:52
{ "title": [ -0.008392334, 0.02418518, -0.026733398, -0.008628845, -0.023422241, -0.027313232, -0.038848877, -0.047088623, -0.005001068, 0.04953003, -0.028671265, 0.0047912598, 0.051971436, -0.027938843, -0.014038086, 0.014266968, 0.01550293, 0.0...
http://arxiv.org/abs/2308.14683v1
Fine-Tuning Llama 2 Large Language Models for Detecting Online Sexual Predatory Chats and Abusive Texts
Thanh Thi Nguyen, Campbell Wilson, Janis Dalins
cs.CL, cs.AI, cs.LG
Detecting online sexual predatory behaviours and abusive language on social media platforms has become a critical area of research due to the growing concerns about online safety, especially for vulnerable populations such as children and adolescents. Researchers have been exploring various techniques and approaches to...
http://arxiv.org/pdf/2308.14683v1
2023-08-28T16:18:50
2023-08-28T16:18:50
{ "title": [ -0.018173218, 0.00504303, -0.016967773, -0.041381836, 0.0072898865, 0.005126953, -0.025665283, -0.0042533875, 0.009796143, 0.0317688, -0.022766113, 0.021118164, 0.013954163, -0.012176514, -0.020431519, -0.0016393661, -0.0146865845, 0...
http://arxiv.org/abs/2308.14669v1
ANER: Arabic and Arabizi Named Entity Recognition using Transformer-Based Approach
Abdelrahman "Boda" Sadallah, Omar Ahmed, Shimaa Mohamed, Omar Hatem, Doaa Hesham, Ahmed H. Yousef
cs.CL, cs.AI
One of the main tasks of Natural Language Processing (NLP), is Named Entity Recognition (NER). It is used in many applications and also can be used as an intermediate step for other tasks. We present ANER, a web-based named entity recognizer for the Arabic, and Arabizi languages. The model is built upon BERT, which is ...
http://arxiv.org/pdf/2308.14669v1
2023-08-28T15:54:48
2023-08-28T15:54:48
{ "title": [ -0.02949524, -0.0043563843, -0.042785645, -0.07104492, 0.024307251, 0.0037899017, -0.04284668, 0.011634827, -0.0015859604, 0.019485474, -0.0184021, 0.014373779, 0.013267517, -0.018203735, -0.02230835, 0.0026664734, 0.011054993, 0.042...
http://arxiv.org/abs/2308.14654v1
Joint Multiple Intent Detection and Slot Filling with Supervised Contrastive Learning and Self-Distillation
Nguyen Anh Tu, Hoang Thi Thu Uyen, Tu Minh Phuong, Ngo Xuan Bach
cs.CL
Multiple intent detection and slot filling are two fundamental and crucial tasks in spoken language understanding. Motivated by the fact that the two tasks are closely related, joint models that can detect intents and extract slots simultaneously are preferred to individual models that perform each task independently. ...
http://arxiv.org/pdf/2308.14654v1
2023-08-28T15:36:33
2023-08-28T15:36:33
{ "title": [ 0.016815186, 0.037719727, -0.039001465, -0.032592773, 0.03274536, -0.024627686, 0.0012950897, -0.056274414, 0.00806427, 0.01499939, -0.0597229, 0.011238098, -0.01939392, -0.025634766, -0.025756836, -0.03353882, -0.023864746, 0.045776...
http://arxiv.org/abs/2308.14641v2
Challenges of GPT-3-based Conversational Agents for Healthcare
Fabian Lechner, Allison Lahnala, Charles Welch, Lucie Flek
cs.CL
The potential to provide patients with faster information access while allowing medical specialists to concentrate on critical tasks makes medical domain dialog agents appealing. However, the integration of large-language models (LLMs) into these agents presents certain limitations that may result in serious consequenc...
http://arxiv.org/pdf/2308.14641v2
2023-08-29T15:48:23
2023-08-28T15:12:34
{ "title": [ 0.012939453, 0.022201538, -0.031951904, -0.05279541, 0.029403687, -0.011566162, -0.04071045, -0.016601562, 0.0040626526, 0.032165527, -0.037231445, 0.005504608, -0.0034008026, -0.016204834, -0.0390625, 0.00447464, -0.014152527, 0.026...
http://arxiv.org/abs/2308.14634v1
Breaking the Bank with ChatGPT: Few-Shot Text Classification for Finance
Lefteris Loukas, Ilias Stogiannidis, Prodromos Malakasiotis, Stavros Vassos
cs.CL, cs.AI, cs.LG, q-fin.CP
We propose the use of conversational GPT models for easy and quick few-shot text classification in the financial domain using the Banking77 dataset. Our approach involves in-context learning with GPT-3.5 and GPT-4, which minimizes the technical expertise required and eliminates the need for expensive GPU computing whil...
http://arxiv.org/pdf/2308.14634v1
2023-08-28T15:04:16
2023-08-28T15:04:16
{ "title": [ -0.023117065, -0.015167236, -0.03149414, -0.02116394, -0.0084991455, -0.01083374, -0.02255249, -0.013198853, -0.0032482147, 0.03012085, -0.03552246, 0.026168823, -0.003414154, 0.0052452087, -0.05041504, -0.020324707, -0.0256958, 0.03...
http://arxiv.org/abs/2308.14608v1
AI in the Gray: Exploring Moderation Policies in Dialogic Large Language Models vs. Human Answers in Controversial Topics
Vahid Ghafouri, Vibhor Agarwal, Yong Zhang, Nishanth Sastry, Jose Such, Guillermo Suarez-Tangil
cs.LG, cs.CL, cs.CY, cs.SI
The introduction of ChatGPT and the subsequent improvement of Large Language Models (LLMs) have prompted more and more individuals to turn to the use of ChatBots, both for information and assistance with decision-making. However, the information the user is after is often not formulated by these ChatBots objectively en...
http://arxiv.org/pdf/2308.14608v1
2023-08-28T14:23:04
2023-08-28T14:23:04
{ "title": [ -0.004425049, -0.0028209686, -0.02368164, -0.038360596, -0.01007843, -0.022064209, -0.016555786, -0.017364502, -0.013259888, 0.018661499, -0.033355713, 0.0155181885, -0.0065460205, -0.011299133, -0.03265381, -0.016860962, -0.025756836, ...
http://arxiv.org/abs/2308.14536v1
Spoken Language Intelligence of Large Language Models for Language Learning
Linkai Peng, Baorian Nuchged, Yingming Gao
cs.CL, cs.AI, cs.LG, cs.SD, eess.AS
People have long hoped for a conversational system that can assist in real-life situations, and recent progress on large language models (LLMs) is bringing this idea closer to reality. While LLMs are often impressive in performance, their efficacy in real-world scenarios that demand expert knowledge remains unclear. LL...
http://arxiv.org/pdf/2308.14536v1
2023-08-28T12:47:41
2023-08-28T12:47:41
{ "title": [ -0.0357666, -0.012382507, -0.030929565, -0.030883789, -0.020446777, -0.005935669, -0.028137207, 0.013824463, -0.007827759, 0.032348633, -0.03451538, -0.004512787, -0.007537842, -0.010955811, -0.041290283, -0.0107421875, -0.012512207, ...
http://arxiv.org/abs/2308.14533v1
A Multi-Task Semantic Decomposition Framework with Task-specific Pre-training for Few-Shot NER
Guanting Dong, Zechen Wang, Jinxu Zhao, Gang Zhao, Daichi Guo, Dayuan Fu, Tingfeng Hui, Chen Zeng, Keqing He, Xuefeng Li, Liwen Wang, Xinyue Cui, Weiran Xu
cs.CL
The objective of few-shot named entity recognition is to identify named entities with limited labeled instances. Previous works have primarily focused on optimizing the traditional token-wise classification framework, while neglecting the exploration of information based on NER data characteristics. To address this iss...
http://arxiv.org/pdf/2308.14533v1
2023-08-28T12:46:21
2023-08-28T12:46:21
{ "title": [ 0.016677856, -0.023040771, -0.02267456, -0.07086182, -0.019317627, 0.01737976, -0.03955078, 0.027389526, -0.011177063, 0.0289917, 0.02017212, 0.024047852, 0.016189575, -0.043273926, 0.013717651, 0.0020542145, 0.025024414, 0.007087707...
http://arxiv.org/abs/2308.14508v1
LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding
Yushi Bai, Xin Lv, Jiajie Zhang, Hongchang Lyu, Jiankai Tang, Zhidian Huang, Zhengxiao Du, Xiao Liu, Aohan Zeng, Lei Hou, Yuxiao Dong, Jie Tang, Juanzi Li
cs.CL
Although large language models (LLMs) demonstrate impressive performance for many language tasks, most of them can only handle texts a few thousand tokens long, limiting their applications on longer sequence inputs, such as books, reports, and codebases. Recent works have proposed methods to improve LLMs' long context ...
http://arxiv.org/pdf/2308.14508v1
2023-08-28T11:53:40
2023-08-28T11:53:40
{ "title": [ -0.028167725, -0.008331299, -0.034210205, -0.08380127, 0.0019350052, 0.008918762, -0.060577393, 0.0015459061, 0.013877869, 0.03173828, -0.0068244934, 0.003873825, -0.007083893, -0.022781372, -0.028884888, 0.014228821, -0.0069885254, ...
http://arxiv.org/abs/2308.14484v1
Multimodal Detection of Social Spambots in Twitter using Transformers
Loukas Ilias, Ioannis Michail Kazelidis, Dimitris Askounis
cs.CL
Although not all bots are malicious, the vast majority of them are responsible for spreading misinformation and manipulating the public opinion about several issues, i.e., elections and many more. Therefore, the early detection of social spambots is crucial. Although there have been proposed methods for detecting bots ...
http://arxiv.org/pdf/2308.14484v1
2023-08-28T10:51:11
2023-08-28T10:51:11
{ "title": [ 0.04586792, 0.02331543, 0.010528564, -0.060699463, 0.012763977, 0.0084991455, -0.024429321, -0.04397583, 0.008995056, 0.04119873, -0.03994751, 0.024597168, 0.006690979, -0.0030288696, -0.04626465, -0.0035057068, -0.021636963, 0.08044...
http://arxiv.org/abs/2308.14482v1
An Empirical Study of Consistency Regularization for End-to-End Speech-to-Text Translation
Pengzhi Gao, Ruiqing Zhang, Zhongjun He, Hua Wu, Haifeng Wang
cs.CL
Consistency regularization methods, such as R-Drop (Liang et al., 2021) and CrossConST (Gao et al., 2023), have achieved impressive supervised and zero-shot performance in the neural machine translation (NMT) field. Can we also boost end-to-end (E2E) speech-to-text translation (ST) by leveraging consistency regularizat...
http://arxiv.org/pdf/2308.14482v1
2023-08-28T10:44:18
2023-08-28T10:44:18
{ "title": [ -0.021072388, 0.012870789, -0.029052734, -0.06286621, 0.013725281, 0.008605957, -0.052093506, 0.011817932, -0.012413025, 0.063964844, 0.00069475174, 0.012649536, 0.035736084, -0.021759033, -0.05340576, 0.02243042, -0.009529114, 0.022...
http://arxiv.org/abs/2308.14436v1
Bridging the KB-Text Gap: Leveraging Structured Knowledge-aware Pre-training for KBQA
Guanting Dong, Rumei Li, Sirui Wang, Yupeng Zhang, Yunsen Xian, Weiran Xu
cs.CL, cs.IR
Knowledge Base Question Answering (KBQA) aims to answer natural language questions with factual information such as entities and relations in KBs. However, traditional Pre-trained Language Models (PLMs) are directly pre-trained on large-scale natural language corpus, which poses challenges for them in understanding and...
http://arxiv.org/pdf/2308.14436v1
2023-08-28T09:22:02
2023-08-28T09:22:02
{ "title": [ -0.039642334, -0.027420044, -0.038024902, -0.066101074, -0.04788208, 0.010932922, -0.08807373, -0.0028953552, 0.012435913, 0.030532837, -0.01689148, -0.004611969, 0.0039253235, -0.02394104, -0.039276123, -0.01777649, 0.012268066, 0.0...
http://arxiv.org/abs/2308.14429v1
Biomedical Entity Linking with Triple-aware Pre-Training
Xi Yan, Cedric Möller, Ricardo Usbeck
cs.CL, cs.AI
Linking biomedical entities is an essential aspect in biomedical natural language processing tasks, such as text mining and question answering. However, a difficulty of linking the biomedical entities using current large language models (LLM) trained on a general corpus is that biomedical entities are scarcely distribu...
http://arxiv.org/pdf/2308.14429v1
2023-08-28T09:06:28
2023-08-28T09:06:28
{ "title": [ 0.019195557, 0.008270264, -0.038848877, -0.105651855, 0.024383545, 0.0010023117, -0.041625977, 0.0007944107, 0.0026664734, 0.036499023, 0.014022827, 0.006706238, 0.018127441, -0.02998352, -0.0155181885, 0.021408081, 0.0012893677, 0.0...
http://arxiv.org/abs/2308.14423v1
GADePo: Graph-Assisted Declarative Pooling Transformers for Document-Level Relation Extraction
Andrei C. Coman, Christos Theodoropoulos, Marie-Francine Moens, James Henderson
cs.CL
Document-level relation extraction aims to identify relationships between entities within a document. Current methods rely on text-based encoders and employ various hand-coded pooling heuristics to aggregate information from entity mentions and associated contexts. In this paper, we replace these rigid pooling function...
http://arxiv.org/pdf/2308.14423v1
2023-08-28T09:04:03
2023-08-28T09:04:03
{ "title": [ -0.007331848, -0.012878418, -0.049926758, -0.101867676, -0.00020694733, 0.015449524, -0.062316895, 0.018066406, 0.02796936, 0.020339966, -0.025436401, 0.017608643, 0.010040283, -0.00003117323, -0.044433594, 0.00016474724, -0.020202637, ...
http://arxiv.org/abs/2308.14391v1
FIRE: Food Image to REcipe generation
Prateek Chhikara, Dhiraj Chaurasia, Yifan Jiang, Omkar Masur, Filip Ilievski
cs.CV, cs.CL
Food computing has emerged as a prominent multidisciplinary field of research in recent years. An ambitious goal of food computing is to develop end-to-end intelligent systems capable of autonomously producing recipe information for a food image. Current image-to-recipe methods are retrieval-based and their success dep...
http://arxiv.org/pdf/2308.14391v1
2023-08-28T08:14:20
2023-08-28T08:14:20
{ "title": [ 0.0020484924, 0.008544922, -0.036865234, -0.03982544, 0.0129470825, -0.013076782, -0.011657715, -0.025375366, -0.007881165, 0.031921387, -0.0131073, 0.015129089, 0.0075950623, -0.019989014, -0.022201538, 0.005634308, 0.004886627, 0.0...
http://arxiv.org/abs/2308.14359v3
Effect of Attention and Self-Supervised Speech Embeddings on Non-Semantic Speech Tasks
Payal Mohapatra, Akash Pandey, Yueyuan Sui, Qi Zhu
cs.AI, cs.CL
Human emotion understanding is pivotal in making conversational technology mainstream. We view speech emotion understanding as a perception task which is a more realistic setting. With varying contexts (languages, demographics, etc.) different share of people perceive the same speech segment as a non-unanimous emotion....
http://arxiv.org/pdf/2308.14359v3
2023-09-27T18:00:54
2023-08-28T07:11:27
{ "title": [ 0.011703491, 0.015365601, -0.052459717, -0.061279297, 0.008636475, -0.036254883, -0.039733887, 0.015655518, -0.0103302, 0.04095459, -0.00806427, -0.002380371, 0.018981934, -0.024734497, -0.031799316, 0.0101623535, -0.012649536, 0.040...
http://arxiv.org/abs/2308.14353v1
ZhuJiu: A Multi-dimensional, Multi-faceted Chinese Benchmark for Large Language Models
Baoli Zhang, Haining Xie, Pengfan Du, Junhao Chen, Pengfei Cao, Yubo Chen, Shengping Liu, Kang Liu, Jun Zhao
cs.CL
The unprecedented performance of large language models (LLMs) requires comprehensive and accurate evaluation. We argue that for LLMs evaluation, benchmarks need to be comprehensive and systematic. To this end, we propose the ZhuJiu benchmark, which has the following strengths: (1) Multi-dimensional ability coverage: We...
http://arxiv.org/pdf/2308.14353v1
2023-08-28T06:56:44
2023-08-28T06:56:44
{ "title": [ -0.0541687, -0.014625549, -0.044006348, -0.05795288, 0.0021514893, 0.010437012, -0.055236816, 0.013679504, 0.007385254, 0.046020508, 0.000641346, 0.0067100525, 0.016433716, -0.022720337, -0.050567627, 0.01727295, -0.017196655, 0.0348...
http://arxiv.org/abs/2308.14352v1
EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models
Rongjie Yi, Liwei Guo, Shiyun Wei, Ao Zhou, Shangguang Wang, Mengwei Xu
cs.LG, cs.AI, cs.CL
Large Language Models (LLMs) such as GPTs and LLaMa have ushered in a revolution in machine intelligence, owing to their exceptional capabilities in a wide range of machine learning tasks. However, the transition of LLMs from data centers to edge devices presents a set of challenges and opportunities. While this shift ...
http://arxiv.org/pdf/2308.14352v1
2023-08-28T06:56:08
2023-08-28T06:56:08
{ "title": [ -0.05517578, -0.034606934, -0.032958984, -0.07122803, -0.0082473755, 0.015014648, -0.07421875, 0.03466797, -0.0020256042, 0.017059326, -0.0138168335, 0.007789612, 0.01953125, -0.027832031, -0.012481689, 0.009727478, -0.011375427, 0.0...
http://arxiv.org/abs/2308.14346v1
DISC-MedLLM: Bridging General Large Language Models and Real-World Medical Consultation
Zhijie Bao, Wei Chen, Shengze Xiao, Kuang Ren, Jiaao Wu, Cheng Zhong, Jiajie Peng, Xuanjing Huang, Zhongyu Wei
cs.CL, cs.AI
We propose DISC-MedLLM, a comprehensive solution that leverages Large Language Models (LLMs) to provide accurate and truthful medical response in end-to-end conversational healthcare services. To construct high-quality Supervised Fine-Tuning (SFT) datasets, we employ three strategies: utilizing medical knowledge-graphs...
http://arxiv.org/pdf/2308.14346v1
2023-08-28T06:41:49
2023-08-28T06:41:49
{ "title": [ -0.016433716, -0.02229309, -0.0340271, -0.09490967, -0.0029010773, 0.017852783, -0.08215332, -0.0035915375, 0.0072669983, 0.037322998, 0.000019609928, 0.0026836395, 0.01902771, -0.019424438, -0.03286743, 0.019989014, 0.003320694, 0.0...
http://arxiv.org/abs/2308.14337v1
Cognitive Effects in Large Language Models
Jonathan Shaki, Sarit Kraus, Michael Wooldridge
cs.AI, cs.CL
Large Language Models (LLMs) such as ChatGPT have received enormous attention over the past year and are now used by hundreds of millions of people every day. The rapid adoption of this technology naturally raises questions about the possible biases such models might exhibit. In this work, we tested one of these models...
http://arxiv.org/pdf/2308.14337v1
2023-08-28T06:30:33
2023-08-28T06:30:33
{ "title": [ -0.01864624, -0.009178162, -0.032440186, -0.054870605, 0.0014104843, -0.013824463, -0.039367676, -0.009155273, 0.005344391, 0.036376953, -0.014091492, -0.017074585, 0.013427734, -0.018447876, -0.04876709, 0.016342163, -0.008796692, 0...
http://arxiv.org/abs/2308.14321v1
Leveraging A Medical Knowledge Graph into Large Language Models for Diagnosis Prediction
Yanjun Gao, Ruizhe Li, John Caskey, Dmitriy Dligach, Timothy Miller, Matthew M. Churpek, Majid Afshar
cs.CL, cs.AI
Electronic Health Records (EHRs) and routine documentation practices play a vital role in patients' daily care, providing a holistic record of health, diagnoses, and treatment. However, complex and verbose EHR narratives overload healthcare providers, risking diagnostic inaccuracies. While Large Language Models (LLMs) ...
http://arxiv.org/pdf/2308.14321v1
2023-08-28T06:05:18
2023-08-28T06:05:18
{ "title": [ -0.014404297, -0.010345459, -0.04550171, -0.09967041, 0.0002156496, 0.011833191, -0.07354736, 0.008087158, 0.00057697296, 0.029067993, -0.011001587, -0.0041007996, 0.0014781952, -0.01890564, -0.032348633, -0.0003888607, 0.008369446, ...
http://arxiv.org/abs/2308.14306v2
Evaluating the Robustness to Instructions of Large Language Models
Yuansheng Ni, Sichao Jiang, Xinyu wu, Hui Shen, Yuli Zhou
cs.CL, cs.AI
Recently, Instruction fine-tuning has risen to prominence as a potential method for enhancing the zero-shot capabilities of Large Language Models (LLMs) on novel tasks. This technique has shown an exceptional ability to boost the performance of moderately sized LLMs, sometimes even reaching performance levels comparabl...
http://arxiv.org/pdf/2308.14306v2
2023-08-29T20:10:50
2023-08-28T04:57:07
{ "title": [ -0.02911377, -0.005554199, -0.03567505, -0.06707764, 0.007835388, 0.0040397644, -0.08135986, -0.011802673, 0.017318726, 0.02999878, -0.01966858, -0.00096559525, 0.0058135986, -0.008338928, -0.055358887, 0.017715454, -0.0069389343, 0....
http://arxiv.org/abs/2308.14280v2
FonMTL: Towards Multitask Learning for the Fon Language
Bonaventure F. P. Dossou, Iffanice Houndayi, Pamely Zantou, Gilles Hacheme
cs.CL, cs.AI
The Fon language, spoken by an average 2 million of people, is a truly low-resourced African language, with a limited online presence, and existing datasets (just to name but a few). Multitask learning is a learning paradigm that aims to improve the generalization capacity of a model by sharing knowledge across differe...
http://arxiv.org/pdf/2308.14280v2
2023-09-11T22:51:25
2023-08-28T03:26:21
{ "title": [ -0.028396606, -0.0071525574, -0.052978516, -0.09246826, 0.016540527, 0.0079422, -0.072387695, 0.023422241, -0.022460938, 0.020126343, -0.027069092, 0.016860962, -0.005897522, -0.022583008, -0.010848999, -0.013999939, -0.023132324, 0....
http://arxiv.org/abs/2308.14272v1
Goodhart's Law Applies to NLP's Explanation Benchmarks
Jennifer Hsia, Danish Pruthi, Aarti Singh, Zachary C. Lipton
cs.CL, cs.LG
Despite the rising popularity of saliency-based explanations, the research community remains at an impasse, facing doubts concerning their purpose, efficacy, and tendency to contradict each other. Seeking to unite the community's efforts around common goals, several recent works have proposed evaluation metrics. In thi...
http://arxiv.org/pdf/2308.14272v1
2023-08-28T03:03:03
2023-08-28T03:03:03
{ "title": [ 0.02986145, -0.012298584, 0.009963989, -0.03741455, -0.045562744, -0.014549255, -0.036499023, -0.025680542, -0.013206482, 0.033813477, -0.03149414, 0.019515991, 0.006816864, -0.020614624, 0.008361816, -0.010597229, -0.0002065897, 0.0...
http://arxiv.org/abs/2308.14266v1
SalesBot 2.0: A Human-Like Intent-Guided Chit-Chat Dataset
Wen-Yu Chang, Yun-Nung Chen
cs.CL, cs.AI
In recent research on dialogue systems and corpora, there has been a significant focus on two distinct categories: task-oriented (TOD) and open-domain (chit-chat) dialogues. TOD systems aim to satisfy specific user goals, such as finding a movie to watch, whereas open-domain systems primarily focus on generating engagi...
http://arxiv.org/pdf/2308.14266v1
2023-08-28T02:48:49
2023-08-28T02:48:49
{ "title": [ 0.0049591064, -0.010505676, -0.022277832, -0.0501709, 0.0010957718, -0.0061035156, -0.017059326, 0.0011873245, -0.010559082, 0.05645752, -0.031585693, 0.03050232, 0.023086548, 0.007358551, -0.038116455, 0.010414124, -0.014465332, 0.0...
http://arxiv.org/abs/2308.14242v1
The Cultural Psychology of Large Language Models: Is ChatGPT a Holistic or Analytic Thinker?
Chuanyang Jin, Songyang Zhang, Tianmin Shu, Zhihan Cui
cs.AI, cs.CL
The prevalent use of Large Language Models (LLMs) has necessitated studying their mental models, yielding noteworthy theoretical and practical implications. Current research has demonstrated that state-of-the-art LLMs, such as ChatGPT, exhibit certain theory of mind capabilities and possess relatively stable Big Five a...
http://arxiv.org/pdf/2308.14242v1
2023-08-28T01:05:18
2023-08-28T01:05:18
{ "title": [ -0.015129089, -0.02281189, 0.0057754517, 0.012588501, -0.04397583, -0.0127334595, -0.008506775, -0.02029419, -0.027786255, 0.0016403198, -0.000015437603, 0.007820129, 0.003604889, -0.04107666, 0.022781372, -0.0065994263, -0.004814148, ...
http://arxiv.org/abs/2308.14217v1
Generations of Knowledge Graphs: The Crazy Ideas and the Business Impact
Xin Luna Dong
cs.DB, cs.AI, cs.CL
Knowledge Graphs (KGs) have been used to support a wide range of applications, from web search to personal assistant. In this paper, we describe three generations of knowledge graphs: entity-based KGs, which have been supporting general search and question answering (e.g., at Google and Bing); text-rich KGs, which have...
http://arxiv.org/pdf/2308.14217v1
2023-08-27T22:35:27
2023-08-27T22:35:27
{ "title": [ -0.010643005, -0.011413574, -0.07098389, -0.067993164, 0.0073013306, -0.024597168, -0.023986816, 0.012771606, -0.0032749176, -0.00242424, -0.04058838, -0.0020961761, -0.040039062, -0.028533936, -0.029418945, -0.01625061, 0.02204895, ...
http://arxiv.org/abs/2308.14199v1
Symbolic and Language Agnostic Large Language Models
Walid S. Saba
cs.AI, cs.CL
We argue that the relative success of large language models (LLMs) is not a reflection on the symbolic vs. subsymbolic debate but a reflection on employing an appropriate strategy of bottom-up reverse engineering of language at scale. However, due to the subsymbolic nature of these models whatever knowledge these syste...
http://arxiv.org/pdf/2308.14199v1
2023-08-27T20:24:33
2023-08-27T20:24:33
{ "title": [ -0.046783447, -0.009857178, -0.05355835, -0.07122803, 0.0009217262, 0.007881165, -0.07733154, 0.01776123, -0.0006914139, 0.043182373, -0.0043029785, -0.019363403, 0.005683899, -0.014045715, -0.046417236, 0.013954163, -0.0071525574, 0...
http://arxiv.org/abs/2308.14186v1
Empowering Cross-lingual Abilities of Instruction-tuned Large Language Models by Translation-following demonstrations
Leonardo Ranaldi, Giulia Pucci, Andre Freitas
cs.CL, cs.AI
The language ability of Large Language Models (LLMs) is often unbalanced towards English because of the imbalance in the distribution of the pre-training data. This disparity is demanded in further fine-tuning and affecting the cross-lingual abilities of LLMs. In this paper, we propose to empower Instructiontuned LLMs ...
http://arxiv.org/pdf/2308.14186v1
2023-08-27T19:22:12
2023-08-27T19:22:12
{ "title": [ -0.027618408, -0.0146102905, -0.018447876, -0.06744385, -0.019119263, 0.011833191, -0.07312012, -0.0073394775, -0.00041389465, 0.032806396, -0.016906738, 0.012252808, 0.02279663, -0.012535095, -0.028930664, 0.013374329, -0.0010004044, ...
http://arxiv.org/abs/2308.14182v1
Generative AI for Business Strategy: Using Foundation Models to Create Business Strategy Tools
Son The Nguyen, Theja Tulabandhula
cs.CL
Generative models (foundation models) such as LLMs (large language models) are having a large impact on multiple fields. In this work, we propose the use of such models for business decision making. In particular, we combine unstructured textual data sources (e.g., news data) with multiple foundation models (namely, GP...
http://arxiv.org/pdf/2308.14182v1
2023-08-27T19:03:12
2023-08-27T19:03:12
{ "title": [ -0.027008057, -0.001958847, -0.056488037, -0.02166748, -0.011138916, -0.017028809, -0.020843506, -0.013343811, -0.01071167, 0.02658081, -0.05355835, 0.018478394, -0.006801605, -0.006965637, -0.02130127, -0.027832031, 0.0058517456, 0....
http://arxiv.org/abs/2308.14179v1
Towards Vision-Language Mechanistic Interpretability: A Causal Tracing Tool for BLIP
Vedant Palit, Rohan Pandey, Aryaman Arora, Paul Pu Liang
cs.CL, cs.AI, cs.CV
Mechanistic interpretability seeks to understand the neural mechanisms that enable specific behaviors in Large Language Models (LLMs) by leveraging causality-based methods. While these approaches have identified neural circuits that copy spans of text, capture factual knowledge, and more, they remain unusable for multi...
http://arxiv.org/pdf/2308.14179v1
2023-08-27T18:46:47
2023-08-27T18:46:47
{ "title": [ 0.0033721924, -0.021652222, -0.017288208, -0.06060791, -0.03250122, -0.009979248, -0.06958008, -0.02053833, 0.016311646, -0.008140564, -0.012466431, -0.0053710938, -0.0006365776, -0.0395813, -0.0030670166, -0.0137786865, 0.0110321045, ...
http://arxiv.org/abs/2308.14149v1
Examining User-Friendly and Open-Sourced Large GPT Models: A Survey on Language, Multimodal, and Scientific GPT Models
Kaiyuan Gao, Sunan He, Zhenyu He, Jiacheng Lin, QiZhi Pei, Jie Shao, Wei Zhang
cs.CL, cs.AI
Generative pre-trained transformer (GPT) models have revolutionized the field of natural language processing (NLP) with remarkable performance in various tasks and also extend their power to multimodal domains. Despite their success, large GPT models like GPT-4 face inherent limitations such as considerable size, high ...
http://arxiv.org/pdf/2308.14149v1
2023-08-27T16:14:19
2023-08-27T16:14:19
{ "title": [ -0.045318604, 0.008140564, -0.03010559, -0.037597656, -0.022460938, 0.010269165, -0.076293945, -0.035705566, -0.01828003, 0.01096344, 0.0015764236, -0.023345947, -0.0037078857, -0.043121338, -0.020523071, 0.0016374588, -0.017913818, ...
http://arxiv.org/abs/2308.14132v3
Detecting Language Model Attacks with Perplexity
Gabriel Alon, Michael Kamfonas
cs.CL, cs.AI, cs.CR, cs.LG
A novel hack involving Large Language Models (LLMs) has emerged, exploiting adversarial suffixes to deceive models into generating perilous responses. Such jailbreaks can trick LLMs into providing intricate instructions to a malicious user for creating explosives, orchestrating a bank heist, or facilitating the creatio...
http://arxiv.org/pdf/2308.14132v3
2023-11-07T03:30:15
2023-08-27T15:20:06
{ "title": [ -0.01838684, -0.010421753, -0.017807007, -0.044433594, -0.013206482, -0.009933472, -0.049957275, -0.007484436, -0.025680542, 0.024383545, 0.004421234, 0.021377563, 0.00081443787, -0.026123047, -0.026504517, -0.001876831, -0.021652222, ...
http://arxiv.org/abs/2308.14120v3
Large Language Models Streamline Automated Machine Learning for Clinical Studies
Soroosh Tayebi Arasteh, Tianyu Han, Mahshad Lotfinia, Christiane Kuhl, Jakob Nikolas Kather, Daniel Truhn, Sven Nebelung
cs.LG, cs.AI, cs.CL
A knowledge gap persists between machine learning (ML) developers (e.g., data scientists) and practitioners (e.g., clinicians), hampering the full utilization of ML for clinical data analysis. We investigated the potential of the ChatGPT Advanced Data Analysis (ADA), an extension of GPT-4, to bridge this gap and perfor...
http://arxiv.org/pdf/2308.14120v3
2023-10-09T18:01:12
2023-08-27T14:28:38
{ "title": [ -0.032287598, -0.013290405, -0.04296875, -0.06506348, -0.02017212, -0.0034275055, -0.07385254, -0.002210617, -0.019226074, 0.035186768, -0.013221741, -0.0073432922, 0.006641388, -0.034576416, -0.018249512, 0.00233078, 0.015312195, 0....
http://arxiv.org/abs/2308.14115v1
Situated Natural Language Explanations
Zining Zhu, Haoming Jiang, Jingfeng Yang, Sreyashi Nag, Chao Zhang, Jie Huang, Yifan Gao, Frank Rudzicz, Bing Yin
cs.CL
Natural language is among the most accessible tools for explaining decisions to humans, and large pretrained language models (PLMs) have demonstrated impressive abilities to generate coherent natural language explanations (NLE). The existing NLE research perspectives do not take the audience into account. An NLE can ha...
http://arxiv.org/pdf/2308.14115v1
2023-08-27T14:14:28
2023-08-27T14:14:28
{ "title": [ -0.0040283203, -0.024124146, -0.00579834, -0.09307861, -0.019973755, 0.011199951, -0.04348755, 0.0009622574, -0.009933472, 0.012779236, -0.026000977, -0.005302429, -0.026245117, -0.020843506, -0.012718201, -0.011543274, -0.0072669983, ...
http://arxiv.org/abs/2308.14089v1
MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records
Scott L. Fleming, Alejandro Lozano, William J. Haberkorn, Jenelle A. Jindal, Eduardo P. Reis, Rahul Thapa, Louis Blankemeier, Julian Z. Genkins, Ethan Steinberg, Ashwin Nayak, Birju S. Patel, Chia-Chun Chiang, Alison Callahan, Zepeng Huo, Sergios Gatidis, Scott J. Adams, Oluseyi Fayanju, Shreya J. Shah, Thomas Savage, ...
cs.CL, cs.AI, cs.LG
The ability of large language models (LLMs) to follow natural language instructions with human-level fluency suggests many opportunities in healthcare to reduce administrative burden and improve quality of care. However, evaluating LLMs on realistic text generation tasks for healthcare remains challenging. Existing que...
http://arxiv.org/pdf/2308.14089v1
2023-08-27T12:24:39
2023-08-27T12:24:39
{ "title": [ 0.0057640076, -0.015014648, -0.030319214, -0.056365967, -0.006580353, -0.017730713, -0.024261475, -0.009147644, -0.012237549, 0.036071777, -0.017425537, 0.016311646, 0.02128601, -0.03237915, -0.010765076, 0.0012578964, 0.005794525, 0...
http://arxiv.org/abs/2308.14077v1
An Analysis of On-the-fly Determinization of Finite-state Automata
Ivan Baburin, Ryan Cotterell
cs.FL, cs.CL
In this paper we establish an abstraction of on-the-fly determinization of finite-state automata using transition monoids and demonstrate how it can be applied to bound the asymptotics. We present algebraic and combinatorial properties that are sufficient for a polynomial state complexity of the deterministic automaton...
http://arxiv.org/pdf/2308.14077v1
2023-08-27T11:51:27
2023-08-27T11:51:27
{ "title": [ -0.03338623, -0.00052928925, -0.015777588, -0.06896973, 0.009422302, 0.013648987, -0.03338623, 0.007484436, -0.005012512, 0.022994995, -0.016647339, 0.01876831, -0.021438599, -0.0043754578, -0.026351929, 0.001745224, -0.003540039, 0....
http://arxiv.org/abs/2308.14034v1
Confucius: Iterative Tool Learning from Introspection Feedback by Easy-to-Difficult Curriculum
Shen Gao, Zhengliang Shi, Minghang Zhu, Bowen Fang, Xin Xin, Pengjie Ren, Zhumin Chen, Jun Ma
cs.AI, cs.CL
Augmenting large language models (LLMs) with external tools has emerged as a promising approach to extending the capability of LLMs. Although some works employ open-source LLMs for the tool learning task, most of them are trained in a controlled environment in which LLMs only learn to execute the human-provided tools. ...
http://arxiv.org/pdf/2308.14034v1
2023-08-27T07:53:00
2023-08-27T07:53:00
{ "title": [ -0.036254883, -0.0038394928, -0.03616333, 0.01902771, -0.0289917, -0.023117065, -0.010246277, -0.043273926, 0.00274086, 0.011749268, -0.029205322, -0.0054512024, 0.0059776306, -0.036193848, 0.002916336, -0.020812988, -0.008674622, 0....
http://arxiv.org/abs/2308.14763v1
VoiceBank-2023: A Multi-Speaker Mandarin Speech Corpus for Constructing Personalized TTS Systems for the Speech Impaired
Jia-Jyu Su, Pang-Chen Liao, Yen-Ting Lin, Wu-Hao Li, Guan-Ting Liou, Cheng-Che Kao, Wei-Cheng Chen, Jen-Chieh Chiang, Wen-Yang Chang, Pin-Han Lin, Chen-Yu Chiang
eess.AS, cs.CL, cs.SD
Services of personalized TTS systems for the Mandarin-speaking speech impaired are rarely mentioned. Taiwan started the VoiceBanking project in 2020, aiming to build a complete set of services to deliver personalized Mandarin TTS systems to amyotrophic lateral sclerosis patients. This paper reports the corpus design, c...
http://arxiv.org/pdf/2308.14763v1
2023-08-27T07:35:30
2023-08-27T07:35:30
{ "title": [ 0.026794434, 0.012641907, -0.04156494, -0.088256836, 0.012565613, 0.0005993843, -0.04385376, 0.01424408, 0.005722046, 0.057861328, -0.01776123, 0.0121154785, 0.018203735, -0.025634766, -0.044189453, 0.0087890625, -0.028625488, 0.0485...
http://arxiv.org/abs/2308.13961v1
Translate Meanings, Not Just Words: IdiomKB's Role in Optimizing Idiomatic Translation with Language Models
Shuang Li, Jiangjie Chen, Siyu Yuan, Xinyi Wu, Hao Yang, Shimin Tao, Yanghua Xiao
cs.CL
To translate well, machine translation (MT) systems and general-purposed language models (LMs) need a deep understanding of both source and target languages and cultures. Therefore, idioms, with their non-compositional nature, pose particular challenges for Transformer-based systems, as literal translations often miss ...
http://arxiv.org/pdf/2308.13961v1
2023-08-26T21:38:31
2023-08-26T21:38:31
{ "title": [ -0.061676025, -0.0061149597, 0.009552002, -0.0015249252, -0.05053711, 0.029891968, -0.06463623, -0.045715332, 0.00026130676, 0.051574707, -0.01109314, 0.0236969, 0.046875, -0.016494751, -0.028442383, 0.027435303, -0.0029659271, 0.013...
http://arxiv.org/abs/2308.13958v1
Improving Knowledge Distillation for BERT Models: Loss Functions, Mapping Methods, and Weight Tuning
Apoorv Dankar, Adeem Jassani, Kartikaeya Kumar
cs.CL, cs.AI
The use of large transformer-based models such as BERT, GPT, and T5 has led to significant advancements in natural language processing. However, these models are computationally expensive, necessitating model compression techniques that reduce their size and complexity while maintaining accuracy. This project investiga...
http://arxiv.org/pdf/2308.13958v1
2023-08-26T20:59:21
2023-08-26T20:59:21
{ "title": [ -0.024291992, -0.036743164, -0.021392822, -0.066223145, -0.042419434, 0.018051147, -0.09515381, 0.0006246567, 0.029449463, 0.023269653, -0.029190063, -0.015945435, 0.016738892, -0.03201294, -0.038208008, 0.0039863586, 0.0064926147, -...
http://arxiv.org/abs/2308.13916v3
Exploring Large Language Models for Knowledge Graph Completion
Liang Yao, Jiazhen Peng, Chengsheng Mao, Yuan Luo
cs.CL, cs.AI
Knowledge graphs play a vital role in numerous artificial intelligence tasks, yet they frequently face the issue of incompleteness. In this study, we explore utilizing Large Language Models (LLM) for knowledge graph completion. We consider triples in knowledge graphs as text sequences and introduce an innovative framew...
http://arxiv.org/pdf/2308.13916v3
2023-09-10T17:42:37
2023-08-26T16:51:17
{ "title": [ -0.03692627, -0.017181396, -0.062927246, -0.10021973, -0.003168106, 0.019317627, -0.087524414, 0.03643799, 0.003873825, 0.03111267, -0.028457642, 0.011512756, 0.0074806213, 0.0012607574, -0.039123535, 0.0045547485, -0.010070801, 0.05...
http://arxiv.org/abs/2308.13911v1
A Wide Evaluation of ChatGPT on Affective Computing Tasks
Mostafa M. Amin, Rui Mao, Erik Cambria, Björn W. Schuller
cs.AI, cs.CL
With the rise of foundation models, a new artificial intelligence paradigm has emerged, by simply using general purpose foundation models with prompting to solve problems instead of training a separate machine learning model for each problem. Such models have been shown to have emergent properties of solving problems t...
http://arxiv.org/pdf/2308.13911v1
2023-08-26T16:10:30
2023-08-26T16:10:30
{ "title": [ -0.0037269592, -0.009742737, -0.012641907, -0.0044937134, -0.0035266876, -0.011932373, 0.017120361, -0.021820068, 0.02130127, 0.023956299, -0.025787354, 0.009269714, 0.007041931, -0.028747559, -0.029708862, 0.02168274, -0.018753052, ...
http://arxiv.org/abs/2308.13904v2
LMSanitator: Defending Prompt-Tuning Against Task-Agnostic Backdoors
Chengkun Wei, Wenlong Meng, Zhikun Zhang, Min Chen, Minghu Zhao, Wenjing Fang, Lei Wang, Zihui Zhang, Wenzhi Chen
cs.CL, cs.CR, cs.LG
Prompt-tuning has emerged as an attractive paradigm for deploying large-scale language models due to its strong downstream task performance and efficient multitask serving ability. Despite its wide adoption, we empirically show that prompt-tuning is vulnerable to downstream task-agnostic backdoors, which reside in the ...
http://arxiv.org/pdf/2308.13904v2
2023-10-14T15:07:37
2023-08-26T15:21:47
{ "title": [ -0.019058228, 0.0018014908, -0.013702393, -0.050109863, -0.0058631897, 0.006401062, -0.045532227, -0.030700684, -0.061767578, 0.0019388199, 0.02645874, 0.020019531, -0.0446167, -0.04171753, -0.00058460236, -0.017486572, -0.00060272217, ...
http://arxiv.org/abs/2308.13844v1
Solving Math Word Problem with Problem Type Classification
Jie Yao, Zihao Zhou, Qiufeng Wang
cs.CL
Math word problems (MWPs) require analyzing text descriptions and generating mathematical equations to derive solutions. Existing works focus on solving MWPs with two types of solvers: tree-based solver and large language model (LLM) solver. However, these approaches always solve MWPs by a single solver, which will bri...
http://arxiv.org/pdf/2308.13844v1
2023-08-26T10:35:16
2023-08-26T10:35:16
{ "title": [ -0.0011024475, 0.00037837029, -0.028411865, -0.015838623, -0.009407043, -0.03930664, 0.013999939, -0.009994507, -0.028167725, 0.012084961, -0.026260376, -0.0038967133, -0.037261963, -0.030563354, 0.002948761, -0.034698486, -0.0016498566,...
http://arxiv.org/abs/2308.13782v1
Planning with Logical Graph-based Language Model for Instruction Generation
Fan Zhang, Kebing Jin, Hankz Hankui Zhuo
cs.CL, cs.AI
Despite the superior performance of large language models to generate natural language texts, it is hard to generate texts with correct logic according to a given task, due to the difficulties for neural models to capture implied rules from free-form texts. In this paper, we propose a novel graph-based language model, ...
http://arxiv.org/pdf/2308.13782v1
2023-08-26T06:28:14
2023-08-26T06:28:14
{ "title": [ -0.060760498, -0.023620605, -0.040649414, -0.039916992, -0.03479004, 0.019195557, -0.072265625, 0.007118225, -0.04031372, 0.02053833, -0.028121948, 0.02909851, -0.046875, -0.010017395, -0.009490967, -0.028625488, -0.015609741, 0.0392...
http://arxiv.org/abs/2308.13775v2
EditSum: A Retrieve-and-Edit Framework for Source Code Summarization
Jia Li, Yongmin Li, Ge Li, Xing Hu, Xin Xia, Zhi Jin
cs.SE, cs.CL
Existing studies show that code summaries help developers understand and maintain source code. Unfortunately, these summaries are often missing or outdated in software projects. Code summarization aims to generate natural language descriptions automatically for source code. Code summaries are highly structured and have...
http://arxiv.org/pdf/2308.13775v2
2023-09-07T11:19:30
2023-08-26T05:48:57
{ "title": [ -0.019943237, -0.025787354, -0.046813965, -0.09814453, -0.0010652542, 0.014198303, -0.061065674, 0.020217896, 0.012863159, 0.0385437, -0.037322998, 0.01260376, 0.022125244, -0.0061416626, -0.04083252, 0.00041484833, -0.0009098053, 0....
http://arxiv.org/abs/2308.13768v1
Adversarial Fine-Tuning of Language Models: An Iterative Optimisation Approach for the Generation and Detection of Problematic Content
Charles O'Neill, Jack Miller, Ioana Ciuca, Yuan-Sen Ting, Thang Bui
cs.CL, cs.LG
In this paper, we tackle the emerging challenge of unintended harmful content generation in Large Language Models (LLMs) with a novel dual-stage optimisation technique using adversarial fine-tuning. Our two-pronged approach employs an adversarial model, fine-tuned to generate potentially harmful prompts, and a judge mo...
http://arxiv.org/pdf/2308.13768v1
2023-08-26T05:20:58
2023-08-26T05:20:58
{ "title": [ -0.005317688, 0.0055160522, -0.017166138, -0.05407715, 0.0050315857, 0.008514404, -0.054656982, -0.026519775, -0.013336182, 0.03439331, 0.0032749176, 0.011390686, -0.010894775, -0.019241333, -0.04699707, -0.0018358231, -0.029281616, ...
http://arxiv.org/abs/2308.13760v1
How Can Context Help? Exploring Joint Retrieval of Passage and Personalized Context
Hui Wan, Hongkang Li, Songtao Lu, Xiaodong Cui, Marina Danilevsky
cs.AI, cs.CL, cs.IR
The integration of external personalized context information into document-grounded conversational systems has significant potential business value, but has not been well-studied. Motivated by the concept of personalized context-aware document-grounded conversational systems, we introduce the task of context-aware pass...
http://arxiv.org/pdf/2308.13760v1
2023-08-26T04:49:46
2023-08-26T04:49:46
{ "title": [ 0.020599365, 0.013809204, -0.035369873, -0.08288574, 0.029067993, -0.00831604, -0.016616821, -0.010177612, -0.033355713, 0.06585693, -0.009208679, 0.042785645, 0.017059326, -0.010757446, -0.023605347, 0.020065308, 0.0051612854, 0.028...
http://arxiv.org/abs/2308.13754v2
ZC3: Zero-Shot Cross-Language Code Clone Detection
Jia Li, Chongyang Tao, Zhi Jin, Fang Liu, Jia Li, Ge Li
cs.SE, cs.CL, cs.IR
Developers introduce code clones to improve programming productivity. Many existing studies have achieved impressive performance in monolingual code clone detection. However, during software development, more and more developers write semantically equivalent programs with different languages to support different platfo...
http://arxiv.org/pdf/2308.13754v2
2023-09-07T11:22:59
2023-08-26T03:48:10
{ "title": [ -0.010025024, -0.010299683, -0.03164673, -0.061309814, 0.0043678284, -0.0072402954, -0.03878784, -0.0051345825, -0.004421234, 0.039886475, -0.014533997, 0.021987915, 0.0055389404, -0.0206604, -0.01928711, 0.015731812, -0.0045547485, ...
http://arxiv.org/abs/2308.13738v1
On Philomatics and Psychomatics for Combining Philosophy and Psychology with Mathematics
Benyamin Ghojogh, Morteza Babaie
math.HO, cs.AI, cs.CL
We propose the concepts of philomatics and psychomatics as hybrid combinations of philosophy and psychology with mathematics. We explain four motivations for this combination which are fulfilling the desire of analytical philosophy, proposing science of philosophy, justifying mathematical algorithms by philosophy, and ...
http://arxiv.org/pdf/2308.13738v1
2023-08-26T02:52:42
2023-08-26T02:52:42
{ "title": [ 0.03778076, 0.002576828, -0.023010254, 0.016082764, -0.00944519, -0.048583984, 0.061035156, -0.017990112, -0.022842407, 0.029144287, -0.020874023, 0.0069389343, 0.013946533, -0.034454346, 0.01638794, -0.015991211, 0.008666992, 0.0321...
http://arxiv.org/abs/2308.13715v1
A Computational Evaluation Framework for Singable Lyric Translation
Haven Kim, Kento Watanabe, Masataka Goto, Juhan Nam
cs.CL
Lyric translation plays a pivotal role in amplifying the global resonance of music, bridging cultural divides, and fostering universal connections. Translating lyrics, unlike conventional translation tasks, requires a delicate balance between singability and semantics. In this paper, we present a computational framewor...
http://arxiv.org/pdf/2308.13715v1
2023-08-26T00:27:08
2023-08-26T00:27:08
{ "title": [ -0.04083252, -0.01663208, -0.0079956055, -0.057281494, -0.015129089, 0.021575928, -0.049072266, 0.018188477, 0.00047898293, 0.038726807, -0.035705566, 0.004234314, -0.0041122437, -0.0035934448, -0.061920166, -0.011489868, -0.02078247, ...
http://arxiv.org/abs/2308.13710v1
WellXplain: Wellness Concept Extraction and Classification in Reddit Posts for Mental Health Analysis
Muskan Garg
cs.CL, cs.AI
During the current mental health crisis, the importance of identifying potential indicators of mental issues from social media content has surged. Overlooking the multifaceted nature of mental and social well-being can have detrimental effects on one's mental state. In traditional therapy sessions, professionals manual...
http://arxiv.org/pdf/2308.13710v1
2023-08-25T23:50:05
2023-08-25T23:50:05
{ "title": [ 0.028793335, 0.000055253506, -0.039794922, -0.041992188, 0.023895264, -0.03375244, 0.0070495605, -0.02420044, -0.023925781, 0.033416748, -0.008583069, 0.04296875, 0.002117157, -0.015075684, -0.020584106, 0.011100769, 0.0062179565, 0....
http://arxiv.org/abs/2308.13696v1
On the Depth between Beam Search and Exhaustive Search for Text Generation
Yuu Jinnai, Tetsuro Morimura, Ukyo Honda
cs.CL, cs.AI
Beam search and exhaustive search are two extreme ends of text decoding algorithms with respect to the search depth. Beam search is limited in both search width and depth, whereas exhaustive search is a global search that has no such limitations. Surprisingly, beam search is not only computationally cheaper but also pe...
http://arxiv.org/pdf/2308.13696v1
2023-08-25T22:57:53
2023-08-25T22:57:53
{ "title": [ -0.02734375, -0.014434814, -0.005783081, -0.02407837, -0.023010254, 0.013710022, -0.042144775, -0.03338623, 0.022140503, 0.050048828, -0.008216858, 0.0020256042, -0.0010976791, -0.0057525635, -0.07940674, 0.0070762634, -0.027328491, ...
http://arxiv.org/abs/2308.13687v1
1.5 million materials narratives generated by chatbots
Yang Jeong Park, Sung Eun Jerng, Jin-Sung Park, Choah Kwon, Chia-Wei Hsu, Zhichu Ren, Sungroh Yoon, Ju Li
cond-mat.mtrl-sci, cs.CL
The advent of artificial intelligence (AI) has enabled a comprehensive exploration of materials for various applications. However, AI models often prioritize frequently encountered materials in the scientific literature, limiting the selection of suitable candidates based on inherent physical and chemical properties. T...
http://arxiv.org/pdf/2308.13687v1
2023-08-25T22:00:53
2023-08-25T22:00:53
{ "title": [ 0.014930725, -0.020065308, -0.005519867, -0.059570312, -0.053009033, 0.025146484, -0.07055664, 0.007858276, -0.036621094, 0.027557373, -0.0004940033, 0.03237915, -0.003490448, -0.03060913, 0.048217773, -0.02609253, -0.010467529, 0.04...
http://arxiv.org/abs/2308.13676v1
Rethinking Language Models as Symbolic Knowledge Graphs
Vishwas Mruthyunjaya, Pouya Pezeshkpour, Estevam Hruschka, Nikita Bhutani
cs.CL, cs.AI, cs.LG
Symbolic knowledge graphs (KGs) play a pivotal role in knowledge-centric applications such as search, question answering and recommendation. As contemporary language models (LMs) trained on extensive textual data have gained prominence, researchers have extensively explored whether the parametric knowledge within these...
http://arxiv.org/pdf/2308.13676v1
2023-08-25T21:25:08
2023-08-25T21:25:08
{ "title": [ -0.050567627, -0.029067993, -0.037200928, -0.07128906, -0.022583008, 0.018692017, -0.080322266, 0.015159607, -0.0039596558, 0.03994751, -0.013923645, -0.001996994, -0.015136719, 0.00024461746, -0.072509766, -0.004688263, -0.010017395, ...
http://arxiv.org/abs/2308.13646v1
GRASP: A Rehearsal Policy for Efficient Online Continual Learning
Md Yousuf Harun, Jhair Gallardo, Christopher Kanan
cs.LG, cs.CL, cs.CV
Continual learning (CL) in deep neural networks (DNNs) involves incrementally accumulating knowledge in a DNN from a growing data stream. A major challenge in CL is that non-stationary data streams cause catastrophic forgetting of previously learned abilities. Rehearsal is a popular and effective way to mitigate this p...
http://arxiv.org/pdf/2308.13646v1
2023-08-25T19:34:21
2023-08-25T19:34:21
{ "title": [ -0.0045204163, -0.01171875, -0.042907715, -0.09552002, 0.025802612, -0.013504028, -0.02180481, 0.013885498, 0.0020217896, 0.011077881, -0.017059326, 0.020553589, -0.011993408, -0.014877319, -0.023071289, -0.0039482117, 0.0070152283, ...
http://arxiv.org/abs/2308.13517v1
ChatGPT as Data Augmentation for Compositional Generalization: A Case Study in Open Intent Detection
Yihao Fang, Xianzhi Li, Stephen W. Thomas, Xiaodan Zhu
cs.CL, cs.AI
Open intent detection, a crucial aspect of natural language understanding, involves the identification of previously unseen intents in user-generated text. Despite the progress made in this field, challenges persist in handling new combinations of language components, which is essential for compositional generalization...
http://arxiv.org/pdf/2308.13517v1
2023-08-25T17:51:23
2023-08-25T17:51:23
{ "title": [ 0.025466919, 0.008934021, -0.01966858, -0.056396484, 0.0062713623, 0.0024166107, -0.032592773, -0.0126953125, -0.0062713623, 0.035003662, -0.025558472, 0.03729248, -0.017532349, -0.0058403015, -0.031677246, -0.0006875992, -0.02835083, ...
http://arxiv.org/abs/2308.13506v2
Training and Meta-Evaluating Machine Translation Evaluation Metrics at the Paragraph Level
Daniel Deutsch, Juraj Juraska, Mara Finkelstein, Markus Freitag
cs.CL
As research on machine translation moves to translating text beyond the sentence level, it remains unclear how effective automatic evaluation metrics are at scoring longer translations. In this work, we first propose a method for creating paragraph-level data for training and meta-evaluating metrics from existing sente...
http://arxiv.org/pdf/2308.13506v2
2023-08-28T17:46:59
2023-08-25T17:31:46
{ "title": [ -0.039154053, -0.008102417, -0.016983032, 0.020263672, -0.041625977, -0.0075531006, -0.0032463074, -0.021224976, -0.032928467, 0.06298828, -0.0017824173, 0.0063171387, 0.04916382, -0.02482605, -0.0010032654, 0.0049057007, -0.0013866425, ...
http://arxiv.org/abs/2308.13590v1
LSTM-based QoE Evaluation for Web Microservices' Reputation Scoring
Maha Driss
cs.IR, cs.CL
Sentiment analysis is the task of mining the authors' opinions about specific entities. It allows organizations to monitor different services in real time and act accordingly. Reputation is what is generally said or believed about people or things. Informally, reputation combines the measure of reliability derived from...
http://arxiv.org/pdf/2308.13590v1
2023-08-25T17:23:12
2023-08-25T17:23:12
{ "title": [ 0.022338867, 0.006767273, -0.029449463, -0.011253357, 0.030014038, -0.03955078, 0.036590576, -0.005897522, -0.011634827, 0.044677734, 0.0010471344, -0.033355713, 0.0075798035, -0.0317688, -0.05987549, -0.015655518, -0.024017334, 0.01...
http://arxiv.org/abs/2308.13497v1
Ngambay-French Neural Machine Translation (sba-Fr)
Sakayo Toadoum Sari, Angela Fan, Lema Logamou Seknewna
cs.CL, cs.LG
In Africa, and the world at large, there is an increasing focus on developing Neural Machine Translation (NMT) systems to overcome language barriers. NMT for Low-resource language is particularly compelling as it involves learning with limited labelled data. However, obtaining a well-aligned parallel corpus for low-res...
http://arxiv.org/pdf/2308.13497v1
2023-08-25T17:13:20
2023-08-25T17:13:20
{ "title": [ -0.029388428, -0.0003259182, -0.0231781, 0.0030498505, -0.036224365, 0.017700195, -0.040039062, -0.01210022, 0.02267456, 0.057922363, -0.031188965, -0.035949707, 0.09008789, -0.03591919, -0.019317627, 0.0025100708, 0.008255005, 0.013...
http://arxiv.org/abs/2308.13479v1
Prompting a Large Language Model to Generate Diverse Motivational Messages: A Comparison with Human-Written Messages
Samuel Rhys Cox, Ashraf Abdul, Wei Tsang Ooi
cs.CL, cs.HC
Large language models (LLMs) are increasingly capable and prevalent, and can be used to produce creative content. The quality of content is influenced by the prompt used, with more specific prompts that incorporate examples generally producing better results. On from this, it could be seen that using instructions writt...
http://arxiv.org/pdf/2308.13479v1
2023-08-25T16:35:06
2023-08-25T16:35:06
{ "title": [ -0.01763916, -0.00466156, -0.024917603, -0.031311035, -0.005306244, -0.0057868958, -0.058380127, -0.024383545, -0.013648987, 0.028045654, 0.00034213066, 0.0011472702, -0.0015735626, -0.025924683, -0.028259277, -0.0032901764, -0.012863159...
http://arxiv.org/abs/2308.13467v1
Leveraging Knowledge and Reinforcement Learning for Enhanced Reliability of Language Models
Nancy Tyagi, Surjodeep Sarkar, Manas Gaur
cs.CL, cs.AI, cs.IR
The Natural Language Processing(NLP) community has been using crowd sourcing techniques to create benchmark datasets such as General Language Understanding and Evaluation(GLUE) for training modern Language Models such as BERT. GLUE tasks measure the reliability scores using inter annotator metrics i.e. Cohens Kappa. Ho...
http://arxiv.org/pdf/2308.13467v1
2023-08-25T16:11:08
2023-08-25T16:11:08
{ "title": [ -0.008590698, -0.012084961, -0.036895752, -0.08239746, 0.0024280548, -0.004962921, -0.063964844, 0.015197754, -0.011077881, 0.025253296, -0.018218994, 0.012001038, -0.009010315, -0.0103302, -0.038330078, -0.0016899109, -0.011230469, ...
http://arxiv.org/abs/2308.13458v1
ARTIST: ARTificial Intelligence for Simplified Text
Lorenzo Corti, Jie Yang
cs.CL
Complex text is a major barrier for many citizens when accessing public information and knowledge. While often done manually, Text Simplification is a key Natural Language Processing task that aims for reducing the linguistic complexity of a text while preserving the original meaning. Recent advances in Generative Arti...
http://arxiv.org/pdf/2308.13458v1
2023-08-25T16:06:06
2023-08-25T16:06:06
{ "title": [ -0.015472412, -0.023590088, -0.030960083, -0.007926941, -0.06640625, -0.010238647, -0.010604858, 0.010475159, 0.013954163, 0.076049805, -0.037384033, -0.0049438477, 0.06036377, -0.02168274, -0.038208008, -0.015304565, 0.016326904, -0...
http://arxiv.org/abs/2308.13449v1
The Poison of Alignment
Aibek Bekbayev, Sungbae Chun, Yerzat Dulat, James Yamazaki
cs.CL
From the perspective of content safety issues, alignment has shown to limit large language models' (LLMs) harmful content generation. This intentional method of reinforcing models to not respond to certain user inputs seem to be present in many modern open-source instruction tuning datasets such as OpenAssistant or Gua...
http://arxiv.org/pdf/2308.13449v1
2023-08-25T15:51:15
2023-08-25T15:51:15
{ "title": [ 0.042388916, -0.023071289, 0.0023765564, -0.020645142, -0.02279663, -0.013175964, 0.017364502, -0.011428833, -0.024215698, 0.06555176, -0.034088135, 0.009796143, 0.01020813, -0.0074386597, -0.007675171, -0.013755798, 0.0013284683, 0....
http://arxiv.org/abs/2309.12342v1
Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensions
Reem I. Masoud, Ziquan Liu, Martin Ferianc, Philip Treleaven, Miguel Rodrigues
cs.CY, cs.CL, cs.LG
The deployment of large language models (LLMs) raises concerns regarding their cultural misalignment and potential ramifications on individuals from various cultural norms. Existing work investigated political and social biases and public opinions rather than their cultural values. To address this limitation, the propo...
http://arxiv.org/pdf/2309.12342v1
2023-08-25T14:50:13
2023-08-25T14:50:13
{ "title": [ -0.037231445, -0.010322571, -0.04925537, -0.03793335, -0.003042221, -0.02619934, -0.033081055, 0.0042419434, -0.0041656494, 0.039764404, 0.0032653809, 0.0012903214, 0.021728516, -0.027633667, -0.026123047, 0.022338867, -0.012542725, ...
http://arxiv.org/abs/2308.13399v2
EntropyRank: Unsupervised Keyphrase Extraction via Side-Information Optimization for Language Model-based Text Compression
Alexander Tsvetkov, Alon Kipnis
cs.CL, cs.IT, cs.LG, math.IT
We propose an unsupervised method to extract keywords and keyphrases from texts based on a pre-trained language model (LM) and Shannon's information maximization. Specifically, our method extracts phrases having the highest conditional entropy under the LM. The resulting set of keyphrases turns out to solve a relevant ...
http://arxiv.org/pdf/2308.13399v2
2023-08-29T18:28:13
2023-08-25T14:23:40
{ "title": [ -0.07342529, -0.0009036064, -0.032958984, -0.02230835, -0.016845703, 0.020751953, -0.048034668, 0.0031604767, 0.002374649, 0.06414795, -0.0030937195, 0.01676941, 0.016677856, -0.01234436, -0.061706543, 0.018081665, -0.009300232, 0.02...
http://arxiv.org/abs/2308.13387v2
Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs
Yuxia Wang, Haonan Li, Xudong Han, Preslav Nakov, Timothy Baldwin
cs.CL
With the rapid evolution of large language models (LLMs), new and hard-to-predict harmful capabilities are emerging. This requires developers to be able to identify risks through the evaluation of "dangerous capabilities" in order to responsibly deploy LLMs. In this work, we collect the first open-source dataset to eva...
http://arxiv.org/pdf/2308.13387v2
2023-09-04T01:47:30
2023-08-25T14:02:12
{ "title": [ -0.022445679, 0.009864807, -0.05480957, -0.03274536, -0.005683899, -0.022476196, -0.026260376, -0.016815186, -0.028244019, 0.003917694, -0.03604126, 0.041503906, -0.035491943, -0.022994995, -0.0003528595, -0.017837524, -0.02279663, 0...
http://arxiv.org/abs/2308.13383v1
Assessing Keyness using Permutation Tests
Thoralf Mildenberger
cs.CL, stat.AP
We propose a resampling-based approach for assessing keyness in corpus linguistics based on suggestions by Gries (2006, 2022). Traditional approaches based on hypothesis tests (e.g. Likelihood Ratio) model the copora as independent identically distributed samples of tokens. This model does not account for the often obs...
http://arxiv.org/pdf/2308.13383v1
2023-08-25T13:52:57
2023-08-25T13:52:57
{ "title": [ -0.0019426346, 0.018447876, -0.009407043, 0.0033340454, -0.008979797, -0.014953613, 0.021713257, -0.037902832, -0.023834229, 0.03527832, -0.030380249, -0.010032654, -0.043701172, -0.037322998, -0.036254883, -0.017913818, -0.014671326, ...
http://arxiv.org/abs/2308.13577v2
Text Style Transfer Evaluation Using Large Language Models
Phil Ostheimer, Mayank Nagda, Marius Kloft, Sophie Fellenz
cs.CL, cs.LG
Evaluating Text Style Transfer (TST) is a complex task due to its multifaceted nature. The quality of the generated text is measured based on challenging factors, such as style transfer accuracy, content preservation, and overall fluency. While human evaluation is considered to be the gold standard in TST assessment, i...
http://arxiv.org/pdf/2308.13577v2
2023-09-23T06:05:22
2023-08-25T13:07:33
{ "title": [ -0.037322998, 0.00062704086, -0.042907715, -0.022857666, -0.002243042, -0.0055999756, -0.03866577, -0.00818634, 0.0057296753, 0.048339844, -0.0010166168, 0.015625, 0.031143188, -0.013679504, -0.030853271, 0.024490356, -0.006790161, 0...
http://arxiv.org/abs/2308.13354v1
On the Impact of Language Selection for Training and Evaluating Programming Language Models
Jonathan Katzy, Maliheh Izadi, Arie van Deursen
cs.SE, cs.AI, cs.CL, cs.LG, cs.PL
The recent advancements in Transformer-based Language Models have demonstrated significant potential in enhancing the multilingual capabilities of these models. The remarkable progress made in this domain not only applies to natural language tasks but also extends to the domain of programming languages. Despite the abi...
http://arxiv.org/pdf/2308.13354v1
2023-08-25T12:57:59
2023-08-25T12:57:59
{ "title": [ -0.03111267, -0.02003479, -0.039855957, -0.028060913, -0.017028809, -0.02557373, -0.03086853, 0.009155273, -0.03164673, 0.021026611, -0.01386261, 0.007247925, -0.017715454, -0.027938843, -0.0149002075, -0.014572144, -0.014129639, 0.0...
http://arxiv.org/abs/2308.13576v1
An Ensemble Approach to Personalized Real Time Predictive Writing for Experts
Sourav Prosad, Viswa Datha Polavarapu, Shrutendra Harsola
cs.CL
Completing a sentence, phrase or word after typing few words / characters is very helpful for Intuit financial experts, while taking notes or having a live chat with users, since they need to write complex financial concepts more efficiently and accurately many times in a day. In this paper, we tie together different a...
http://arxiv.org/pdf/2308.13576v1
2023-08-25T12:45:46
2023-08-25T12:45:46
{ "title": [ -0.023971558, -0.0044555664, -0.034820557, -0.06604004, -0.010971069, 0.01184845, -0.04135132, 0.004886627, -0.008087158, 0.03286743, -0.012283325, 0.013511658, -0.019073486, -0.008293152, -0.051879883, -0.010536194, 0.005317688, 0.0...
http://arxiv.org/abs/2308.13345v1
Decoupled Structure for Improved Adaptability of End-to-End Models
Keqi Deng, Philip C. Woodland
eess.AS, cs.CL, cs.SD
Although end-to-end (E2E) trainable automatic speech recognition (ASR) has shown great success by jointly learning acoustic and linguistic information, it still suffers from the effect of domain shifts, thus limiting potential applications. The E2E ASR model implicitly learns an internal language model (LM) which chara...
http://arxiv.org/pdf/2308.13345v1
2023-08-25T12:31:12
2023-08-25T12:31:12
{ "title": [ 0.017364502, 0.0069351196, -0.012664795, -0.08648682, -0.019729614, 0.016296387, -0.07385254, -0.008872986, 0.005859375, 0.0068626404, 0.014556885, -0.035888672, 0.010063171, -0.07269287, 0.022415161, 0.012664795, 0.026855469, -0.026...
http://arxiv.org/abs/2308.13317v1
Transforming the Output of Generative Pre-trained Transformer: The Influence of the PGI Framework on Attention Dynamics
Aline Ioste
cs.AI, cs.CL, cs.LG
This paper presents a novel approach named Persona-Grouping-Intelligence (PGI), which has been crafted to tackle the challenges posed by GPT models when applied to real-world business issues. PGI leverages the inherent capabilities of the GPT model to comprehend intricate language structures and generate responses that...
http://arxiv.org/pdf/2308.13317v1
2023-08-25T11:41:05
2023-08-25T11:41:05
{ "title": [ 0.012512207, 0.031280518, -0.051239014, -0.0423584, 0.008934021, -0.02911377, -0.011016846, -0.023666382, -0.004371643, 0.035705566, -0.014503479, -0.00036525726, 0.013877869, -0.02748108, -0.03656006, -0.008369446, 0.0032253265, 0.0...
http://arxiv.org/abs/2308.13315v2
Construction Grammar and Language Models
Harish Tayyar Madabushi, Laurence Romain, Petar Milin, Dagmar Divjak
cs.CL
Recent progress in deep learning and natural language processing has given rise to powerful models that are primarily trained on a cloze-like task and show some evidence of having access to substantial linguistic information, including some constructional knowledge. This groundbreaking discovery presents an exciting op...
http://arxiv.org/pdf/2308.13315v2
2023-09-04T21:03:51
2023-08-25T11:37:56
{ "title": [ -0.0473938, -0.015808105, -0.03302002, -0.06530762, -0.007896423, 0.028396606, -0.06829834, 0.006565094, -0.023040771, 0.026290894, -0.009246826, -0.0018320084, -0.030334473, -0.02268982, -0.032684326, -0.004676819, -0.024536133, 0.0...
http://arxiv.org/abs/2309.00639v1
Misinformation Concierge: A Proof-of-Concept with Curated Twitter Dataset on COVID-19 Vaccination
Shakshi Sharma, Anwitaman Datta, Vigneshwaran Shankaran, Rajesh Sharma
cs.CL, cs.SI
We demonstrate the Misinformation Concierge, a proof-of-concept that provides actionable intelligence on misinformation prevalent in social media. Specifically, it uses language processing and machine learning tools to identify subtopics of discourse and discern non/misleading posts; presents statistical reports for po...
http://arxiv.org/pdf/2309.00639v1
2023-08-25T10:06:05
2023-08-25T10:06:05
{ "title": [ 0.024398804, -0.0126571655, -0.03262329, -0.07141113, -0.01260376, -0.01953125, 0.003501892, 0.042266846, -0.0395813, 0.03555298, -0.031066895, 0.040618896, -0.008285522, -0.025634766, -0.005794525, -0.033325195, -0.008255005, 0.0441...
http://arxiv.org/abs/2308.13259v2
Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answering
Keheng Wang, Feiyu Duan, Sirui Wang, Peiguang Li, Yunsen Xian, Chuantao Yin, Wenge Rong, Zhang Xiong
cs.CL, cs.AI
Equipped with Chain-of-Thought (CoT), Large language models (LLMs) have shown impressive reasoning ability in various downstream tasks. Even so, suffering from hallucinations and the inability to access external knowledge, LLMs often come with incorrect or unfaithful intermediate reasoning steps, especially in the cont...
http://arxiv.org/pdf/2308.13259v2
2023-10-28T12:19:29
2023-08-25T09:23:55
{ "title": [ -0.045654297, -0.02355957, -0.05340576, -0.051757812, -0.036254883, -0.0042877197, -0.054534912, 0.021972656, 0.013206482, 0.03451538, -0.044830322, 0.014389038, 0.014709473, -0.010887146, -0.053985596, -0.0066337585, 0.0003902912, 0...
http://arxiv.org/abs/2308.13207v1
LLM2KB: Constructing Knowledge Bases using instruction tuned context aware Large Language Models
Anmol Nayak, Hari Prasad Timmapathini
cs.CL
The advent of Large Language Models (LLM) has revolutionized the field of natural language processing, enabling significant progress in various applications. One key area of interest is the construction of Knowledge Bases (KB) using these powerful models. Knowledge bases serve as repositories of structured information,...
http://arxiv.org/pdf/2308.13207v1
2023-08-25T07:04:16
2023-08-25T07:04:16
{ "title": [ -0.052520752, -0.022247314, -0.049468994, -0.0847168, -0.0065345764, 0.026763916, -0.086120605, 0.023406982, 0.0009660721, 0.03100586, -0.044647217, 0.014312744, 0.004852295, -0.0052337646, -0.02331543, -0.003364563, -0.0026302338, 0...
http://arxiv.org/abs/2308.13198v1
Journey to the Center of the Knowledge Neurons: Discoveries of Language-Independent Knowledge Neurons and Degenerate Knowledge Neurons
Yuheng Chen, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao
cs.CL
Pre-trained language models (PLMs) contain vast amounts of factual knowledge, but how the knowledge is stored in the parameters remains unclear. This paper delves into the complex task of understanding how factual knowledge is stored in multilingual PLMs, and introduces the Architecture-adapted Multilingual Integrated ...
http://arxiv.org/pdf/2308.13198v1
2023-08-25T06:26:05
2023-08-25T06:26:05
{ "title": [ -0.025497437, -0.0017576218, -0.05380249, -0.011619568, -0.0008416176, -0.029830933, 0.012107849, -0.019805908, 0.00045490265, 0.034698486, -0.015167236, -0.023742676, 0.023880005, -0.031982422, -0.041412354, 0.007724762, 0.010101318, ...
http://arxiv.org/abs/2308.13192v1
Formalising Natural Language Quantifiers for Human-Robot Interactions
Stefan Morar, Adrian Groza, Mihai Pomarlan
cs.AI, cs.CL, cs.RO
We present a method for formalising quantifiers in natural language in the context of human-robot interactions. The solution is based on first-order logic extended with capabilities to represent the cardinality of variables, operating similarly to generalised quantifiers. To demonstrate the method, we designed an end-t...
http://arxiv.org/pdf/2308.13192v1
2023-08-25T06:05:57
2023-08-25T06:05:57
{ "title": [ -0.0063056946, 0.0038166046, -0.027694702, -0.0413208, 0.0074653625, -0.016052246, -0.030685425, -0.025268555, -0.010810852, 0.03479004, -0.02645874, -0.009048462, -0.0057411194, -0.0072746277, -0.054260254, -0.02558899, -0.012901306, ...
http://arxiv.org/abs/2308.13191v1
Chunk, Align, Select: A Simple Long-sequence Processing Method for Transformers
Jiawen Xie, Pengyu Cheng, Xiao Liang, Yong Dai, Nan Du
cs.CL, cs.AI
Although dominant in natural language processing, transformer-based models remain challenged by the task of long-sequence processing, because the computational cost of self-attention operations in transformers swells quadratically with the input sequence length. To alleviate the complexity of long-sequence processing, ...
http://arxiv.org/pdf/2308.13191v1
2023-08-25T05:52:05
2023-08-25T05:52:05
{ "title": [ -0.013046265, 0.004699707, -0.059326172, -0.09082031, 0.033599854, -0.0072402954, -0.04888916, -0.004840851, 0.0019369125, 0.047058105, -0.034210205, 0.0051345825, 0.016616821, -0.020095825, -0.046051025, 0.008850098, 0.008834839, 0....
http://arxiv.org/abs/2308.13569v1
Discovering Mental Health Research Topics with Topic Modeling
Xin Gao, Cem Sazara
cs.CL, cs.LG
Mental health significantly influences various aspects of our daily lives, and its importance has been increasingly recognized by the research community and the general public, particularly in the wake of the COVID-19 pandemic. This heightened interest is evident in the growing number of publications dedicated to menta...
http://arxiv.org/pdf/2308.13569v1
2023-08-25T05:25:05
2023-08-25T05:25:05
{ "title": [ -0.010498047, 0.010810852, -0.03488159, -0.036590576, 0.022140503, -0.005027771, -0.025344849, -0.033233643, -0.00019550323, 0.02822876, -0.011100769, 0.030288696, -0.017150879, -0.018753052, -0.030883789, 0.015914917, -0.009727478, ...
http://arxiv.org/abs/2308.13177v1
How to Evaluate the Generalization of Detection? A Benchmark for Comprehensive Open-Vocabulary Detection
Yiyang Yao, Peng Liu, Tiancheng Zhao, Qianqian Zhang, Jiajia Liao, Chunxin Fang, Kyusong Lee, Qing Wang
cs.CV, cs.CL
Object detection (OD) in computer vision has made significant progress in recent years, transitioning from closed-set labels to open-vocabulary detection (OVD) based on large-scale vision-language pre-training (VLP). However, current evaluation methods and datasets are limited to testing generalization over object type...
http://arxiv.org/pdf/2308.13177v1
2023-08-25T04:54:32
2023-08-25T04:54:32
{ "title": [ -0.011947632, 0.021209717, -0.023971558, -0.044830322, 0.033477783, -0.0028648376, -0.030761719, -0.05026245, -0.0011587143, 0.017730713, -0.039489746, 0.038269043, -0.036102295, -0.006565094, -0.033081055, -0.0107421875, -0.0076179504, ...
http://arxiv.org/abs/2308.13173v1
DISGO: Automatic End-to-End Evaluation for Scene Text OCR
Mei-Yuh Hwang, Yangyang Shi, Ankit Ramchandani, Guan Pang, Praveen Krishnan, Lucas Kabela, Frank Seide, Samyak Datta, Jun Liu
cs.CV, cs.CL
This paper discusses the challenges of optical character recognition (OCR) on natural scenes, which is harder than OCR on documents due to the wild content and various image backgrounds. We propose to uniformly use word error rates (WER) as a new measurement for evaluating scene-text OCR, both end-to-end (e2e) performa...
http://arxiv.org/pdf/2308.13173v1
2023-08-25T04:45:37
2023-08-25T04:45:37
{ "title": [ -0.032928467, 0.0031089783, -0.017959595, -0.03744507, -0.00045681, -0.00015258789, -0.013717651, -0.018249512, 0.010917664, 0.039642334, -0.04925537, 0.009529114, 0.015220642, -0.0012922287, -0.04650879, -0.014274597, -0.008049011, ...
http://arxiv.org/abs/2308.13170v1
Measuring Spurious Correlation in Classification: 'Clever Hans' in Translationese
Angana Borah, Daria Pylypenko, Cristina Espana-Bonet, Josef van Genabith
cs.CL
Recent work has shown evidence of 'Clever Hans' behavior in high-performance neural translationese classifiers, where BERT-based classifiers capitalize on spurious correlations, in particular topic information, between data and target classification labels, rather than genuine translationese signals. Translationese sig...
http://arxiv.org/pdf/2308.13170v1
2023-08-25T04:19:58
2023-08-25T04:19:58
{ "title": [ -0.03543091, 0.010505676, -0.035369873, 0.040802002, -0.018218994, -0.038848877, 0.028320312, -0.050994873, -0.017456055, 0.035003662, -0.009605408, 0.0032577515, 0.012916565, -0.039123535, -0.03289795, -0.014633179, -0.032989502, 0....
http://arxiv.org/abs/2309.08609v2
Media of Langue
Goki Muramoto, Atsuki Sato, Takayoshi Koyama
cs.CL, cs.HC
This paper aims to archive the materials behind "Media of Langue" by Goki Muramoto et al. Media of Langue is a new dictionary and public sculpture that depicts the map of meaning on the boundary between languages solely from the vast events of "this word was translated into that word" and two forces: repulsion between ...
http://arxiv.org/pdf/2309.08609v2
2023-09-19T02:00:36
2023-08-25T03:54:20
{ "title": [ -0.025115967, -0.016677856, -0.018966675, 0.0013589859, -0.070129395, -0.020095825, -0.023330688, -0.00061035156, -0.025436401, 0.059448242, 0.008361816, -0.03286743, 0.020736694, -0.043701172, 0.015670776, 0.0008773804, 0.01651001, ...
http://arxiv.org/abs/2308.13149v1
SciEval: A Multi-Level Large Language Model Evaluation Benchmark for Scientific Research
Liangtai Sun, Yang Han, Zihan Zhao, Da Ma, Zhennan Shen, Baocai Chen, Lu Chen, Kai Yu
cs.CL
Recently, there has been growing interest in using Large Language Models (LLMs) for scientific research. Numerous benchmarks have been proposed to evaluate the ability of LLMs for scientific research. However, current benchmarks are mostly based on pre-collected objective questions. This design suffers from data leakag...
http://arxiv.org/pdf/2308.13149v1
2023-08-25T03:05:33
2023-08-25T03:05:33
{ "title": [ -0.046844482, -0.021606445, -0.04547119, -0.05847168, 0.003271103, 0.010421753, -0.070129395, -0.0138168335, 0.017410278, 0.032226562, 0.01751709, -0.003993988, 0.006832123, -0.037994385, -0.054748535, 0.027526855, -0.019348145, 0.04...
http://arxiv.org/abs/2308.13139v1
MatchXML: An Efficient Text-label Matching Framework for Extreme Multi-label Text Classification
Hui Ye, Rajshekhar Sunderraman, Shihao Ji
cs.CL, cs.LG
The eXtreme Multi-label text Classification(XMC) refers to training a classifier that assigns a text sample with relevant labels from an extremely large-scale label set (e.g., millions of labels). We propose MatchXML, an efficient text-label matching framework for XMC. We observe that the label embeddings generated fro...
http://arxiv.org/pdf/2308.13139v1
2023-08-25T02:32:36
2023-08-25T02:32:36
{ "title": [ -0.020462036, -0.002714157, -0.03918457, -0.052520752, 0.018936157, -0.0017967224, -0.005886078, -0.0038433075, 0.01083374, 0.018569946, -0.022720337, 0.024902344, -0.01550293, -0.0107803345, -0.042297363, -0.0017251968, -0.0262146, ...
http://arxiv.org/abs/2308.13137v2
OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
Wenqi Shao, Mengzhao Chen, Zhaoyang Zhang, Peng Xu, Lirui Zhao, Zhiqian Li, Kaipeng Zhang, Peng Gao, Yu Qiao, Ping Luo
cs.LG, cs.CL
Large language models (LLMs) have revolutionized natural language processing tasks. However, their practical deployment is hindered by their immense memory and computation requirements. Although recent post-training quantization (PTQ) methods are effective in reducing memory footprint and improving the computational ef...
http://arxiv.org/pdf/2308.13137v2
2023-10-22T08:10:10
2023-08-25T02:28:35
{ "title": [ -0.045043945, -0.0033073425, -0.056152344, -0.04611206, -0.00037670135, -0.006134033, -0.05908203, -0.002943039, 0.00299263, 0.05029297, -0.020401001, 0.000808239, 0.022827148, -0.011955261, -0.04562378, 0.014465332, -0.0017366409, 0...
http://arxiv.org/abs/2308.13566v2
MLLM-DataEngine: An Iterative Refinement Approach for MLLM
Zhiyuan Zhao, Linke Ouyang, Bin Wang, Siyuan Huang, Pan Zhang, Xiaoyi Dong, Jiaqi Wang, Conghui He
cs.LG, cs.AI, cs.CL, cs.CV
Despite the great advance of Multimodal Large Language Models (MLLMs) in both instruction dataset building and benchmarking, the independence of training and evaluation makes current MLLMs hard to further improve their capability under the guidance of evaluation results with a relatively low human cost. In this paper, ...
http://arxiv.org/pdf/2308.13566v2
2023-09-11T08:28:40
2023-08-25T01:41:04
{ "title": [ -0.057159424, -0.009513855, -0.06652832, -0.08288574, -0.004333496, 0.019119263, -0.08001709, 0.018661499, -0.011199951, 0.055603027, 0.027145386, -0.006427765, 0.020126343, -0.036621094, -0.030578613, 0.020233154, -0.015319824, 0.01...
http://arxiv.org/abs/2308.13565v1
DARWIN Series: Domain Specific Large Language Models for Natural Science
Tong Xie, Yuwei Wan, Wei Huang, Zhenyu Yin, Yixuan Liu, Shaozhou Wang, Qingyuan Linghu, Chunyu Kit, Clara Grazian, Wenjie Zhang, Imran Razzak, Bram Hoex
cs.CL, cond-mat.mtrl-sci, physics.app-ph
Emerging tools bring forth fresh approaches to work, and the field of natural science is no different. In natural science, traditional manual, serial, and labour-intensive work is being augmented by automated, parallel, and iterative processes driven by artificial intelligence-based experimental automation and more. To...
http://arxiv.org/pdf/2308.13565v1
2023-08-25T01:40:48
2023-08-25T01:40:48
{ "title": [ -0.05026245, -0.021148682, -0.057922363, -0.063964844, 0.012107849, 0.0042304993, -0.047210693, 0.00566864, 0.010795593, 0.029785156, 0.011657715, -0.005130768, 0.012008667, -0.03515625, -0.042510986, 0.030319214, -0.015975952, 0.049...
http://arxiv.org/abs/2308.13563v1
Large Language Models in Analyzing Crash Narratives -- A Comparative Study of ChatGPT, BARD and GPT-4
Maroa Mumtarin, Md Samiullah Chowdhury, Jonathan Wood
cs.CL, cs.AI, cs.IR, cs.LG
In traffic safety research, extracting information from crash narratives using text analysis is a common practice. With recent advancements of large language models (LLM), it would be useful to know how the popular LLM interfaces perform in classifying or extracting information from crash narratives. To explore this, o...
http://arxiv.org/pdf/2308.13563v1
2023-08-25T00:09:16
2023-08-25T00:09:16
{ "title": [ -0.01689148, -0.026519775, -0.018859863, -0.053588867, -0.018920898, 0.0032043457, -0.044677734, 0.011947632, -0.04510498, 0.03741455, 0.010665894, 0.030166626, -0.011230469, -0.022232056, -0.01763916, 0.0077171326, -0.039733887, 0.0...
http://arxiv.org/abs/2308.13116v1
Sentence Embedding Models for Ancient Greek Using Multilingual Knowledge Distillation
Kevin Krahn, Derrick Tate, Andrew C. Lamicela
cs.CL, I.2.7
Contextual language models have been trained on Classical languages, including Ancient Greek and Latin, for tasks such as lemmatization, morphological tagging, part of speech tagging, authorship attribution, and detection of scribal errors. However, high-quality sentence embedding models for these historical languages ...
http://arxiv.org/pdf/2308.13116v1
2023-08-24T23:38:44
2023-08-24T23:38:44
{ "title": [ -0.00014519691, 0.008712769, -0.01876831, -0.039642334, -0.0066337585, -0.007850647, -0.046142578, -0.0046424866, 0.0552063, 0.021072388, -0.05429077, -0.011810303, 0.024673462, -0.014541626, -0.054534912, -0.0050582886, 0.008354187, ...