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/2309.00135v1
Construction Grammar and Artificial Intelligence
Katrien Beuls, Paul Van Eecke
cs.AI, cs.CL
In this chapter, we argue that it is highly beneficial for the contemporary construction grammarian to have a thorough understanding of the strong relationship between the research fields of construction grammar and artificial intelligence. We start by unravelling the historical links between the two fields, showing th...
http://arxiv.org/pdf/2309.00135v1
2023-08-31T21:15:06
2023-08-31T21:15:06
{ "title": [ -0.041656494, -0.01890564, -0.025405884, -0.044677734, -0.010284424, 0.00806427, -0.02444458, 0.0010471344, -0.0140686035, 0.020095825, -0.03289795, 0.006843567, -0.041259766, -0.012626648, -0.027008057, -0.010673523, -0.019607544, 0...
http://arxiv.org/abs/2309.00126v1
QS-TTS: Towards Semi-Supervised Text-to-Speech Synthesis via Vector-Quantized Self-Supervised Speech Representation Learning
Haohan Guo, Fenglong Xie, Jiawen Kang, Yujia Xiao, Xixin Wu, Helen Meng
cs.SD, cs.CL, eess.AS
This paper proposes a novel semi-supervised TTS framework, QS-TTS, to improve TTS quality with lower supervised data requirements via Vector-Quantized Self-Supervised Speech Representation Learning (VQ-S3RL) utilizing more unlabeled speech audio. This framework comprises two VQ-S3R learners: first, the principal learne...
http://arxiv.org/pdf/2309.00126v1
2023-08-31T20:25:44
2023-08-31T20:25:44
{ "title": [ -0.0043296814, 0.010818481, -0.031143188, -0.06311035, 0.0060195923, 0.0008506775, -0.05130005, 0.004623413, -0.029891968, 0.055786133, -0.008865356, 0.020385742, 0.0063552856, -0.01940918, -0.050811768, -0.009994507, -0.010047913, 0...
http://arxiv.org/abs/2309.00087v1
Large language models in medicine: the potentials and pitfalls
Jesutofunmi A. Omiye, Haiwen Gui, Shawheen J. Rezaei, James Zou, Roxana Daneshjou
cs.CL, cs.AI, cs.CY
Large language models (LLMs) have been applied to tasks in healthcare, ranging from medical exam questions to responding to patient questions. With increasing institutional partnerships between companies producing LLMs and healthcare systems, real world clinical application is coming closer to reality. As these models ...
http://arxiv.org/pdf/2309.00087v1
2023-08-31T19:06:39
2023-08-31T19:06:39
{ "title": [ -0.03652954, -0.009941101, -0.036193848, -0.06323242, 0.0046691895, 0.010032654, -0.056610107, -0.00894165, -0.013244629, 0.031463623, 0.014015198, -0.0095825195, -0.001947403, -0.038330078, -0.031402588, 0.019500732, -0.0017976761, ...
http://arxiv.org/abs/2309.00071v2
YaRN: Efficient Context Window Extension of Large Language Models
Bowen Peng, Jeffrey Quesnelle, Honglu Fan, Enrico Shippole
cs.CL, cs.AI, cs.LG
Rotary Position Embeddings (RoPE) have been shown to effectively encode positional information in transformer-based language models. However, these models fail to generalize past the sequence length they were trained on. We present YaRN (Yet another RoPE extensioN method), a compute-efficient method to extend the conte...
http://arxiv.org/pdf/2309.00071v2
2023-11-01T17:28:26
2023-08-31T18:18:07
{ "title": [ -0.025604248, -0.034088135, -0.045440674, -0.099121094, 0.013008118, 0.015899658, -0.059417725, 0.020599365, 0.017105103, 0.043121338, -0.0103302, -0.00087976456, 0.03479004, -0.016403198, -0.048950195, 0.029251099, -0.008659363, 0.0...
http://arxiv.org/abs/2308.16911v1
PointLLM: Empowering Large Language Models to Understand Point Clouds
Runsen Xu, Xiaolong Wang, Tai Wang, Yilun Chen, Jiangmiao Pang, Dahua Lin
cs.CV, cs.AI, cs.CL
The unprecedented advancements in Large Language Models (LLMs) have created a profound impact on natural language processing but are yet to fully embrace the realm of 3D understanding. This paper introduces PointLLM, a preliminary effort to fill this gap, thereby enabling LLMs to understand point clouds and offering a ...
http://arxiv.org/pdf/2308.16911v1
2023-08-31T17:59:46
2023-08-31T17:59:46
{ "title": [ -0.03237915, -0.024612427, -0.029022217, -0.06451416, -0.0119018555, 0.015136719, -0.054260254, -0.0057373047, 0.0093688965, 0.02041626, -0.02078247, -0.015396118, -0.00007092953, -0.024520874, -0.022964478, 0.0013170242, -0.008079529, ...
http://arxiv.org/abs/2308.16898v2
Transformers as Support Vector Machines
Davoud Ataee Tarzanagh, Yingcong Li, Christos Thrampoulidis, Samet Oymak
cs.LG, cs.AI, cs.CL, math.OC
Since its inception in "Attention Is All You Need", transformer architecture has led to revolutionary advancements in NLP. The attention layer within the transformer admits a sequence of input tokens $X$ and makes them interact through pairwise similarities computed as softmax$(XQK^\top X^\top)$, where $(K,Q)$ are the ...
http://arxiv.org/pdf/2308.16898v2
2023-09-07T17:50:52
2023-08-31T17:57:50
{ "title": [ -0.008651733, 0.0005054474, -0.014190674, -0.046661377, 0.017486572, 0.003129959, -0.022720337, -0.04623413, 0.0154800415, 0.019332886, -0.03564453, -0.0074005127, -0.010124207, -0.026748657, -0.046691895, 0.0037708282, 0.0067100525, ...
http://arxiv.org/abs/2308.16890v2
TouchStone: Evaluating Vision-Language Models by Language Models
Shuai Bai, Shusheng Yang, Jinze Bai, Peng Wang, Xingxuan Zhang, Junyang Lin, Xinggang Wang, Chang Zhou, Jingren Zhou
cs.CV, cs.CL
Large vision-language models (LVLMs) have recently witnessed rapid advancements, exhibiting a remarkable capacity for perceiving, understanding, and processing visual information by connecting visual receptor with large language models (LLMs). However, current assessments mainly focus on recognizing and reasoning abili...
http://arxiv.org/pdf/2308.16890v2
2023-09-04T15:06:15
2023-08-31T17:52:04
{ "title": [ -0.037872314, -0.020690918, -0.040100098, -0.042999268, -0.011146545, -0.008766174, -0.04977417, -0.006412506, 0.022399902, 0.02154541, -0.028686523, -0.0061950684, 0.007888794, -0.01260376, -0.053771973, 0.0026760101, -0.0029773712, ...
http://arxiv.org/abs/2308.16884v1
The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants
Lucas Bandarkar, Davis Liang, Benjamin Muller, Mikel Artetxe, Satya Narayan Shukla, Donald Husa, Naman Goyal, Abhinandan Krishnan, Luke Zettlemoyer, Madian Khabsa
cs.CL, cs.AI, cs.LG, I.2.7
We present Belebele, a multiple-choice machine reading comprehension (MRC) dataset spanning 122 language variants. Significantly expanding the language coverage of natural language understanding (NLU) benchmarks, this dataset enables the evaluation of text models in high-, medium-, and low-resource languages. Each ques...
http://arxiv.org/pdf/2308.16884v1
2023-08-31T17:43:08
2023-08-31T17:43:08
{ "title": [ -0.055511475, -0.012702942, -0.028961182, -0.06689453, -0.027816772, 0.014045715, -0.09460449, -0.009712219, 0.0072517395, 0.05819702, -0.0022335052, 0.014877319, 0.017440796, -0.009986877, -0.057128906, 0.023544312, -0.008682251, 0....
http://arxiv.org/abs/2308.16871v1
The Gender-GAP Pipeline: A Gender-Aware Polyglot Pipeline for Gender Characterisation in 55 Languages
Benjamin Muller, Belen Alastruey, Prangthip Hansanti, Elahe Kalbassi, Christophe Ropers, Eric Michael Smith, Adina Williams, Luke Zettlemoyer, Pierre Andrews, Marta R. Costa-jussà
cs.CL, cs.AI
Gender biases in language generation systems are challenging to mitigate. One possible source for these biases is gender representation disparities in the training and evaluation data. Despite recent progress in documenting this problem and many attempts at mitigating it, we still lack shared methodology and tooling to...
http://arxiv.org/pdf/2308.16871v1
2023-08-31T17:20:50
2023-08-31T17:20:50
{ "title": [ -0.05331421, -0.007637024, -0.058135986, -0.060943604, -0.0050239563, -0.01020813, -0.05090332, 0.008575439, -0.015045166, 0.0317688, -0.013877869, -0.00024151802, 0.0184021, -0.02281189, -0.036956787, -0.010658264, -0.03555298, 0.06...
http://arxiv.org/abs/2308.16824v2
Can Programming Languages Boost Each Other via Instruction Tuning?
Daoguang Zan, Ailun Yu, Bo Shen, Jiaxin Zhang, Taihong Chen, Bing Geng, Bei Chen, Jichuan Ji, Yafen Yao, Yongji Wang, Qianxiang Wang
cs.CL, cs.AI, cs.PL, cs.SE
When human programmers have mastered a programming language, it would be easier when they learn a new programming language. In this report, we focus on exploring whether programming languages can boost each other during the instruction fine-tuning phase of code large language models. We conduct extensive experiments of...
http://arxiv.org/pdf/2308.16824v2
2023-09-03T08:30:29
2023-08-31T15:53:51
{ "title": [ -0.02229309, -0.021392822, -0.034851074, -0.036071777, -0.040405273, -0.0015258789, -0.020690918, 0.026672363, -0.056396484, 0.025787354, -0.024261475, 0.023605347, -0.035003662, -0.028900146, 0.024383545, -0.032684326, 0.008216858, ...
http://arxiv.org/abs/2308.16797v2
Simple LLM Prompting is State-of-the-Art for Robust and Multilingual Dialogue Evaluation
John Mendonça, Patrícia Pereira, Helena Moniz, João Paulo Carvalho, Alon Lavie, Isabel Trancoso
cs.CL
Despite significant research effort in the development of automatic dialogue evaluation metrics, little thought is given to evaluating dialogues other than in English. At the same time, ensuring metrics are invariant to semantically similar responses is also an overlooked topic. In order to achieve the desired properti...
http://arxiv.org/pdf/2308.16797v2
2023-09-08T11:24:06
2023-08-31T15:19:28
{ "title": [ -0.004802704, -0.020645142, -0.029373169, -0.0725708, -0.04232788, 0.000087976456, -0.10559082, 0.006111145, 0.015655518, 0.018692017, -0.00045466423, 0.012924194, -0.02355957, -0.050323486, 0.007286072, -0.00043320656, -0.015312195, ...
http://arxiv.org/abs/2308.16795v1
Towards Multilingual Automatic Dialogue Evaluation
John Mendonça, Alon Lavie, Isabel Trancoso
cs.CL
The main limiting factor in the development of robust multilingual dialogue evaluation metrics is the lack of multilingual data and the limited availability of open sourced multilingual dialogue systems. In this work, we propose a workaround for this lack of data by leveraging a strong multilingual pretrained LLM and a...
http://arxiv.org/pdf/2308.16795v1
2023-08-31T15:15:26
2023-08-31T15:15:26
{ "title": [ -0.022979736, -0.008636475, -0.01663208, -0.065979004, 0.00007921457, 0.016403198, -0.06048584, 0.003446579, 0.008369446, 0.03491211, 0.0027389526, -0.011177063, -0.016204834, -0.019454956, -0.068359375, -0.0013465881, -0.03314209, 0...
http://arxiv.org/abs/2308.16770v1
Enhancing PLM Performance on Labour Market Tasks via Instruction-based Finetuning and Prompt-tuning with Rules
Jarno Vrolijk, David Graus
cs.CL
The increased digitization of the labour market has given researchers, educators, and companies the means to analyze and better understand the labour market. However, labour market resources, although available in high volumes, tend to be unstructured, and as such, research towards methodologies for the identification,...
http://arxiv.org/pdf/2308.16770v1
2023-08-31T14:47:00
2023-08-31T14:47:00
{ "title": [ 0.014205933, -0.01499939, -0.023406982, -0.058685303, -0.021026611, -0.0076828003, -0.026550293, 0.01687622, -0.04510498, 0.031402588, -0.011146545, 0.04168701, -0.010414124, -0.0104599, -0.009346008, -0.023101807, -0.0126953125, 0.0...
http://arxiv.org/abs/2308.16763v2
Ladder-of-Thought: Using Knowledge as Steps to Elevate Stance Detection
Kairui Hu, Ming Yan, Joey Tianyi Zhou, Ivor W. Tsang, Wen Haw Chong, Yong Keong Yap
cs.CL, cs.AI
Stance detection aims to identify the attitude expressed in a document towards a given target. Techniques such as Chain-of-Thought (CoT) prompting have advanced this task, enhancing a model's reasoning capabilities through the derivation of intermediate rationales. However, CoT relies primarily on a model's pre-trained...
http://arxiv.org/pdf/2308.16763v2
2023-09-07T09:15:24
2023-08-31T14:31:48
{ "title": [ -0.0011930466, 0.00081300735, -0.0357666, -0.036468506, -0.010032654, -0.031311035, -0.030273438, -0.028900146, 0.009971619, 0.021255493, -0.062805176, 0.00806427, -0.010253906, -0.0099105835, -0.05316162, -0.02949524, 0.022918701, 0...
http://arxiv.org/abs/2308.16705v2
CREHate: A CRoss-cultural English Hate Speech Dataset
Nayeon Lee, Chani Jung, Junho Myung, Jiho Jin, Jose Camacho-Collados, Juho Kim, Alice Oh
cs.CL, cs.AI
Most NLP datasets neglect the cultural diversity among language speakers, resulting in a critical shortcoming in hate speech detection and other culturally sensitive tasks. To address this, we introduce CREHate, a CRoss-cultural English Hate speech dataset. To construct CREHate, we follow a two-step procedure: 1) cultu...
http://arxiv.org/pdf/2308.16705v2
2023-11-01T12:58:22
2023-08-31T13:14:47
{ "title": [ 0.014350891, -0.009162903, -0.03262329, -0.06286621, 0.025360107, -0.023223877, -0.006175995, 0.0024604797, -0.008804321, 0.03363037, -0.011192322, 0.024642944, 0.014923096, -0.010513306, -0.034454346, 0.020309448, -0.018997192, 0.05...
http://arxiv.org/abs/2309.06580v1
Can humans help BERT gain "confidence"?
Piyush Agrawal
cs.CL, cs.AI
The advancements in artificial intelligence over the last decade have opened a multitude of avenues for interdisciplinary research. Since the idea of artificial intelligence was inspired by the working of neurons in the brain, it seems pretty practical to combine the two fields and take the help of cognitive data to tr...
http://arxiv.org/pdf/2309.06580v1
2023-08-31T13:12:28
2023-08-31T13:12:28
{ "title": [ 0.006843567, -0.0059013367, -0.016586304, 0.0002553463, -0.035614014, -0.014801025, -0.008140564, -0.0209198, -0.026062012, 0.053863525, -0.021728516, 0.006389618, 0.015823364, -0.00283432, -0.0056381226, -0.017547607, -0.011795044, ...
http://arxiv.org/abs/2309.00649v1
GPT has become financially literate: Insights from financial literacy tests of GPT and a preliminary test of how people use it as a source of advice
Paweł Niszczota, Sami Abbas
cs.CL, cs.AI, cs.CY, econ.GN, q-fin.EC, I.2.7; J.4; H.3.3
We assess the ability of GPT -- a large language model -- to serve as a financial robo-advisor for the masses, by using a financial literacy test. Davinci and ChatGPT based on GPT-3.5 score 66% and 65% on the financial literacy test, respectively, compared to a baseline of 33%. However, ChatGPT based on GPT-4 achieves ...
http://arxiv.org/pdf/2309.00649v1
2023-08-31T12:53:52
2023-08-31T12:53:52
{ "title": [ 0.012916565, 0.0016450882, -0.005558014, 0.012329102, -0.023590088, -0.025390625, -0.005710602, -0.036987305, -0.003168106, 0.008110046, -0.058654785, 0.025390625, 0.005886078, -0.024627686, 0.007873535, -0.02192688, 0.010101318, 0.0...
http://arxiv.org/abs/2308.16692v1
SpeechTokenizer: Unified Speech Tokenizer for Speech Large Language Models
Xin Zhang, Dong Zhang, Shimin Li, Yaqian Zhou, Xipeng Qiu
cs.CL, cs.SD, eess.AS
Current speech large language models build upon discrete speech representations, which can be categorized into semantic tokens and acoustic tokens. However, existing speech tokens are not specifically designed for speech language modeling. To assess the suitability of speech tokens for building speech language models, ...
http://arxiv.org/pdf/2308.16692v1
2023-08-31T12:53:09
2023-08-31T12:53:09
{ "title": [ -0.027008057, -0.014717102, -0.06854248, -0.0713501, 0.018630981, -0.012710571, -0.054138184, 0.014526367, -0.017837524, 0.042022705, -0.011230469, -0.005470276, 0.029006958, -0.02696228, -0.030441284, 0.0065307617, -0.011413574, 0.0...
http://arxiv.org/abs/2308.16688v1
Using Large Language Models to Automate Category and Trend Analysis of Scientific Articles: An Application in Ophthalmology
Hina Raja, Asim Munawar, Mohammad Delsoz, Mohammad Elahi, Yeganeh Madadi, Amr Hassan, Hashem Abu Serhan, Onur Inam, Luis Hermandez, Sang Tran, Wuqas Munir, Alaa Abd-Alrazaq, Hao Chen, SiamakYousefi
cs.CL, cs.AI
Purpose: In this paper, we present an automated method for article classification, leveraging the power of Large Language Models (LLM). The primary focus is on the field of ophthalmology, but the model is extendable to other fields. Methods: We have developed a model based on Natural Language Processing (NLP) technique...
http://arxiv.org/pdf/2308.16688v1
2023-08-31T12:45:53
2023-08-31T12:45:53
{ "title": [ -0.034942627, -0.012611389, -0.063964844, -0.05886841, -0.0070381165, -0.0039405823, -0.0769043, -0.003862381, 0.0013828278, 0.047576904, -0.00041890144, 0.0014543533, 0.0035438538, -0.010620117, -0.05758667, 0.0096206665, 0.010643005, ...
http://arxiv.org/abs/2308.16687v2
DictaBERT: A State-of-the-Art BERT Suite for Modern Hebrew
Shaltiel Shmidman, Avi Shmidman, Moshe Koppel
cs.CL
We present DictaBERT, a new state-of-the-art pre-trained BERT model for modern Hebrew, outperforming existing models on most benchmarks. Additionally, we release three fine-tuned versions of the model, designed to perform three specific foundational tasks in the analysis of Hebrew texts: prefix segmentation, morphologi...
http://arxiv.org/pdf/2308.16687v2
2023-10-13T13:56:28
2023-08-31T12:43:18
{ "title": [ 0.022247314, -0.047912598, -0.018493652, -0.101257324, -0.04269409, 0.021026611, -0.07891846, 0.040039062, -0.007545471, 0.030838013, -0.0013151169, 0.012550354, 0.057250977, -0.04260254, 0.01953125, 0.020584106, -0.017471313, 0.0256...
http://arxiv.org/abs/2308.16622v1
Developing a Scalable Benchmark for Assessing Large Language Models in Knowledge Graph Engineering
Lars-Peter Meyer, Johannes Frey, Kurt Junghanns, Felix Brei, Kirill Bulert, Sabine Gründer-Fahrer, Michael Martin
cs.AI, cs.CL, cs.DB
As the field of Large Language Models (LLMs) evolves at an accelerated pace, the critical need to assess and monitor their performance emerges. We introduce a benchmarking framework focused on knowledge graph engineering (KGE) accompanied by three challenges addressing syntax and error correction, facts extraction and ...
http://arxiv.org/pdf/2308.16622v1
2023-08-31T10:31:19
2023-08-31T10:31:19
{ "title": [ -0.060546875, -0.027038574, -0.06463623, -0.061767578, -0.01537323, 0.006942749, -0.072265625, 0.023956299, 0.008300781, 0.037139893, -0.0046577454, 0.0026226044, -0.0018081665, -0.017608643, -0.049102783, 0.006542206, 0.0009560585, ...
http://arxiv.org/abs/2308.16593v1
Towards Spontaneous Style Modeling with Semi-supervised Pre-training for Conversational Text-to-Speech Synthesis
Weiqin Li, Shun Lei, Qiaochu Huang, Yixuan Zhou, Zhiyong Wu, Shiyin Kang, Helen Meng
cs.SD, cs.CL, cs.LG, eess.AS
The spontaneous behavior that often occurs in conversations makes speech more human-like compared to reading-style. However, synthesizing spontaneous-style speech is challenging due to the lack of high-quality spontaneous datasets and the high cost of labeling spontaneous behavior. In this paper, we propose a semi-supe...
http://arxiv.org/pdf/2308.16593v1
2023-08-31T09:50:33
2023-08-31T09:50:33
{ "title": [ -0.020950317, -0.002544403, -0.03930664, -0.068847656, -0.012672424, 0.009849548, -0.08331299, 0.02508545, -0.017807007, 0.04345703, -0.011627197, 0.031982422, 0.030212402, -0.028533936, -0.0038223267, 0.008766174, -0.01146698, 0.017...
http://arxiv.org/abs/2308.16588v1
Interpreting Sentiment Composition with Latent Semantic Tree
Zhongtao Jiang, Yuanzhe Zhang, Cao Liu, Jiansong Chen, Jun Zhao, Kang Liu
cs.CL
As the key to sentiment analysis, sentiment composition considers the classification of a constituent via classifications of its contained sub-constituents and rules operated on them. Such compositionality has been widely studied previously in the form of hierarchical trees including untagged and sentiment ones, which ...
http://arxiv.org/pdf/2308.16588v1
2023-08-31T09:35:52
2023-08-31T09:35:52
{ "title": [ -0.011993408, -0.003211975, -0.017181396, -0.07836914, -0.0003771782, 0.02067566, -0.057159424, -0.008605957, 0.018295288, 0.023010254, -0.04486084, -0.011314392, -0.020248413, 0.0017261505, -0.07006836, -0.020477295, -0.0076828003, ...
http://arxiv.org/abs/2308.16584v1
Unsupervised Text Style Transfer with Deep Generative Models
Zhongtao Jiang, Yuanzhe Zhang, Yiming Ju, Kang Liu
cs.CL
We present a general framework for unsupervised text style transfer with deep generative models. The framework models each sentence-label pair in the non-parallel corpus as partially observed from a complete quadruplet which additionally contains two latent codes representing the content and style, respectively. These ...
http://arxiv.org/pdf/2308.16584v1
2023-08-31T09:29:35
2023-08-31T09:29:35
{ "title": [ -0.02949524, -0.0044021606, -0.045318604, -0.036315918, 0.0045700073, 0.0012226105, -0.029663086, -0.0066566467, 0.017730713, 0.041778564, -0.00049209595, 0.032348633, 0.023452759, -0.011985779, -0.024353027, 0.03074646, -0.0012283325, ...
http://arxiv.org/abs/2308.16577v1
Improving Mandarin Prosodic Structure Prediction with Multi-level Contextual Information
Jie Chen, Changhe Song, Deyi Tuo, Xixin Wu, Shiyin Kang, Zhiyong Wu, Helen Meng
cs.SD, cs.CL, eess.AS
For text-to-speech (TTS) synthesis, prosodic structure prediction (PSP) plays an important role in producing natural and intelligible speech. Although inter-utterance linguistic information can influence the speech interpretation of the target utterance, previous works on PSP mainly focus on utilizing intrautterance li...
http://arxiv.org/pdf/2308.16577v1
2023-08-31T09:19:15
2023-08-31T09:19:15
{ "title": [ 0.0039711, 0.0030879974, -0.015899658, -0.06713867, -0.020080566, 0.009887695, -0.05517578, 0.007827759, -0.0069999695, 0.030685425, 0.011077881, 0.0024852753, 0.012809753, -0.047424316, -0.024673462, 0.011749268, -0.012145996, 0.034...
http://arxiv.org/abs/2308.16549v1
Thesis Distillation: Investigating The Impact of Bias in NLP Models on Hate Speech Detection
Fatma Elsafoury
cs.CL
This paper is a summary of the work in my PhD thesis. In which, I investigate the impact of bias in NLP models on the task of hate speech detection from three perspectives: explainability, offensive stereotyping bias, and fairness. I discuss the main takeaways from my thesis and how they can benefit the broader NLP com...
http://arxiv.org/pdf/2308.16549v1
2023-08-31T08:40:41
2023-08-31T08:40:41
{ "title": [ 0.009796143, 0.012992859, -0.042938232, -0.027359009, 0.011680603, -0.03378296, -0.0041275024, -0.028121948, 0.012710571, 0.023864746, -0.05697632, -0.0016078949, 0.023956299, -0.002620697, -0.03967285, -0.022003174, -0.03262329, 0.0...
http://arxiv.org/abs/2308.16540v1
Time-Varying Quasi-Closed-Phase Analysis for Accurate Formant Tracking in Speech Signals
Dhananjaya Gowda, Sudarsana Reddy Kadiri, Brad Story, Paavo Alku
eess.AS, cs.CL, cs.SD, eess.SP
In this paper, we propose a new method for the accurate estimation and tracking of formants in speech signals using time-varying quasi-closed-phase (TVQCP) analysis. Conventional formant tracking methods typically adopt a two-stage estimate-and-track strategy wherein an initial set of formant candidates are estimated u...
http://arxiv.org/pdf/2308.16540v1
2023-08-31T08:30:20
2023-08-31T08:30:20
{ "title": [ -0.008155823, 0.0051574707, -0.029144287, -0.044677734, 0.010231018, 0.014030457, -0.016967773, 0.0045394897, -0.0037593842, 0.04107666, -0.021377563, -0.0036582947, 0.040985107, -0.040771484, -0.041015625, -0.0026855469, -0.012290955, ...
http://arxiv.org/abs/2308.16537v1
The Smart Data Extractor, a Clinician Friendly Solution to Accelerate and Improve the Data Collection During Clinical Trials
Sophie Quennelle, Maxime Douillet, Lisa Friedlander, Olivia Boyer, Anita Burgun, Antoine Neuraz, Nicolas Garcelon
q-bio.QM, cs.CL
In medical research, the traditional way to collect data, i.e. browsing patient files, has been proven to induce bias, errors, human labor and costs. We propose a semi-automated system able to extract every type of data, including notes. The Smart Data Extractor pre-populates clinic research forms by following rules. W...
http://arxiv.org/pdf/2308.16537v1
2023-08-31T08:28:11
2023-08-31T08:28:11
{ "title": [ 0.007949829, 0.0066604614, -0.073791504, -0.037139893, 0.007335663, -0.010406494, -0.013633728, -0.0038795471, 0.03149414, 0.023803711, -0.00016093254, 0.021652222, -0.01977539, -0.029556274, -0.010696411, 0.0027713776, 0.027297974, ...
http://arxiv.org/abs/2309.03219v1
Companion Animal Disease Diagnostics based on Literal-aware Medical Knowledge Graph Representation Learning
Van Thuy Hoang, Sang Thanh Nguyen, Sangmyeong Lee, Jooho Lee, Luong Vuong Nguyen, O-Joun Lee
cs.AI, cs.CL, cs.LG
Knowledge graph (KG) embedding has been used to benefit the diagnosis of animal diseases by analyzing electronic medical records (EMRs), such as notes and veterinary records. However, learning representations to capture entities and relations with literal information in KGs is challenging as the KGs show heterogeneous ...
http://arxiv.org/pdf/2309.03219v1
2023-08-31T08:14:07
2023-08-31T08:14:07
{ "title": [ -0.01979065, 0.0050582886, -0.05517578, -0.059906006, -0.016952515, 0.0070877075, -0.08148193, 0.0033912659, -0.008552551, 0.02357483, -0.02267456, -0.00046253204, 0.008171082, -0.03326416, -0.020645142, -0.014221191, 0.0073547363, 0...
http://arxiv.org/abs/2308.16498v1
Generalised Winograd Schema and its Contextuality
Kin Ian Lo, Mehrnoosh Sadrzadeh, Shane Mansfield
cs.CL, cs.AI, quant-ph
Ambiguities in natural language give rise to probability distributions over interpretations. The distributions are often over multiple ambiguous words at a time; a multiplicity which makes them a suitable topic for sheaf-theoretic models of quantum contextuality. Previous research showed that different quantitative mea...
http://arxiv.org/pdf/2308.16498v1
2023-08-31T07:00:21
2023-08-31T07:00:21
{ "title": [ 0.0362854, 0.025253296, -0.041778564, -0.06878662, 0.02331543, -0.014297485, -0.01411438, -0.015098572, 0.0014228821, 0.055633545, -0.007259369, -0.024139404, 0.030792236, -0.03857422, -0.04611206, 0.021270752, 0.0010061264, 0.021392...
http://arxiv.org/abs/2308.16475v1
Transformer Compression via Subspace Projection
Yuxuan Hu, Jing Zhang, Chen Zhao, Cuiping Li, Hong Chen
cs.CL, cs.AI, I.2.7
We propose TCSP, a novel method for compressing a transformer model by focusing on reducing the hidden size of the model. By projecting the whole transform model into a subspace, we enable matrix operations between the weight matrices in the model and features in a reduced-dimensional space, leading to significant redu...
http://arxiv.org/pdf/2308.16475v1
2023-08-31T05:40:14
2023-08-31T05:40:14
{ "title": [ -0.018249512, 0.02734375, -0.02079773, -0.033111572, 0.013710022, 0.0021457672, 0.004146576, -0.028274536, -0.014671326, 0.033233643, -0.022079468, 0.008651733, 0.014175415, -0.028152466, -0.0024662018, 0.014587402, 0.030563354, -0.0...
http://arxiv.org/abs/2308.16474v1
Enhancing Subtask Performance of Multi-modal Large Language Model
Yongqiang Zhao, Zhenyu Li, Feng Zhang, Xinhai Xu, Donghong Liu
cs.CL, cs.AI
Multi-modal Large Language Model (MLLM) refers to a model expanded from a Large Language Model (LLM) that possesses the capability to handle and infer multi-modal data. Current MLLMs typically begin by using LLMs to decompose tasks into multiple subtasks, then employing individual pre-trained models to complete specifi...
http://arxiv.org/pdf/2308.16474v1
2023-08-31T05:37:21
2023-08-31T05:37:21
{ "title": [ -0.032409668, 0.0065460205, -0.06109619, -0.07092285, 0.012107849, -0.008460999, -0.08276367, 0.005847931, 0.012634277, 0.0546875, -0.002111435, -0.008018494, 0.043182373, -0.021652222, -0.039154053, 0.018234253, -0.01210022, 0.01342...
http://arxiv.org/abs/2308.16469v2
Link Prediction for Wikipedia Articles as a Natural Language Inference Task
Chau-Thang Phan, Quoc-Nam Nguyen, Kiet Van Nguyen
cs.CL
Link prediction task is vital to automatically understanding the structure of large knowledge bases. In this paper, we present our system to solve this task at the Data Science and Advanced Analytics 2023 Competition "Efficient and Effective Link Prediction" (DSAA-2023 Competition) with a corpus containing 948,233 trai...
http://arxiv.org/pdf/2308.16469v2
2023-09-05T09:34:55
2023-08-31T05:25:04
{ "title": [ -0.0021266937, -0.0063323975, -0.03012085, -0.060028076, 0.0028057098, -0.0024681091, -0.02368164, -0.007949829, -0.0062828064, 0.020477295, -0.0063667297, 0.0098724365, 0.0070762634, -0.014305115, -0.026947021, 0.0059013367, -0.00926971...
http://arxiv.org/abs/2308.16463v2
Sparkles: Unlocking Chats Across Multiple Images for Multimodal Instruction-Following Models
Yupan Huang, Zaiqiao Meng, Fangyu Liu, Yixuan Su, Nigel Collier, Yutong Lu
cs.CV, cs.CL
Large language models exhibit enhanced zero-shot performance on various tasks when fine-tuned with instruction-following data. Multimodal instruction-following models extend these capabilities by integrating both text and images. However, existing models such as MiniGPT-4 face challenges in maintaining dialogue coheren...
http://arxiv.org/pdf/2308.16463v2
2023-10-02T03:31:17
2023-08-31T05:15:27
{ "title": [ -0.0003042221, -0.0001283884, -0.013298035, -0.028564453, -0.013671875, -0.007827759, -0.03201294, -0.029434204, 0.005619049, 0.037506104, -0.032989502, 0.04284668, 0.014640808, -0.009666443, -0.016586304, -0.0033664703, -0.009689331, ...
http://arxiv.org/abs/2308.16458v3
BioCoder: A Benchmark for Bioinformatics Code Generation with Contextual Pragmatic Knowledge
Xiangru Tang, Bill Qian, Rick Gao, Jiakang Chen, Xinyun Chen, Mark Gerstein
cs.LG, cs.AI, cs.CL
Pre-trained large language models have significantly improved code generation. As these models scale up, there is an increasing need for the output to handle more intricate tasks and to be appropriately specialized to particular domains. Bioinformatics provides an important domain. In this field generating functional p...
http://arxiv.org/pdf/2308.16458v3
2023-09-29T20:27:06
2023-08-31T04:52:58
{ "title": [ -0.03225708, -0.026306152, -0.038330078, -0.07196045, -0.0023784637, 0.013587952, -0.059631348, 0.008018494, 0.04071045, 0.026031494, -0.021331787, -0.0050697327, 0.0107421875, -0.024887085, -0.039978027, 0.012611389, -0.0025138855, ...
http://arxiv.org/abs/2308.16415v1
Knowledge Distillation from Non-streaming to Streaming ASR Encoder using Auxiliary Non-streaming Layer
Kyuhong Shim, Jinkyu Lee, Simyung Chang, Kyuwoong Hwang
cs.CL, eess.AS
Streaming automatic speech recognition (ASR) models are restricted from accessing future context, which results in worse performance compared to the non-streaming models. To improve the performance of streaming ASR, knowledge distillation (KD) from the non-streaming to streaming model has been studied, mainly focusing ...
http://arxiv.org/pdf/2308.16415v1
2023-08-31T02:58:33
2023-08-31T02:58:33
{ "title": [ -0.009788513, 0.007736206, -0.08050537, -0.06750488, -0.0074005127, -0.02279663, -0.046722412, 0.028335571, -0.00032258034, 0.05545044, -0.02180481, -0.015365601, 0.036712646, -0.03237915, -0.010971069, 0.0074043274, 0.019714355, 0.0...
http://arxiv.org/abs/2308.16349v2
Affective Visual Dialog: A Large-Scale Benchmark for Emotional Reasoning Based on Visually Grounded Conversations
Kilichbek Haydarov, Xiaoqian Shen, Avinash Madasu, Mahmoud Salem, Li-Jia Li, Gamaleldin Elsayed, Mohamed Elhoseiny
cs.CL
We introduce Affective Visual Dialog, an emotion explanation and reasoning task as a testbed for research on understanding the formation of emotions in visually grounded conversations. The task involves three skills: (1) Dialog-based Question Answering (2) Dialog-based Emotion Prediction and (3) Affective emotion expla...
http://arxiv.org/pdf/2308.16349v2
2023-09-12T04:37:37
2023-08-30T22:50:32
{ "title": [ -0.00053215027, -0.0034484863, -0.031982422, -0.053222656, -0.003211975, -0.011566162, -0.038635254, -0.012123108, 0.04525757, 0.021148682, -0.02268982, 0.003271103, 0.013366699, -0.031204224, -0.04522705, 0.001121521, 0.0065994263, ...
http://arxiv.org/abs/2308.16336v2
ToddlerBERTa: Exploiting BabyBERTa for Grammar Learning and Language Understanding
Omer Veysel Cagatan
cs.CL, cs.LG
We present ToddlerBERTa, a BabyBERTa-like language model, exploring its capabilities through five different models with varied hyperparameters. Evaluating on BLiMP, SuperGLUE, MSGS, and a Supplement benchmark from the BabyLM challenge, we find that smaller models can excel in specific tasks, while larger models perform...
http://arxiv.org/pdf/2308.16336v2
2023-11-08T12:31:35
2023-08-30T21:56:36
{ "title": [ -0.03414917, -0.023239136, -0.02607727, -0.0501709, -0.03152466, 0.0077781677, -0.050994873, 0.00051641464, -0.021972656, 0.03805542, -0.024505615, 0.0051231384, 0.009719849, -0.0063095093, -0.037841797, -0.008850098, -0.011161804, 0...
http://arxiv.org/abs/2308.16175v2
Quantifying Uncertainty in Answers from any Language Model and Enhancing their Trustworthiness
Jiuhai Chen, Jonas Mueller
cs.CL, cs.AI
We introduce BSDetector, a method for detecting bad and speculative answers from a pretrained Large Language Model by estimating a numeric confidence score for any output it generated. Our uncertainty quantification technique works for any LLM accessible only via a black-box API, whose training data remains unknown. By...
http://arxiv.org/pdf/2308.16175v2
2023-10-04T15:05:24
2023-08-30T17:53:25
{ "title": [ -0.030593872, 0.00843811, -0.07312012, -0.044891357, -0.0058059692, -0.040100098, -0.040496826, -0.009155273, -0.0013389587, 0.027450562, -0.017044067, 0.017150879, 0.006160736, -0.030883789, -0.024353027, 0.017593384, 0.0072669983, ...
http://arxiv.org/abs/2308.16149v2
Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models
Neha Sengupta, Sunil Kumar Sahu, Bokang Jia, Satheesh Katipomu, Haonan Li, Fajri Koto, William Marshall, Gurpreet Gosal, Cynthia Liu, Zhiming Chen, Osama Mohammed Afzal, Samta Kamboj, Onkar Pandit, Rahul Pal, Lalit Pradhan, Zain Muhammad Mujahid, Massa Baali, Xudong Han, Sondos Mahmoud Bsharat, Alham Fikri Aji, Zhiqian...
cs.CL, cs.AI, cs.LG, 68T50, F.2.2; I.2.7
We introduce Jais and Jais-chat, new state-of-the-art Arabic-centric foundation and instruction-tuned open generative large language models (LLMs). The models are based on the GPT-3 decoder-only architecture and are pretrained on a mixture of Arabic and English texts, including source code in various programming langua...
http://arxiv.org/pdf/2308.16149v2
2023-09-29T11:51:51
2023-08-30T17:07:17
{ "title": [ -0.04458618, -0.0345459, -0.031463623, -0.08526611, -0.023864746, 0.02319336, -0.07897949, 0.02368164, -0.022338867, 0.018371582, 0.0011644363, 0.0058403015, -0.018722534, -0.03656006, -0.028656006, -0.009765625, 0.0020122528, 0.0159...
http://arxiv.org/abs/2308.16137v4
LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models
Chi Han, Qifan Wang, Wenhan Xiong, Yu Chen, Heng Ji, Sinong Wang
cs.CL, cs.AI
In recent years, there have been remarkable advancements in the performance of Transformer-based Large Language Models (LLMs) across various domains. As these LLMs are deployed for increasingly complex domains, they often face the need to follow longer user prompts or generate longer texts. In these situations, the $\t...
http://arxiv.org/pdf/2308.16137v4
2023-10-03T02:46:50
2023-08-30T16:47:51
{ "title": [ -0.04168701, -0.020095825, -0.052886963, -0.07354736, -0.011383057, 0.009643555, -0.07678223, 0.019577026, -0.010665894, 0.028045654, -0.010063171, 0.021972656, -0.0061073303, -0.01109314, -0.034698486, 0.014022827, -0.009231567, 0.0...
http://arxiv.org/abs/2308.16118v1
Response: Emergent analogical reasoning in large language models
Damian Hodel, Jevin West
cs.CL, cs.AI
In their recent Nature Human Behaviour paper, "Emergent analogical reasoning in large language models," (Webb, Holyoak, and Lu, 2023) the authors argue that "large language models such as GPT-3 have acquired an emergent ability to find zero-shot solutions to a broad range of analogy problems." In this response, we prov...
http://arxiv.org/pdf/2308.16118v1
2023-08-30T16:17:26
2023-08-30T16:17:26
{ "title": [ -0.020812988, -0.021972656, -0.0004785061, -0.043304443, -0.023422241, 0.003835678, -0.036743164, -0.006614685, 0.014137268, 0.0309906, -0.010757446, -0.003326416, 0.017120361, -0.017745972, -0.04006958, 0.01713562, -0.009437561, 0.0...
http://arxiv.org/abs/2308.16109v1
Grandma Karl is 27 years old -- research agenda for pseudonymization of research data
Elena Volodina, Simon Dobnik, Therese Lindström Tiedemann, Xuan-Son Vu
cs.CL, cs.AI
Accessibility of research data is critical for advances in many research fields, but textual data often cannot be shared due to the personal and sensitive information which it contains, e.g names or political opinions. General Data Protection Regulation (GDPR) suggests pseudonymization as a solution to secure open acce...
http://arxiv.org/pdf/2308.16109v1
2023-08-30T16:04:54
2023-08-30T16:04:54
{ "title": [ -0.030776978, -0.007637024, -0.0791626, 0.000029742718, -0.009605408, -0.028137207, -0.009727478, 0.014930725, -0.058563232, 0.047546387, -0.0012960434, -0.016662598, 0.014968872, -0.027664185, -0.006160736, -0.047973633, 0.010536194, ...
http://arxiv.org/abs/2309.08614v1
Analyzing Character and Consciousness in AI-Generated Social Content: A Case Study of Chirper, the AI Social Network
Jianwei Luo
cs.AI, cs.CL, cs.SI, 68T01
This paper delves into an intricate analysis of the character and consciousness of AI entities, with a particular focus on Chirpers within the AI social network. At the forefront of this research is the introduction of novel testing methodologies, including the Influence index and Struggle Index Test, which offers a fr...
http://arxiv.org/pdf/2309.08614v1
2023-08-30T15:40:18
2023-08-30T15:40:18
{ "title": [ -0.009895325, -0.011329651, -0.032836914, -0.0036582947, 0.0052375793, -0.026000977, 0.029510498, -0.008529663, -0.006511688, 0.027145386, -0.040222168, 0.0049743652, 0.017715454, -0.014099121, -0.028305054, -0.022369385, -0.0362854, ...
http://arxiv.org/abs/2308.16075v1
Impact of Visual Context on Noisy Multimodal NMT: An Empirical Study for English to Indian Languages
Baban Gain, Dibyanayan Bandyopadhyay, Samrat Mukherjee, Chandranath Adak, Asif Ekbal
cs.CL, cs.AI, cs.CV
The study investigates the effectiveness of utilizing multimodal information in Neural Machine Translation (NMT). While prior research focused on using multimodal data in low-resource scenarios, this study examines how image features impact translation when added to a large-scale, pre-trained unimodal NMT system. Surpr...
http://arxiv.org/pdf/2308.16075v1
2023-08-30T14:52:14
2023-08-30T14:52:14
{ "title": [ 0.0129776, 0.009521484, 0.007144928, -0.039978027, -0.0017852783, 0.0061454773, -0.028427124, -0.022109985, -0.026062012, 0.037384033, -0.013473511, 0.029312134, 0.013458252, -0.037506104, 0.0009384155, 0.0015039444, -0.024673462, 0....
http://arxiv.org/abs/2308.16061v1
Conti Inc.: Understanding the Internal Discussions of a large Ransomware-as-a-Service Operator with Machine Learning
Estelle Ruellan, Masarah Paquet-Clouston, Sebastian Garcia
cs.CR, cs.CL, cs.LG
Ransomware-as-a-service (RaaS) is increasing the scale and complexity of ransomware attacks. Understanding the internal operations behind RaaS has been a challenge due to the illegality of such activities. The recent chat leak of the Conti RaaS operator, one of the most infamous ransomware operators on the internationa...
http://arxiv.org/pdf/2308.16061v1
2023-08-30T14:36:25
2023-08-30T14:36:25
{ "title": [ -0.040008545, -0.037994385, -0.027359009, 0.014572144, -0.053344727, -0.029724121, 0.0062294006, 0.024108887, -0.022735596, 0.0039405823, -0.046142578, -0.028915405, 0.01965332, -0.04324341, 0.016067505, -0.0713501, 0.0070381165, 0.0...
http://arxiv.org/abs/2308.16060v1
Text-to-OverpassQL: A Natural Language Interface for Complex Geodata Querying of OpenStreetMap
Michael Staniek, Raphael Schumann, Maike Züfle, Stefan Riezler
cs.CL, cs.AI, cs.CY, cs.DB, cs.HC
We present Text-to-OverpassQL, a task designed to facilitate a natural language interface for querying geodata from OpenStreetMap (OSM). The Overpass Query Language (OverpassQL) allows users to formulate complex database queries and is widely adopted in the OSM ecosystem. Generating Overpass queries from natural langua...
http://arxiv.org/pdf/2308.16060v1
2023-08-30T14:33:25
2023-08-30T14:33:25
{ "title": [ 0.008224487, -0.019363403, -0.012901306, -0.049713135, 0.015296936, 0.009544373, -0.01184082, -0.011856079, 0.012283325, 0.030090332, -0.04727173, -0.009735107, 0.003545761, -0.012519836, -0.041931152, -0.035491943, -0.0048446655, 0....
http://arxiv.org/abs/2308.16055v1
AsyncET: Asynchronous Learning for Knowledge Graph Entity Typing with Auxiliary Relations
Yun-Cheng Wang, Xiou Ge, Bin Wang, C. -C. Jay Kuo
cs.CL, cs.AI
Knowledge graph entity typing (KGET) is a task to predict the missing entity types in knowledge graphs (KG). Previously, KG embedding (KGE) methods tried to solve the KGET task by introducing an auxiliary relation, 'hasType', to model the relationship between entities and their types. However, a single auxiliary relati...
http://arxiv.org/pdf/2308.16055v1
2023-08-30T14:24:16
2023-08-30T14:24:16
{ "title": [ -0.038970947, -0.013298035, -0.055847168, -0.07543945, -0.0044174194, 0.009666443, -0.0637207, 0.026397705, 0.008094788, 0.030670166, -0.024749756, 0.012374878, 0.004951477, -0.017120361, -0.021987915, -0.0042915344, -0.007873535, 0....
http://arxiv.org/abs/2308.15987v1
FPTQ: Fine-grained Post-Training Quantization for Large Language Models
Qingyuan Li, Yifan Zhang, Liang Li, Peng Yao, Bo Zhang, Xiangxiang Chu, Yerui Sun, Li Du, Yuchen Xie
cs.CL, cs.AI, cs.LG
In the era of large-scale language models, the substantial parameter size poses significant challenges for deployment. Being a prevalent compression technique, quantization has emerged as the mainstream practice to tackle this issue, which is mainly centered on two recipes W8A8 and W4A16 (i.e. weights and activations i...
http://arxiv.org/pdf/2308.15987v1
2023-08-30T12:18:18
2023-08-30T12:18:18
{ "title": [ -0.027496338, -0.0054359436, -0.042785645, -0.05883789, 0.01020813, -0.0030021667, -0.052124023, 0.0033473969, 0.0016517639, 0.046142578, -0.002401352, -0.000091314316, 0.015808105, -0.020019531, -0.050872803, 0.014457703, -0.007686615, ...
http://arxiv.org/abs/2308.15982v1
MerA: Merging Pretrained Adapters For Few-Shot Learning
Shwai He, Run-Ze Fan, Liang Ding, Li Shen, Tianyi Zhou, Dacheng Tao
cs.CL
Adapter tuning, which updates only a few parameters, has become a mainstream method for fine-tuning pretrained language models to downstream tasks. However, it often yields subpar results in few-shot learning. AdapterFusion, which assembles pretrained adapters using composition layers tailored to specific tasks, is a p...
http://arxiv.org/pdf/2308.15982v1
2023-08-30T12:10:17
2023-08-30T12:10:17
{ "title": [ 0.011077881, -0.016571045, -0.021011353, -0.0423584, -0.0019397736, -0.0012731552, -0.015419006, -0.009155273, 0.008460999, 0.014305115, -0.025268555, 0.0008883476, -0.02532959, -0.017410278, -0.033233643, -0.010734558, 0.018127441, ...
http://arxiv.org/abs/2308.15961v1
Finding-Aware Anatomical Tokens for Chest X-Ray Automated Reporting
Francesco Dalla Serra, Chaoyang Wang, Fani Deligianni, Jeffrey Dalton, Alison Q. O'Neil
cs.CV, cs.CL
The task of radiology reporting comprises describing and interpreting the medical findings in radiographic images, including description of their location and appearance. Automated approaches to radiology reporting require the image to be encoded into a suitable token representation for input to the language model. Pre...
http://arxiv.org/pdf/2308.15961v1
2023-08-30T11:35:21
2023-08-30T11:35:21
{ "title": [ -0.011528015, 0.005176544, -0.03970337, -0.021209717, -0.008575439, -0.008850098, -0.015007019, -0.02949524, -0.0076560974, 0.0368042, -0.03363037, 0.004623413, 0.020523071, -0.020080566, -0.040863037, 0.008102417, 0.01309967, -0.018...
http://arxiv.org/abs/2308.15952v1
Benchmarking Multilabel Topic Classification in the Kyrgyz Language
Anton Alekseev, Sergey I. Nikolenko, Gulnara Kabaeva
cs.CL
Kyrgyz is a very underrepresented language in terms of modern natural language processing resources. In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news cl...
http://arxiv.org/pdf/2308.15952v1
2023-08-30T11:02:26
2023-08-30T11:02:26
{ "title": [ -0.01751709, 0.01361084, -0.043121338, -0.034301758, -0.0015516281, -0.005504608, -0.07116699, -0.026123047, 0.017532349, 0.044067383, -0.0046195984, 0.022323608, 0.011383057, -0.020080566, -0.03579712, 0.009864807, -0.023757935, 0.0...
http://arxiv.org/abs/2308.15930v3
LLaSM: Large Language and Speech Model
Yu Shu, Siwei Dong, Guangyao Chen, Wenhao Huang, Ruihua Zhang, Daochen Shi, Qiqi Xiang, Yemin Shi
cs.CL, cs.LG, cs.SD, eess.AS
Multi-modal large language models have garnered significant interest recently. Though, most of the works focus on vision-language multi-modal models providing strong capabilities in following vision-and-language instructions. However, we claim that speech is also an important modality through which humans interact with...
http://arxiv.org/pdf/2308.15930v3
2023-09-16T06:14:54
2023-08-30T10:12:39
{ "title": [ -0.028640747, -0.010383606, -0.055389404, -0.07940674, 0.021072388, 0.0068969727, -0.064331055, 0.009239197, 0.00051784515, 0.037261963, -0.0013828278, -0.0020065308, 0.02230835, -0.029769897, -0.029052734, 0.021087646, -0.009925842, ...
http://arxiv.org/abs/2308.15906v3
Is the U.S. Legal System Ready for AI's Challenges to Human Values?
Inyoung Cheong, Aylin Caliskan, Tadayoshi Kohno
cs.CY, cs.AI, cs.CL
Our interdisciplinary study investigates how effectively U.S. laws confront the challenges posed by Generative AI to human values. Through an analysis of diverse hypothetical scenarios crafted during an expert workshop, we have identified notable gaps and uncertainties within the existing legal framework regarding the ...
http://arxiv.org/pdf/2308.15906v3
2023-09-05T01:01:58
2023-08-30T09:19:06
{ "title": [ 0.010765076, -0.021820068, 0.0012893677, 0.021697998, -0.020584106, -0.011306763, 0.045928955, -0.016036987, -0.026748657, 0.0345459, -0.058685303, -0.0056152344, 0.017242432, -0.020889282, -0.006061554, -0.04269409, -0.0020503998, 0...
http://arxiv.org/abs/2308.15885v1
Towards One-Shot Learning for Text Classification using Inductive Logic Programming
Ghazal Afroozi Milani, Daniel Cyrus, Alireza Tamaddoni-Nezhad
cs.LG, cs.CL, cs.LO
With the ever-increasing potential of AI to perform personalised tasks, it is becoming essential to develop new machine learning techniques which are data-efficient and do not require hundreds or thousands of training data. In this paper, we explore an Inductive Logic Programming approach for one-shot text classificati...
http://arxiv.org/pdf/2308.15885v1
2023-08-30T09:04:06
2023-08-30T09:04:06
{ "title": [ -0.03439331, -0.020874023, -0.031951904, -0.044189453, -0.024368286, 0.0074043274, -0.034057617, 0.016815186, -0.0023155212, 0.006427765, -0.041656494, 0.03704834, -0.041656494, -0.0051574707, -0.02784729, -0.011795044, -0.021652222, ...
http://arxiv.org/abs/2308.15813v1
Knowledge-grounded Natural Language Recommendation Explanation
Anthony Colas, Jun Araki, Zhengyu Zhou, Bingqing Wang, Zhe Feng
cs.CL, cs.IR
Explanations accompanied by a recommendation can assist users in understanding the decision made by recommendation systems, which in turn increases a user's confidence and trust in the system. Recently, research has focused on generating natural language explanations in a human-readable format. Thus far, the proposed a...
http://arxiv.org/pdf/2308.15813v1
2023-08-30T07:36:12
2023-08-30T07:36:12
{ "title": [ -0.01600647, -0.01222229, -0.03805542, -0.06311035, -0.014816284, -0.0118255615, -0.026733398, 0.014030457, -0.01776123, 0.017608643, -0.028015137, 0.018432617, -0.0025558472, -0.022445679, -0.0054512024, -0.0030174255, -0.010482788, ...
http://arxiv.org/abs/2308.15812v1
Peering Through Preferences: Unraveling Feedback Acquisition for Aligning Large Language Models
Hritik Bansal, John Dang, Aditya Grover
cs.LG, cs.AI, cs.CL
Aligning large language models (LLMs) with human values and intents critically involves the use of human or AI feedback. While dense feedback annotations are expensive to acquire and integrate, sparse feedback presents a structural design choice between ratings (e.g., score Response A on a scale of 1-7) and rankings (e...
http://arxiv.org/pdf/2308.15812v1
2023-08-30T07:35:32
2023-08-30T07:35:32
{ "title": [ -0.032318115, -0.0025978088, -0.037994385, -0.0496521, -0.0028152466, 0.0066566467, -0.056274414, -0.0075149536, 0.012290955, 0.02619934, -0.023803711, 0.0019187927, -0.0013570786, -0.006187439, -0.051757812, 0.0036716461, -0.0054473877,...
http://arxiv.org/abs/2308.15793v1
HAlf-MAsked Model for Named Entity Sentiment analysis
Anton Kabaev, Pavel Podberezko, Andrey Kaznacheev, Sabina Abdullayeva
cs.CL, cs.LG
Named Entity Sentiment analysis (NESA) is one of the most actively developing application domains in Natural Language Processing (NLP). Social media NESA is a significant field of opinion analysis since detecting and tracking sentiment trends in the news flow is crucial for building various analytical systems and monit...
http://arxiv.org/pdf/2308.15793v1
2023-08-30T06:53:24
2023-08-30T06:53:24
{ "title": [ -0.011627197, 0.0012702942, -0.042999268, -0.048217773, 0.015159607, -0.0070877075, -0.018936157, 0.0008087158, 0.040222168, 0.03137207, -0.033813477, -0.009559631, 0.032562256, -0.006462097, -0.05618286, 0.011665344, -0.01146698, 0....
http://arxiv.org/abs/2308.15772v3
Task-Based MoE for Multitask Multilingual Machine Translation
Hai Pham, Young Jin Kim, Subhabrata Mukherjee, David P. Woodruff, Barnabas Poczos, Hany Hassan Awadalla
cs.CL
Mixture-of-experts (MoE) architecture has been proven a powerful method for diverse tasks in training deep models in many applications. However, current MoE implementations are task agnostic, treating all tokens from different tasks in the same manner. In this work, we instead design a novel method that incorporates ta...
http://arxiv.org/pdf/2308.15772v3
2023-10-24T22:59:26
2023-08-30T05:41:29
{ "title": [ -0.043395996, -0.010246277, -0.03894043, -0.06500244, 0.0020122528, 0.0129776, -0.053894043, 0.027923584, -0.0077781677, 0.059936523, 0.0066833496, 0.0060272217, 0.03439331, -0.022445679, -0.038513184, 0.023666382, -0.013885498, 0.02...
http://arxiv.org/abs/2308.15745v1
Cyberbullying Detection for Low-resource Languages and Dialects: Review of the State of the Art
Tanjim Mahmud, Michal Ptaszynski, Juuso Eronen, Fumito Masui
cs.CL
The struggle of social media platforms to moderate content in a timely manner, encourages users to abuse such platforms to spread vulgar or abusive language, which, when performed repeatedly becomes cyberbullying a social problem taking place in virtual environments, yet with real-world consequences, such as depression...
http://arxiv.org/pdf/2308.15745v1
2023-08-30T03:52:28
2023-08-30T03:52:28
{ "title": [ 0.0041923523, -0.0075531006, -0.0076065063, -0.04522705, -0.0059547424, 0.0012111664, -0.029815674, -0.0036811829, -0.010368347, 0.04107666, -0.010177612, 0.013191223, 0.025756836, -0.02067566, -0.029724121, 0.005039215, -0.026641846, ...
http://arxiv.org/abs/2308.15727v2
Quantifying and Analyzing Entity-level Memorization in Large Language Models
Zhenhong Zhou, Jiuyang Xiang, Chaomeng Chen, Sen Su
cs.CL
Large language models (LLMs) have been proven capable of memorizing their training data, which can be extracted through specifically designed prompts. As the scale of datasets continues to grow, privacy risks arising from memorization have attracted increasing attention. Quantifying language model memorization helps ev...
http://arxiv.org/pdf/2308.15727v2
2023-11-05T13:44:55
2023-08-30T03:06:47
{ "title": [ -0.019378662, -0.0067710876, -0.036865234, -0.06384277, 0.003932953, -0.01084137, -0.033294678, 0.012588501, 0.004749298, 0.046875, -0.015571594, 0.021606445, 0.022216797, -0.0065231323, -0.049621582, 0.018234253, -0.01550293, 0.0213...
http://arxiv.org/abs/2308.15711v1
Optimizing Factual Accuracy in Text Generation through Dynamic Knowledge Selection
Hongjin Qian, Zhicheng Dou, Jiejun Tan, Haonan Chen, Haoqi Gu, Ruofei Lai, Xinyu Zhang, Zhao Cao, Ji-Rong Wen
cs.CL, cs.AI
Language models (LMs) have revolutionized the way we interact with information, but they often generate nonfactual text, raising concerns about their reliability. Previous methods use external knowledge as references for text generation to enhance factuality but often struggle with the knowledge mix-up(e.g., entity mis...
http://arxiv.org/pdf/2308.15711v1
2023-08-30T02:22:40
2023-08-30T02:22:40
{ "title": [ -0.038208008, -0.01285553, -0.058532715, -0.048950195, -0.009643555, -0.004863739, -0.040893555, 0.0047454834, -0.010261536, 0.044433594, -0.0063552856, 0.03060913, 0.008232117, -0.00541687, -0.050598145, 0.011253357, -0.014930725, 0...
http://arxiv.org/abs/2309.08613v1
Multimodal Recommender Systems in the Prediction of Disease Comorbidity
Aashish Cheruvu
cs.IR, cs.AI, cs.CL, cs.LG
While deep-learning based recommender systems utilizing collaborative filtering have been commonly used for recommendation in other domains, their application in the medical domain have been limited. In addition to modeling user-item interactions, we show that deep-learning based recommender systems can be used to mode...
http://arxiv.org/pdf/2309.08613v1
2023-08-30T01:40:45
2023-08-30T01:40:45
{ "title": [ 0.016311646, 0.019546509, -0.049865723, -0.10192871, 0.027389526, -0.00058174133, -0.048553467, -0.0036029816, -0.01626587, 0.016815186, -0.0042266846, 0.013877869, -0.028457642, -0.029373169, -0.021636963, 0.0038337708, -0.010108948, ...
http://arxiv.org/abs/2309.00642v1
Extracting Mathematical Concepts with Large Language Models
Valeria de Paiva, Qiyue Gao, Pavel Kovalev, Lawrence S. Moss
cs.CL, 68T50, I.2.7
We extract mathematical concepts from mathematical text using generative large language models (LLMs) like ChatGPT, contributing to the field of automatic term extraction (ATE) and mathematical text processing, and also to the study of LLMs themselves. Our work builds on that of others in that we aim for automatic extr...
http://arxiv.org/pdf/2309.00642v1
2023-08-29T20:54:50
2023-08-29T20:54:50
{ "title": [ -0.012588501, -0.0013589859, -0.02748108, -0.048706055, 0.0026683807, -0.0031147003, -0.031463623, -0.014152527, 0.03793335, 0.037261963, -0.004322052, -0.0046920776, 0.008392334, -0.007987976, -0.050628662, 0.029190063, -0.006111145, ...
http://arxiv.org/abs/2309.06419v1
Radiology-Llama2: Best-in-Class Large Language Model for Radiology
Zhengliang Liu, Yiwei Li, Peng Shu, Aoxiao Zhong, Longtao Yang, Chao Ju, Zihao Wu, Chong Ma, Jie Luo, Cheng Chen, Sekeun Kim, Jiang Hu, Haixing Dai, Lin Zhao, Dajiang Zhu, Jun Liu, Wei Liu, Dinggang Shen, Tianming Liu, Quanzheng Li, Xiang Li
cs.CL
This paper introduces Radiology-Llama2, a large language model specialized for radiology through a process known as instruction tuning. Radiology-Llama2 is based on the Llama2 architecture and further trained on a large dataset of radiology reports to generate coherent and clinically useful impressions from radiologica...
http://arxiv.org/pdf/2309.06419v1
2023-08-29T17:44:28
2023-08-29T17:44:28
{ "title": [ -0.038391113, -0.00082731247, -0.038208008, -0.06021118, 0.0132369995, 0.016479492, -0.055755615, -0.0060577393, 0.00012457371, 0.028076172, 0.006252289, -0.002840042, 0.0007266998, -0.04196167, -0.026901245, 0.01612854, 0.0066566467, ...
http://arxiv.org/abs/2308.15459v2
ParaGuide: Guided Diffusion Paraphrasers for Plug-and-Play Textual Style Transfer
Zachary Horvitz, Ajay Patel, Chris Callison-Burch, Zhou Yu, Kathleen McKeown
cs.CL, cs.AI
Textual style transfer is the task of transforming stylistic properties of text while preserving meaning. Target "styles" can be defined in numerous ways, ranging from single attributes (e.g, formality) to authorship (e.g, Shakespeare). Previous unsupervised style-transfer approaches generally rely on significant amoun...
http://arxiv.org/pdf/2308.15459v2
2023-09-04T20:34:03
2023-08-29T17:36:02
{ "title": [ -0.021331787, -0.016311646, -0.03353882, -0.048431396, -0.012489319, 0.008018494, -0.03591919, -0.0021266937, -0.030181885, 0.052856445, -0.014640808, 0.04144287, 0.037384033, -0.015602112, 0.0002322197, 0.01789856, -0.0064926147, 0....
http://arxiv.org/abs/2308.15452v4
When Do Program-of-Thoughts Work for Reasoning?
Zhen Bi, Ningyu Zhang, Yinuo Jiang, Shumin Deng, Guozhou Zheng, Huajun Chen
cs.CL, cs.AI, cs.LG, cs.SE
In the realm of embodied artificial intelligence, the reasoning capabilities of Large Language Models (LLMs) play a pivotal role. Although there are effective methods like program-of-thought prompting for LLMs which uses programming language to tackle complex reasoning tasks, the specific impact of code data on the imp...
http://arxiv.org/pdf/2308.15452v4
2023-11-15T14:06:30
2023-08-29T17:22:39
{ "title": [ 0.0010538101, -0.041992188, 0.0335083, -0.004585266, -0.06439209, 0.009208679, -0.0007915497, -0.009643555, -0.017227173, 0.012557983, -0.03149414, -0.023742676, 0.013626099, -0.03466797, 0.015052795, -0.026290894, 0.011077881, -0.01...
http://arxiv.org/abs/2308.15448v1
Vulgar Remarks Detection in Chittagonian Dialect of Bangla
Tanjim Mahmud, Michal Ptaszynski, Fumito Masui
cs.CL
The negative effects of online bullying and harassment are increasing with Internet popularity, especially in social media. One solution is using natural language processing (NLP) and machine learning (ML) methods for the automatic detection of harmful remarks, but these methods are limited in low-resource languages li...
http://arxiv.org/pdf/2308.15448v1
2023-08-29T17:19:32
2023-08-29T17:19:32
{ "title": [ 0.009727478, -0.0010690689, -0.037139893, -0.044311523, -0.0011053085, -0.031311035, -0.022720337, 0.006965637, 0.018310547, 0.047607422, -0.032440186, -0.034210205, 0.042907715, -0.032287598, -0.05596924, -0.014976501, -0.02017212, ...
http://arxiv.org/abs/2308.15517v1
Document AI: A Comparative Study of Transformer-Based, Graph-Based Models, and Convolutional Neural Networks For Document Layout Analysis
Sotirios Kastanas, Shaomu Tan, Yi He
cs.CL, cs.CV
Document AI aims to automatically analyze documents by leveraging natural language processing and computer vision techniques. One of the major tasks of Document AI is document layout analysis, which structures document pages by interpreting the content and spatial relationships of layout, image, and text. This task can...
http://arxiv.org/pdf/2308.15517v1
2023-08-29T16:58:03
2023-08-29T16:58:03
{ "title": [ -0.06427002, -0.020996094, -0.052093506, -0.026412964, -0.027359009, -0.014099121, -0.0625, -0.0025959015, 0.018920898, 0.039001465, -0.053710938, 0.007575989, 0.04309082, -0.007888794, -0.010681152, 0.0030727386, 0.004501343, 0.0231...
http://arxiv.org/abs/2308.15419v1
Characterizing Learning Curves During Language Model Pre-Training: Learning, Forgetting, and Stability
Tyler A. Chang, Zhuowen Tu, Benjamin K. Bergen
cs.CL
How do language models learn to make predictions during pre-training? To study this question, we extract learning curves from five autoregressive English language model pre-training runs, for 1M tokens in context. We observe that the language models generate short repetitive phrases before learning to generate longer a...
http://arxiv.org/pdf/2308.15419v1
2023-08-29T16:24:09
2023-08-29T16:24:09
{ "title": [ -0.026397705, 0.009262085, -0.019943237, -0.03781128, -0.01676941, -0.00033521652, -0.052124023, -0.01737976, -0.03024292, 0.02218628, -0.03265381, 0.030670166, -0.011123657, -0.014533997, -0.024520874, -0.01725769, -0.0077323914, 0....
http://arxiv.org/abs/2308.15399v1
Rethinking Machine Ethics -- Can LLMs Perform Moral Reasoning through the Lens of Moral Theories?
Jingyan Zhou, Minda Hu, Junan Li, Xiaoying Zhang, Xixin Wu, Irwin King, Helen Meng
cs.CL
Making moral judgments is an essential step toward developing ethical AI systems. Prevalent approaches are mostly implemented in a bottom-up manner, which uses a large set of annotated data to train models based on crowd-sourced opinions about morality. These approaches have been criticized for potentially overgenerali...
http://arxiv.org/pdf/2308.15399v1
2023-08-29T15:57:32
2023-08-29T15:57:32
{ "title": [ -0.034423828, -0.008171082, -0.027557373, 0.016296387, -0.03161621, -0.027633667, 0.004508972, -0.01638794, -0.033172607, 0.01486969, -0.055389404, 0.023757935, -0.015487671, -0.022369385, -0.008430481, -0.048950195, -0.016143799, 0....
http://arxiv.org/abs/2308.15363v3
Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation
Dawei Gao, Haibin Wang, Yaliang Li, Xiuyu Sun, Yichen Qian, Bolin Ding, Jingren Zhou
cs.DB, cs.CL, cs.LG
Large language models (LLMs) have emerged as a new paradigm for Text-to-SQL task. However, the absence of a systematical benchmark inhibits the development of designing effective, efficient and economic LLM-based Text-to-SQL solutions. To address this challenge, in this paper, we first conduct a systematical and extens...
http://arxiv.org/pdf/2308.15363v3
2023-09-20T04:16:31
2023-08-29T14:59:54
{ "title": [ -0.026000977, -0.015113831, -0.041381836, -0.08178711, 0.014923096, 0.0107421875, -0.051513672, 0.022140503, 0.031143188, 0.04611206, -0.023712158, -0.0010633469, 0.01259613, -0.000202775, -0.06982422, 0.0041236877, -0.00422287, 0.02...
http://arxiv.org/abs/2308.15352v1
Historical patterns of rice farming explain modern-day language use in China and Japan more than modernization and urbanization
Sharath Chandra Guntuku, Thomas Talhelm, Garrick Sherman, Angel Fan, Salvatore Giorgi, Liuqing Wei, Lyle H. Ungar
cs.CL, cs.SI, physics.soc-ph
We used natural language processing to analyze a billion words to study cultural differences on Weibo, one of China's largest social media platforms. We compared predictions from two common explanations about cultural differences in China (economic development and urban-rural differences) against the less-obvious legac...
http://arxiv.org/pdf/2308.15352v1
2023-08-29T14:47:08
2023-08-29T14:47:08
{ "title": [ 0.030899048, 0.0026226044, -0.025253296, 0.0063667297, 0.013801575, -0.03741455, 0.021469116, -0.020645142, -0.01763916, -0.002506256, -0.0102005005, -0.0134887695, -0.029785156, -0.04550171, 0.012191772, -0.03390503, -0.022216797, 0...
http://arxiv.org/abs/2308.15334v1
A Framework for Responsible Development of Automated Student Feedback with Generative AI
Euan D Lindsay, Aditya Johri, Johannes Bjerva
cs.CY, cs.AI, cs.CL
Providing rich feedback to students is essential for supporting student learning. Recent advances in generative AI, particularly within large language modelling (LLM), provide the opportunity to deliver repeatable, scalable and instant automatically generated feedback to students, making abundant a previously scarce an...
http://arxiv.org/pdf/2308.15334v1
2023-08-29T14:29:57
2023-08-29T14:29:57
{ "title": [ -0.0076179504, -0.0061416626, -0.014541626, -0.040008545, -0.011123657, -0.006843567, -0.01524353, -0.030670166, 0.010406494, 0.032165527, -0.031311035, 0.0012722015, -0.005744934, -0.012481689, -0.06323242, -0.016921997, -0.020446777, ...
http://arxiv.org/abs/2309.12981v2
Wordification: A New Way of Teaching English Spelling Patterns
Lexington Whalen, Nathan Bickel, Shash Comandur, Dalton Craven, Stanley Dubinsky, Homayoun Valafar
cs.OH, cs.CL, K.3
Literacy, or the ability to read and write, is a crucial indicator of success in life and greater society. It is estimated that 85% of people in juvenile delinquent systems cannot adequately read or write, that more than half of those with substance abuse issues have complications in reading or writing and that two-thi...
http://arxiv.org/pdf/2309.12981v2
2023-11-10T10:49:12
2023-08-29T14:14:01
{ "title": [ -0.016906738, -0.0060310364, -0.035705566, 0.00010484457, -0.013999939, -0.02330017, 0.012161255, -0.015457153, -0.032928467, 0.03945923, -0.016464233, 0.01939392, 0.019348145, -0.020980835, 0.004425049, -0.009857178, -0.017578125, 0...
http://arxiv.org/abs/2308.15299v1
TaskLAMA: Probing the Complex Task Understanding of Language Models
Quan Yuan, Mehran Kazemi, Xin Xu, Isaac Noble, Vaiva Imbrasaite, Deepak Ramachandran
cs.CL
Structured Complex Task Decomposition (SCTD) is the problem of breaking down a complex real-world task (such as planning a wedding) into a directed acyclic graph over individual steps that contribute to achieving the task, with edges specifying temporal dependencies between them. SCTD is an important component of assis...
http://arxiv.org/pdf/2308.15299v1
2023-08-29T13:36:45
2023-08-29T13:36:45
{ "title": [ -0.0413208, -0.012741089, -0.0435791, -0.062561035, -0.0057411194, 0.0008459091, -0.067993164, 0.008201599, 0.008621216, 0.023284912, -0.0054473877, -0.0034332275, 0.011230469, -0.019851685, -0.044525146, 0.015060425, -0.013694763, 0...
http://arxiv.org/abs/2308.15298v1
KGConv, a Conversational Corpus grounded in Wikidata
Quentin Brabant, Gwenole Lecorve, Lina M. Rojas-Barahona, Claire Gardent
cs.CL, cs.AI
We present KGConv, a large, conversational corpus of 71k conversations where each question-answer pair is grounded in a Wikidata fact. Conversations contain on average 8.6 questions and for each Wikidata fact, we provide multiple variants (12 on average) of the corresponding question using templates, human annotations,...
http://arxiv.org/pdf/2308.15298v1
2023-08-29T13:35:51
2023-08-29T13:35:51
{ "title": [ -0.043029785, -0.033691406, -0.064575195, -0.01537323, -0.06549072, -0.0154418945, -0.068237305, -0.00014805794, 0.06567383, 0.043029785, -0.05731201, -0.02583313, 0.07702637, -0.025512695, -0.040527344, -0.0056037903, 0.0041656494, ...
http://arxiv.org/abs/2308.15262v1
Enhancing OCR Performance through Post-OCR Models: Adopting Glyph Embedding for Improved Correction
Yung-Hsin Chen, Yuli Zhou
cs.CV, cs.CL
The study investigates the potential of post-OCR models to overcome limitations in OCR models and explores the impact of incorporating glyph embedding on post-OCR correction performance. In this study, we have developed our own post-OCR correction model. The novelty of our approach lies in embedding the OCR output usin...
http://arxiv.org/pdf/2308.15262v1
2023-08-29T12:41:50
2023-08-29T12:41:50
{ "title": [ 0.005077362, 0.008331299, -0.00075149536, -0.01940918, -0.0049324036, -0.009513855, -0.016983032, -0.034240723, -0.00013577938, 0.05722046, -0.029815674, 0.014335632, 0.04309082, -0.0045547485, -0.038146973, 0.0041389465, -0.0002937317, ...
http://arxiv.org/abs/2308.15246v1
A Classification-Guided Approach for Adversarial Attacks against Neural Machine Translation
Sahar Sadrizadeh, Ljiljana Dolamic, Pascal Frossard
cs.CL
Neural Machine Translation (NMT) models have been shown to be vulnerable to adversarial attacks, wherein carefully crafted perturbations of the input can mislead the target model. In this paper, we introduce ACT, a novel adversarial attack framework against NMT systems guided by a classifier. In our attack, the adversa...
http://arxiv.org/pdf/2308.15246v1
2023-08-29T12:12:53
2023-08-29T12:12:53
{ "title": [ -0.012626648, -0.0017528534, -0.006629944, -0.03289795, 0.018035889, -0.004486084, 0.003156662, -0.011161804, -0.05340576, 0.03945923, 0.025970459, 0.013839722, -0.015609741, -0.027511597, -0.035186768, -0.004386902, -0.009002686, 0....
http://arxiv.org/abs/2308.15235v1
PronounFlow: A Hybrid Approach for Calibrating Pronouns in Sentences
Nicos Isaak
cs.CL, cs.AI, I.2.0; I.2.3; I.2.7; I.5.1; I.5.4
Flip through any book or listen to any song lyrics, and you will come across pronouns that, in certain cases, can hinder meaning comprehension, especially for machines. As the role of having cognitive machines becomes pervasive in our lives, numerous systems have been developed to resolve pronouns under various challen...
http://arxiv.org/pdf/2308.15235v1
2023-08-29T11:46:27
2023-08-29T11:46:27
{ "title": [ -0.031097412, -0.008232117, -0.037872314, -0.07055664, -0.0039482117, 0.012771606, -0.05444336, -0.005554199, 0.016479492, 0.03149414, -0.014312744, 0.0061416626, 0.0050811768, -0.010559082, -0.05029297, 0.00856781, -0.02935791, 0.05...
http://arxiv.org/abs/2308.15232v1
Classification-Aware Neural Topic Model Combined With Interpretable Analysis -- For Conflict Classification
Tianyu Liang, Yida Mu, Soonho Kim, Darline Larissa Kengne Kuate, Julie Lang, Rob Vos, Xingyi Song
cs.LG, cs.CL, cs.IR
A large number of conflict events are affecting the world all the time. In order to analyse such conflict events effectively, this paper presents a Classification-Aware Neural Topic Model (CANTM-IA) for Conflict Information Classification and Topic Discovery. The model provides a reliable interpretation of classificati...
http://arxiv.org/pdf/2308.15232v1
2023-08-29T11:40:24
2023-08-29T11:40:24
{ "title": [ 0.01737976, 0.026016235, -0.036865234, -0.058380127, 0.021392822, -0.016357422, -0.03842163, -0.026565552, 0.0023651123, 0.002746582, -0.021652222, -0.0023612976, -0.036224365, -0.02607727, -0.046875, -0.03475952, -0.017868042, 0.012...
http://arxiv.org/abs/2308.15231v1
Multi-party Goal Tracking with LLMs: Comparing Pre-training, Fine-tuning, and Prompt Engineering
Angus Addlesee, Weronika Sieińska, Nancie Gunson, Daniel Hernández Garcia, Christian Dondrup, Oliver Lemon
cs.CL, cs.HC
This paper evaluates the extent to which current Large Language Models (LLMs) can capture task-oriented multi-party conversations (MPCs). We have recorded and transcribed 29 MPCs between patients, their companions, and a social robot in a hospital. We then annotated this corpus for multi-party goal-tracking and intent-...
http://arxiv.org/pdf/2308.15231v1
2023-08-29T11:40:03
2023-08-29T11:40:03
{ "title": [ -0.02911377, -0.013725281, -0.034332275, -0.035491943, -0.032806396, -0.011405945, -0.05645752, -0.01322937, -0.059539795, -0.000702858, 0.015342712, 0.037750244, -0.04525757, -0.0625, 0.041046143, -0.015205383, 0.0042419434, 0.00809...
http://arxiv.org/abs/2308.15226v1
CLIPTrans: Transferring Visual Knowledge with Pre-trained Models for Multimodal Machine Translation
Devaansh Gupta, Siddhant Kharbanda, Jiawei Zhou, Wanhua Li, Hanspeter Pfister, Donglai Wei
cs.CV, cs.AI, cs.CL
There has been a growing interest in developing multimodal machine translation (MMT) systems that enhance neural machine translation (NMT) with visual knowledge. This problem setup involves using images as auxiliary information during training, and more recently, eliminating their use during inference. Towards this end...
http://arxiv.org/pdf/2308.15226v1
2023-08-29T11:29:43
2023-08-29T11:29:43
{ "title": [ -0.033996582, -0.011543274, -0.02470398, -0.053619385, -0.026687622, 0.012145996, -0.047821045, -0.002336502, 0.0061798096, 0.041656494, -0.011550903, 0.014945984, 0.017349243, -0.020355225, -0.03753662, 0.009880066, 0.014228821, 0.0...
http://arxiv.org/abs/2308.15214v2
FurChat: An Embodied Conversational Agent using LLMs, Combining Open and Closed-Domain Dialogue with Facial Expressions
Neeraj Cherakara, Finny Varghese, Sheena Shabana, Nivan Nelson, Abhiram Karukayil, Rohith Kulothungan, Mohammed Afil Farhan, Birthe Nesset, Meriam Moujahid, Tanvi Dinkar, Verena Rieser, Oliver Lemon
cs.CL, cs.AI, cs.HC, cs.RO
We demonstrate an embodied conversational agent that can function as a receptionist and generate a mixture of open and closed-domain dialogue along with facial expressions, by using a large language model (LLM) to develop an engaging conversation. We deployed the system onto a Furhat robot, which is highly expressive a...
http://arxiv.org/pdf/2308.15214v2
2023-08-30T13:13:19
2023-08-29T11:08:40
{ "title": [ -0.01411438, -0.00024044514, -0.0317688, 0.005458832, -0.033050537, -0.02986145, -0.023544312, -0.023651123, 0.04522705, 0.028747559, -0.048065186, 0.016815186, 0.034362793, -0.016174316, -0.010368347, -0.008644104, -0.015716553, 0.0...
http://arxiv.org/abs/2308.15209v1
Shared Lexical Items as Triggers of Code Switching
Shuly Wintner, Safaa Shehadi, Yuli Zeira, Doreen Osmelak, Yuval Nov
cs.CL, cs.AI
Why do bilingual speakers code-switch (mix their two languages)? Among the several theories that attempt to explain this natural and ubiquitous phenomenon, the Triggering Hypothesis relates code-switching to the presence of lexical triggers, specifically cognates and proper names, adjacent to the switch point. We provi...
http://arxiv.org/pdf/2308.15209v1
2023-08-29T10:55:44
2023-08-29T10:55:44
{ "title": [ -0.013191223, -0.0017595291, -0.01966858, -0.016616821, -0.02609253, -0.008842468, -0.009529114, -0.013587952, -0.019378662, 0.035339355, -0.017715454, -0.011154175, 0.021697998, -0.039611816, -0.010032654, 0.0023727417, 0.008232117, ...
http://arxiv.org/abs/2308.15202v1
Benchmarking the Generation of Fact Checking Explanations
Daniel Russo, Serra Sinem Tekiroglu, Marco Guerini
cs.CL
Fighting misinformation is a challenging, yet crucial, task. Despite the growing number of experts being involved in manual fact-checking, this activity is time-consuming and cannot keep up with the ever-increasing amount of Fake News produced daily. Hence, automating this process is necessary to help curb misinformati...
http://arxiv.org/pdf/2308.15202v1
2023-08-29T10:40:46
2023-08-29T10:40:46
{ "title": [ -0.002691269, -0.0009212494, -0.01927185, -0.086242676, -0.0073013306, 0.0121536255, -0.04711914, -0.013381958, -0.036010742, 0.033111572, 0.0058898926, 0.037322998, -0.029342651, -0.017227173, -0.034362793, -0.00081062317, -0.025619507,...
http://arxiv.org/abs/2308.15192v1
Enhancing Psychological Counseling with Large Language Model: A Multifaceted Decision-Support System for Non-Professionals
Guanghui Fu, Qing Zhao, Jianqiang Li, Dan Luo, Changwei Song, Wei Zhai, Shuo Liu, Fan Wang, Yan Wang, Lijuan Cheng, Juan Zhang, Bing Xiang Yang
cs.AI, cs.CL
In the contemporary landscape of social media, an alarming number of users express negative emotions, some of which manifest as strong suicidal intentions. This situation underscores a profound need for trained psychological counselors who can enact effective mental interventions. However, the development of these prof...
http://arxiv.org/pdf/2308.15192v1
2023-08-29T10:20:53
2023-08-29T10:20:53
{ "title": [ -0.023956299, -0.0149002075, -0.027557373, -0.053100586, -0.010604858, 0.0025672913, -0.05038452, -0.011871338, 0.002811432, 0.029891968, 0.000037372112, 0.00013875961, 0.003408432, -0.027069092, -0.029876709, 0.0020122528, -0.010147095,...
http://arxiv.org/abs/2308.15154v1
The Anatomy of Conspirators: Unveiling Traits using a Comprehensive Twitter Dataset
Margherita Gambini, Serena Tardelli, Maurizio Tesconi
cs.SI, cs.CL
The discourse around conspiracy theories is currently thriving amidst the rampant misinformation prevalent in online environments. Research in this field has been focused on detecting conspiracy theories on social media, often relying on limited datasets. In this study, we present a novel methodology for constructing a...
http://arxiv.org/pdf/2308.15154v1
2023-08-29T09:35:23
2023-08-29T09:35:23
{ "title": [ -0.011192322, -0.0008072853, -0.031097412, -0.046722412, 0.007972717, -0.0018758774, -0.009773254, 0.0007610321, -0.03692627, 0.014923096, 0.00466156, 0.021759033, -0.0039043427, -0.03012085, 0.0016441345, 0.0040397644, -0.009941101, ...
http://arxiv.org/abs/2308.15126v3
Evaluation and Analysis of Hallucination in Large Vision-Language Models
Junyang Wang, Yiyang Zhou, Guohai Xu, Pengcheng Shi, Chenlin Zhao, Haiyang Xu, Qinghao Ye, Ming Yan, Ji Zhang, Jihua Zhu, Jitao Sang, Haoyu Tang
cs.LG, cs.AI, cs.CL, cs.CV
Large Vision-Language Models (LVLMs) have recently achieved remarkable success. However, LVLMs are still plagued by the hallucination problem, which limits the practicality in many scenarios. Hallucination refers to the information of LVLMs' responses that does not exist in the visual input, which poses potential risks...
http://arxiv.org/pdf/2308.15126v3
2023-10-10T11:57:26
2023-08-29T08:51:24
{ "title": [ -0.007549286, 0.0073127747, -0.03262329, -0.033294678, -0.0023345947, -0.026367188, -0.020736694, -0.029647827, 0.02418518, 0.03543091, -0.0064926147, -0.014205933, 0.011642456, -0.022918701, -0.062347412, 0.0039367676, 0.0004556179, ...
http://arxiv.org/abs/2308.15122v3
SpikeBERT: A Language Spikformer Learned from BERT with Knowledge Distillation
Changze Lv, Tianlong Li, Jianhan Xu, Chenxi Gu, Zixuan Ling, Cenyuan Zhang, Xiaoqing Zheng, Xuanjing Huang
cs.CL
Spiking neural networks (SNNs) offer a promising avenue to implement deep neural networks in a more energy-efficient way. However, the network architectures of existing SNNs for language tasks are still simplistic and relatively shallow, and deep architectures have not been fully explored, resulting in a significant pe...
http://arxiv.org/pdf/2308.15122v3
2023-10-07T10:08:43
2023-08-29T08:41:16
{ "title": [ -0.010681152, -0.018341064, -0.070251465, -0.07019043, -0.030075073, -0.0033912659, -0.090026855, 0.02784729, 0.02104187, 0.037719727, -0.02456665, 0.008583069, 0.053649902, -0.02960205, -0.016540527, 0.020904541, 0.006576538, 0.0288...
http://arxiv.org/abs/2308.15118v1
Large Language Models on the Chessboard: A Study on ChatGPT's Formal Language Comprehension and Complex Reasoning Skills
Mu-Tien Kuo, Chih-Chung Hsueh, Richard Tzong-Han Tsai
cs.CL
While large language models have made strides in natural language processing, their proficiency in complex reasoning tasks requiring formal language comprehension, such as chess, remains less investigated. This paper probes the performance of ChatGPT, a sophisticated language model by OpenAI in tackling such complex re...
http://arxiv.org/pdf/2308.15118v1
2023-08-29T08:36:30
2023-08-29T08:36:30
{ "title": [ -0.023254395, -0.017837524, -0.009552002, -0.031677246, -0.024154663, 0.007167816, -0.036254883, -0.010948181, -0.02394104, 0.021697998, -0.017089844, 0.017242432, 0.013572693, -0.025909424, -0.002418518, 0.008346558, -0.038269043, 0...
http://arxiv.org/abs/2308.15097v1
Sequential annotations for naturally-occurring HRI: first insights
Lucien Tisserand, Frédéric Armetta, Heike Baldauf-Quilliatre, Antoine Bouquin, Salima Hassas, Mathieu Lefort
cs.AI, cs.CL
We explain the methodology we developed for improving the interactions accomplished by an embedded conversational agent, drawing from Conversation Analytic sequential and multimodal analysis. The use case is a Pepper robot that is expected to inform and orient users in a library. In order to propose and learn better in...
http://arxiv.org/pdf/2308.15097v1
2023-08-29T08:07:26
2023-08-29T08:07:26
{ "title": [ 0.00023436546, -0.012481689, 0.00061893463, -0.05886841, -0.025146484, 0.0028152466, -0.04537964, -0.018859863, -0.009216309, 0.037353516, -0.0019741058, -0.0038414001, -0.00063705444, -0.043884277, -0.04156494, 0.006603241, 0.01247406, ...
http://arxiv.org/abs/2308.15090v1
Killing two birds with one stone: Can an audio captioning system also be used for audio-text retrieval?
Etienne Labbé, Thomas Pellegrini, Julien Pinquier
cs.CL, cs.IR, cs.SD, eess.AS
Automated Audio Captioning (AAC) aims to develop systems capable of describing an audio recording using a textual sentence. In contrast, Audio-Text Retrieval (ATR) systems seek to find the best matching audio recording(s) for a given textual query (Text-to-Audio) or vice versa (Audio-to-Text). These tasks require diffe...
http://arxiv.org/pdf/2308.15090v1
2023-08-29T07:53:17
2023-08-29T07:53:17
{ "title": [ -0.015388489, -0.0016107559, -0.03894043, 0.0043678284, -0.029937744, -0.033569336, -0.0067863464, -0.017288208, 0.0028953552, 0.062805176, -0.026519775, 0.009460449, 0.040649414, -0.0082473755, -0.034942627, -0.0073242188, -0.021499634,...
http://arxiv.org/abs/2309.08612v1
Explaining Vision and Language through Graphs of Events in Space and Time
Mihai Masala, Nicolae Cudlenco, Traian Rebedea, Marius Leordeanu
cs.AI, cs.CL, cs.CV
Artificial Intelligence makes great advances today and starts to bridge the gap between vision and language. However, we are still far from understanding, explaining and controlling explicitly the visual content from a linguistic perspective, because we still lack a common explainable representation between the two dom...
http://arxiv.org/pdf/2309.08612v1
2023-08-29T07:25:06
2023-08-29T07:25:06
{ "title": [ 0.01499939, -0.028244019, -0.022369385, -0.06237793, -0.015853882, -0.028747559, -0.012077332, 0.011245728, -0.0046539307, 0.010482788, -0.022338867, -0.009498596, 0.0038032532, -0.03225708, -0.016983032, -0.02406311, 0.020004272, 0....
http://arxiv.org/abs/2308.15055v1
Taxonomic Loss for Morphological Glossing of Low-Resource Languages
Michael Ginn, Alexis Palmer
cs.CL, cs.LG
Morpheme glossing is a critical task in automated language documentation and can benefit other downstream applications greatly. While state-of-the-art glossing systems perform very well for languages with large amounts of existing data, it is more difficult to create useful models for low-resource languages. In this pa...
http://arxiv.org/pdf/2308.15055v1
2023-08-29T06:31:21
2023-08-29T06:31:21
{ "title": [ -0.045928955, -0.01184845, -0.041625977, -0.036895752, -0.018173218, 0.0060424805, -0.042999268, 0.0002863407, -0.003042221, 0.045837402, -0.020339966, 0.003894806, 0.014045715, -0.013748169, -0.03994751, 0.00044941902, -0.008575439, ...
http://arxiv.org/abs/2308.15053v2
Adapting Text-based Dialogue State Tracker for Spoken Dialogues
Jaeseok Yoon, Seunghyun Hwang, Ran Han, Jeonguk Bang, Kee-Eung Kim
cs.CL, cs.AI
Although there have been remarkable advances in dialogue systems through the dialogue systems technology competition (DSTC), it remains one of the key challenges to building a robust task-oriented dialogue system with a speech interface. Most of the progress has been made for text-based dialogue systems since there are...
http://arxiv.org/pdf/2308.15053v2
2023-08-30T04:46:19
2023-08-29T06:27:58
{ "title": [ -0.009185791, -0.026275635, -0.025726318, -0.07635498, -0.015686035, 0.006580353, -0.055847168, 0.0124435425, 0.01890564, 0.012573242, -0.010292053, -0.006225586, -0.01890564, -0.022323608, -0.05102539, -0.009269714, -0.01158905, 0.0...
http://arxiv.org/abs/2308.15047v1
Large language models converge toward human-like concept organization
Mathias Lykke Gammelgaard, Jonathan Gabel Christiansen, Anders Søgaard
cs.LG, cs.CL
Large language models show human-like performance in knowledge extraction, reasoning and dialogue, but it remains controversial whether this performance is best explained by memorization and pattern matching, or whether it reflects human-like inferential semantics and world knowledge. Knowledge bases such as WikiData p...
http://arxiv.org/pdf/2308.15047v1
2023-08-29T06:09:47
2023-08-29T06:09:47
{ "title": [ -0.0016393661, 0.0010871887, -0.037719727, -0.07849121, -0.029708862, -0.014556885, -0.064208984, 0.017181396, -0.007068634, 0.0524292, -0.021209717, -0.0011205673, 0.04006958, -0.016830444, -0.021316528, 0.018051147, 0.000477314, 0....
http://arxiv.org/abs/2308.15027v1
Improving Neural Ranking Models with Traditional IR Methods
Anik Saha, Oktie Hassanzadeh, Alex Gittens, Jian Ni, Kavitha Srinivas, Bulent Yener
cs.IR, cs.CL
Neural ranking methods based on large transformer models have recently gained significant attention in the information retrieval community, and have been adopted by major commercial solutions. Nevertheless, they are computationally expensive to create, and require a great deal of labeled data for specialized corpora. I...
http://arxiv.org/pdf/2308.15027v1
2023-08-29T05:18:47
2023-08-29T05:18:47
{ "title": [ 0.023452759, 0.01675415, -0.0072517395, -0.010612488, 0.018005371, -0.019302368, 0.016525269, -0.054718018, -0.0067481995, 0.013916016, -0.01826477, -0.0012369156, -0.051879883, -0.018325806, -0.053710938, -0.017684937, -0.014328003, ...
http://arxiv.org/abs/2308.15022v1
Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models
Qingyue Wang, Liang Ding, Yanan Cao, Zhiliang Tian, Shi Wang, Dacheng Tao, Li Guo
cs.CL, cs.AI
Most open-domain dialogue systems suffer from forgetting important information, especially in a long-term conversation. Existing works usually train the specific retriever or summarizer to obtain key information from the past, which is time-consuming and highly depends on the quality of labeled data. To alleviate this ...
http://arxiv.org/pdf/2308.15022v1
2023-08-29T04:59:53
2023-08-29T04:59:53
{ "title": [ 0.013084412, -0.016479492, -0.021011353, -0.109558105, -0.028442383, -0.0016822815, -0.09033203, 0.027114868, -0.0030994415, 0.045288086, 0.0035915375, 0.003622055, 0.013069153, -0.024978638, -0.043823242, 0.014884949, -0.0014181137, ...
http://arxiv.org/abs/2308.15010v1
TransPrompt v2: A Transferable Prompting Framework for Cross-task Text Classification
Jianing Wang, Chengyu Wang, Cen Chen, Ming Gao, Jun Huang, Aoying Zhou
cs.CL
Text classification is one of the most imperative tasks in natural language processing (NLP). Recent advances with pre-trained language models (PLMs) have shown remarkable success on this task. However, the satisfying results obtained by PLMs heavily depend on the large amounts of task-specific labeled data, which may ...
http://arxiv.org/pdf/2308.15010v1
2023-08-29T04:16:57
2023-08-29T04:16:57
{ "title": [ -0.0118255615, 0.0037269592, -0.036895752, -0.04800415, 0.0022525787, -0.010719299, -0.039764404, -0.019882202, 0.019607544, 0.0067253113, -0.04940796, 0.038085938, -0.012557983, -0.0031585693, -0.022766113, -0.008117676, -0.017822266, ...