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/2311.03780v1
Ensembling Textual and Structure-Based Models for Knowledge Graph Completion
Ananjan Nandi, Navdeep Kaur, Parag Singla, Mausam
cs.CL, cs.AI, cs.LG, I.2.7
We consider two popular approaches to Knowledge Graph Completion (KGC): textual models that rely on textual entity descriptions, and structure-based models that exploit the connectivity structure of the Knowledge Graph (KG). Preliminary experiments show that these approaches have complementary strengths: structure-base...
http://arxiv.org/pdf/2311.03780v1
2023-11-07T07:53:06
2023-11-07T07:53:06
{ "title": [ -0.04446411, -0.019729614, -0.06640625, -0.097351074, -0.009506226, 0.02709961, -0.09106445, 0.02267456, 0.010574341, 0.027679443, -0.03414917, 0.022384644, -0.0000078082085, 0.008712769, -0.046203613, -0.0021152496, 0.0032634735, 0....
http://arxiv.org/abs/2311.03767v1
Gender Inflected or Bias Inflicted: On Using Grammatical Gender Cues for Bias Evaluation in Machine Translation
Pushpdeep Singh
cs.CL
Neural Machine Translation (NMT) models are state-of-the-art for machine translation. However, these models are known to have various social biases, especially gender bias. Most of the work on evaluating gender bias in NMT has focused primarily on English as the source language. For source languages different from Engl...
http://arxiv.org/pdf/2311.03767v1
2023-11-07T07:09:59
2023-11-07T07:09:59
{ "title": [ -0.058624268, -0.014732361, -0.02709961, -0.010284424, -0.019927979, -0.008003235, -0.015266418, -0.009933472, 0.0047569275, 0.044189453, -0.034057617, 0.0019340515, 0.05203247, -0.007637024, -0.043823242, 0.0026950836, -0.043701172, ...
http://arxiv.org/abs/2311.03755v2
Multilingual Mathematical Autoformalization
Albert Q. Jiang, Wenda Li, Mateja Jamnik
cs.CL, cs.LG
Autoformalization is the task of translating natural language materials into machine-verifiable formalisations. Progress in autoformalization research is hindered by the lack of a sizeable dataset consisting of informal-formal pairs expressing the same essence. Existing methods tend to circumvent this challenge by manu...
http://arxiv.org/pdf/2311.03755v2
2023-11-09T09:58:15
2023-11-07T06:42:15
{ "title": [ -0.026138306, 0.02519226, -0.029922485, -0.068481445, 0.011787415, 0.013473511, -0.07122803, -0.011116028, -0.0071868896, 0.062408447, 0.018447876, -0.010223389, 0.028366089, -0.0357666, -0.013687134, 0.0418396, 0.011100769, -0.00984...
http://arxiv.org/abs/2311.03754v1
Which is better? Exploring Prompting Strategy For LLM-based Metrics
Joonghoon Kim, Saeran Park, Kiyoon Jeong, Sangmin Lee, Seung Hun Han, Jiyoon Lee, Pilsung Kang
cs.CL
This paper describes the DSBA submissions to the Prompting Large Language Models as Explainable Metrics shared task, where systems were submitted to two tracks: small and large summarization tracks. With advanced Large Language Models (LLMs) such as GPT-4, evaluating the quality of Natural Language Generation (NLG) has...
http://arxiv.org/pdf/2311.03754v1
2023-11-07T06:36:39
2023-11-07T06:36:39
{ "title": [ -0.07281494, -0.024414062, -0.030227661, -0.010231018, -0.047454834, 0.011917114, -0.06518555, -0.009979248, -0.027893066, -0.0037841797, -0.017562866, 0.022369385, -0.029083252, -0.046936035, 0.02482605, -0.013923645, 0.0041275024, ...
http://arxiv.org/abs/2311.03753v1
COOL: A Constraint Object-Oriented Logic Programming Language and its Neural-Symbolic Compilation System
Jipeng Han
cs.AI, cs.CL, cs.DC, cs.FL, cs.HC
This paper explores the integration of neural networks with logic programming, addressing the longstanding challenges of combining the generalization and learning capabilities of neural networks with the precision of symbolic logic. Traditional attempts at this integration have been hampered by difficulties in initial ...
http://arxiv.org/pdf/2311.03753v1
2023-11-07T06:29:59
2023-11-07T06:29:59
{ "title": [ -0.07373047, 0.0061569214, -0.0769043, -0.06903076, 0.007045746, 0.016860962, -0.08618164, 0.020965576, -0.021453857, 0.037841797, -0.03729248, 0.018310547, 0.008583069, -0.007820129, -0.024215698, -0.0035381317, -0.024429321, 0.0248...
http://arxiv.org/abs/2311.03748v1
Unified Low-Resource Sequence Labeling by Sample-Aware Dynamic Sparse Finetuning
Sarkar Snigdha Sarathi Das, Ranran Haoran Zhang, Peng Shi, Wenpeng Yin, Rui Zhang
cs.CL
Unified Sequence Labeling that articulates different sequence labeling problems such as Named Entity Recognition, Relation Extraction, Semantic Role Labeling, etc. in a generalized sequence-to-sequence format opens up the opportunity to make the maximum utilization of large language model knowledge toward structured pr...
http://arxiv.org/pdf/2311.03748v1
2023-11-07T06:19:37
2023-11-07T06:19:37
{ "title": [ 0.0042648315, -0.020019531, -0.05645752, -0.13952637, 0.0046653748, -0.003643036, -0.06311035, 0.060699463, -0.0022087097, 0.042510986, -0.025802612, 0.029464722, -0.012145996, -0.00434494, -0.036590576, 0.00881958, 0.027053833, 0.02...
http://arxiv.org/abs/2311.03734v1
Leveraging Structured Information for Explainable Multi-hop Question Answering and Reasoning
Ruosen Li, Xinya Du
cs.CL
Neural models, including large language models (LLMs), achieve superior performance on multi-hop question-answering. To elicit reasoning capabilities from LLMs, recent works propose using the chain-of-thought (CoT) mechanism to generate both the reasoning chain and the answer, which enhances the model's capabilities in...
http://arxiv.org/pdf/2311.03734v1
2023-11-07T05:32:39
2023-11-07T05:32:39
{ "title": [ -0.017974854, -0.029846191, -0.028564453, -0.08660889, -0.016586304, 0.013381958, -0.045532227, 0.03967285, 0.012039185, 0.009780884, -0.034179688, 0.002614975, -0.014183044, -0.003232956, -0.04046631, -0.017044067, 0.007949829, 0.00...
http://arxiv.org/abs/2311.03732v1
Learning to Learn for Few-shot Continual Active Learning
Stella Ho, Ming Liu, Shang Gao, Longxiang Gao
cs.LG, cs.CL
Continual learning strives to ensure stability in solving previously seen tasks while demonstrating plasticity in a novel domain. Recent advances in CL are mostly confined to a supervised learning setting, especially in NLP domain. In this work, we consider a few-shot continual active learning (CAL) setting where label...
http://arxiv.org/pdf/2311.03732v1
2023-11-07T05:22:11
2023-11-07T05:22:11
{ "title": [ 0.0037612915, -0.00819397, -0.017608643, -0.022415161, 0.008338928, -0.013175964, 0.009025574, -0.012863159, 0.005622864, 0.023101807, -0.00970459, 0.016937256, -0.040802002, -0.020523071, -0.033935547, 0.0031108856, 0.016281128, 0.0...
http://arxiv.org/abs/2311.03731v1
A Survey of Large Language Models Attribution
Dongfang Li, Zetian Sun, Xinshuo Hu, Zhenyu Liu, Ziyang Chen, Baotian Hu, Aiguo Wu, Min Zhang
cs.CL
Open-domain generative systems have gained significant attention in the field of conversational AI (e.g., generative search engines). This paper presents a comprehensive review of the attribution mechanisms employed by these systems, particularly large language models. Though attribution or citation improve the factual...
http://arxiv.org/pdf/2311.03731v1
2023-11-07T05:20:09
2023-11-07T05:20:09
{ "title": [ -0.018951416, -0.013076782, -0.03765869, -0.048461914, -0.0047798157, -0.0054397583, -0.057800293, -0.0060424805, -0.010513306, 0.023529053, -0.007843018, -0.0044021606, 0.008743286, -0.029434204, -0.012016296, 0.008857727, -0.027038574,...
http://arxiv.org/abs/2311.03716v1
LLM as an Art Director (LaDi): Using LLMs to improve Text-to-Media Generators
Allen Roush, Emil Zakirov, Artemiy Shirokov, Polina Lunina, Jack Gane, Alexander Duffy, Charlie Basil, Aber Whitcomb, Jim Benedetto, Chris DeWolfe
cs.CL, cs.AI, cs.CV
Recent advancements in text-to-image generation have revolutionized numerous fields, including art and cinema, by automating the generation of high-quality, context-aware images and video. However, the utility of these technologies is often limited by the inadequacy of text prompts in guiding the generator to produce a...
http://arxiv.org/pdf/2311.03716v1
2023-11-07T04:44:40
2023-11-07T04:44:40
{ "title": [ -0.043060303, -0.023239136, -0.049194336, 0.012306213, -0.060943604, -0.008171082, -0.022705078, 0.014297485, -0.058624268, 0.035003662, -0.013999939, 0.026428223, 0.004512787, -0.038635254, 0.03555298, -0.026870728, -0.0135650635, 0...
http://arxiv.org/abs/2311.03696v1
Bilingual Corpus Mining and Multistage Fine-Tuning for Improving Machine Translation of Lecture Transcripts
Haiyue Song, Raj Dabre, Chenhui Chu, Atsushi Fujita, Sadao Kurohashi
cs.CL
Lecture transcript translation helps learners understand online courses, however, building a high-quality lecture machine translation system lacks publicly available parallel corpora. To address this, we examine a framework for parallel corpus mining, which provides a quick and effective way to mine a parallel corpus f...
http://arxiv.org/pdf/2311.03696v1
2023-11-07T03:50:25
2023-11-07T03:50:25
{ "title": [ -0.025253296, -0.01235199, -0.014663696, -0.06317139, -0.008140564, 0.022918701, -0.046661377, 0.0135269165, -0.003255844, 0.06378174, -0.027374268, 0.009033203, 0.008285522, 0.009216309, -0.07678223, -0.017318726, -0.019302368, 0.01...
http://arxiv.org/abs/2311.03687v1
Dissecting the Runtime Performance of the Training, Fine-tuning, and Inference of Large Language Models
Longteng Zhang, Xiang Liu, Zeyu Li, Xinglin Pan, Peijie Dong, Ruibo Fan, Rui Guo, Xin Wang, Qiong Luo, Shaohuai Shi, Xiaowen Chu
cs.PF, cs.CL, cs.LG
Large Language Models (LLMs) have seen great advance in both academia and industry, and their popularity results in numerous open-source frameworks and techniques in accelerating LLM pre-training, fine-tuning, and inference. Training and deploying LLMs are expensive as it requires considerable computing resources and m...
http://arxiv.org/pdf/2311.03687v1
2023-11-07T03:25:56
2023-11-07T03:25:56
{ "title": [ -0.03353882, -0.01965332, -0.025436401, -0.062683105, -0.020568848, -0.008628845, -0.05621338, 0.010444641, -0.019607544, 0.025756836, -0.030532837, 0.023956299, -0.001540184, -0.005332947, -0.010238647, -0.0031852722, -0.007194519, ...
http://arxiv.org/abs/2311.03672v1
CBSiMT: Mitigating Hallucination in Simultaneous Machine Translation with Weighted Prefix-to-Prefix Training
Mengge Liu, Wen Zhang, Xiang Li, Yanzhi Tian, Yuhang Guo, Jian Luan, Bin Wang, Shuoying Chen
cs.CL
Simultaneous machine translation (SiMT) is a challenging task that requires starting translation before the full source sentence is available. Prefix-to-prefix framework is often applied to SiMT, which learns to predict target tokens using only a partial source prefix. However, due to the word order difference between ...
http://arxiv.org/pdf/2311.03672v1
2023-11-07T02:44:45
2023-11-07T02:44:45
{ "title": [ -0.019317627, -0.010528564, 0.0034885406, -0.04434204, -0.024429321, 0.020080566, -0.038909912, -0.011955261, -0.02432251, 0.05444336, 0.024124146, 0.028900146, 0.011100769, -0.024993896, -0.044921875, 0.030578613, -0.020935059, 0.04...
http://arxiv.org/abs/2311.03663v2
Principles from Clinical Research for NLP Model Generalization
Aparna Elangovan, Jiayuan He, Yuan Li, Karin Verspoor
cs.CL
The NLP community typically relies on performance of a model on a held-out test set to assess generalization. Performance drops observed in datasets outside of official test sets are generally attributed to "out-of-distribution'' effects. Here, we explore the foundations of generalizability and study the various factor...
http://arxiv.org/pdf/2311.03663v2
2023-11-09T15:09:04
2023-11-07T02:17:25
{ "title": [ -0.01828003, -0.011177063, -0.026992798, -0.064208984, -0.01134491, 0.00945282, -0.061584473, -0.010932922, -0.008857727, 0.03378296, -0.016464233, 0.011566162, -0.004512787, -0.021331787, -0.039642334, 0.014442444, -0.0128479, 0.059...
http://arxiv.org/abs/2311.03658v1
The Linear Representation Hypothesis and the Geometry of Large Language Models
Kiho Park, Yo Joong Choe, Victor Veitch
cs.CL, cs.AI, cs.LG, stat.ML
Informally, the 'linear representation hypothesis' is the idea that high-level concepts are represented linearly as directions in some representation space. In this paper, we address two closely related questions: What does "linear representation" actually mean? And, how do we make sense of geometric notions (e.g., cos...
http://arxiv.org/pdf/2311.03658v1
2023-11-07T01:59:11
2023-11-07T01:59:11
{ "title": [ -0.028411865, -0.018875122, -0.030166626, -0.05355835, -0.008758545, -0.0014371872, -0.049865723, 0.0032634735, -0.014465332, 0.030914307, -0.006336212, -0.00390625, 0.001625061, -0.014923096, -0.038635254, 0.014915466, -0.0041923523, ...
http://arxiv.org/abs/2311.03633v1
Innovation and Word Usage Patterns in Machine Learning
Vítor Bandeira Borges, Daniel Oliveira Cajueiro
cs.LG, cs.CL
In this study, we delve into the dynamic landscape of machine learning research evolution. Initially, through the utilization of Latent Dirichlet Allocation, we discern pivotal themes and fundamental concepts that have emerged within the realm of machine learning. Subsequently, we undertake a comprehensive analysis to ...
http://arxiv.org/pdf/2311.03633v1
2023-11-07T00:41:15
2023-11-07T00:41:15
{ "title": [ -0.026321411, -0.004776001, -0.049194336, -0.010620117, 0.007881165, -0.028015137, 0.010375977, -0.0060272217, -0.02986145, 0.02986145, -0.021087646, 0.016799927, -0.026443481, -0.0027427673, -0.02796936, 0.0046081543, -0.015449524, ...
http://arxiv.org/abs/2311.03627v1
GNAT: A General Narrative Alignment Tool
Tanzir Pial, Steven Skiena
cs.CL
Algorithmic sequence alignment identifies similar segments shared between pairs of documents, and is fundamental to many NLP tasks. But it is difficult to recognize similarities between distant versions of narratives such as translations and retellings, particularly for summaries and abridgements which are much shorter...
http://arxiv.org/pdf/2311.03627v1
2023-11-07T00:24:14
2023-11-07T00:24:14
{ "title": [ -0.038879395, -0.008071899, -0.05618286, -0.05215454, -0.0038814545, 0.0052719116, -0.04360962, -0.018478394, -0.012741089, 0.024337769, 0.0076675415, 0.0027980804, 0.0049743652, -0.038085938, -0.024215698, 0.0020542145, -0.017715454, ...
http://arxiv.org/abs/2311.04931v1
GPT4All: An Ecosystem of Open Source Compressed Language Models
Yuvanesh Anand, Zach Nussbaum, Adam Treat, Aaron Miller, Richard Guo, Ben Schmidt, GPT4All Community, Brandon Duderstadt, Andriy Mulyar
cs.CL, cs.AI
Large language models (LLMs) have recently achieved human-level performance on a range of professional and academic benchmarks. The accessibility of these models has lagged behind their performance. State-of-the-art LLMs require costly infrastructure; are only accessible via rate-limited, geo-locked, and censored web i...
http://arxiv.org/pdf/2311.04931v1
2023-11-06T23:50:20
2023-11-06T23:50:20
{ "title": [ -0.048980713, -0.01625061, -0.041778564, -0.055389404, -0.01576233, 0.02142334, -0.067993164, 0.009284973, 0.002298355, 0.03250122, 0.026733398, -0.012886047, 0.018371582, -0.03955078, -0.034332275, 0.031707764, 0.0028839111, 0.01120...
http://arxiv.org/abs/2311.03614v1
STONYBOOK: A System and Resource for Large-Scale Analysis of Novels
Charuta Pethe, Allen Kim, Rajesh Prabhakar, Tanzir Pial, Steven Skiena
cs.CL
Books have historically been the primary mechanism through which narratives are transmitted. We have developed a collection of resources for the large-scale analysis of novels, including: (1) an open source end-to-end NLP analysis pipeline for the annotation of novels into a standard XML format, (2) a collection of 49,...
http://arxiv.org/pdf/2311.03614v1
2023-11-06T23:46:40
2023-11-06T23:46:40
{ "title": [ -0.03591919, -0.01914978, -0.030273438, -0.039978027, -0.0030002594, -0.0038032532, -0.035980225, -0.011009216, 0.0010681152, 0.02998352, 0.010017395, 0.028518677, 0.022842407, -0.033325195, -0.037384033, 0.024673462, 0.0077819824, 0...
http://arxiv.org/abs/2311.03584v1
Dimensions of Online Conflict: Towards Modeling Agonism
Matt Canute, Mali Jin, hannah holtzclaw, Alberto Lusoli, Philippa R Adams, Mugdha Pandya, Maite Taboada, Diana Maynard, Wendy Hui Kyong Chun
cs.CL
Agonism plays a vital role in democratic dialogue by fostering diverse perspectives and robust discussions. Within the realm of online conflict there is another type: hateful antagonism, which undermines constructive dialogue. Detecting conflict online is central to platform moderation and monetization. It is also vita...
http://arxiv.org/pdf/2311.03584v1
2023-11-06T22:34:17
2023-11-06T22:34:17
{ "title": [ 0.024978638, 0.013336182, -0.019226074, -0.0413208, 0.030090332, -0.023834229, 0.019973755, -0.012466431, -0.020812988, 0.023406982, -0.0025863647, 0.011352539, -0.020339966, -0.023757935, -0.03970337, -0.016906738, 0.0036830902, -0....
http://arxiv.org/abs/2311.03566v1
Measuring Adversarial Datasets
Yuanchen Bai, Raoyi Huang, Vijay Viswanathan, Tzu-Sheng Kuo, Tongshuang Wu
cs.LG, cs.CL, cs.HC
In the era of widespread public use of AI systems across various domains, ensuring adversarial robustness has become increasingly vital to maintain safety and prevent undesirable errors. Researchers have curated various adversarial datasets (through perturbations) for capturing model deficiencies that cannot be reveale...
http://arxiv.org/pdf/2311.03566v1
2023-11-06T22:08:16
2023-11-06T22:08:16
{ "title": [ -0.022781372, 0.029556274, -0.037750244, -0.01084137, 0.01550293, -0.012748718, -0.008934021, -0.04309082, -0.037994385, 0.01713562, -0.022705078, 0.023162842, -0.057128906, -0.024002075, -0.020339966, -0.015007019, 0.01637268, 0.013...
http://arxiv.org/abs/2311.03551v1
Context Unlocks Emotions: Text-based Emotion Classification Dataset Auditing with Large Language Models
Daniel Yang, Aditya Kommineni, Mohammad Alshehri, Nilamadhab Mohanty, Vedant Modi, Jonathan Gratch, Shrikanth Narayanan
cs.CL, cs.AI
The lack of contextual information in text data can make the annotation process of text-based emotion classification datasets challenging. As a result, such datasets often contain labels that fail to consider all the relevant emotions in the vocabulary. This misalignment between text inputs and labels can degrade the p...
http://arxiv.org/pdf/2311.03551v1
2023-11-06T21:34:49
2023-11-06T21:34:49
{ "title": [ -0.01600647, 0.0063285828, -0.050964355, -0.039794922, 0.008178711, -0.015655518, -0.027313232, -0.008369446, 0.028411865, 0.036224365, -0.05038452, -0.009544373, 0.022201538, -0.009780884, -0.058807373, -0.008346558, -0.003479004, 0...
http://arxiv.org/abs/2311.03533v1
Quantifying Uncertainty in Natural Language Explanations of Large Language Models
Sree Harsha Tanneru, Chirag Agarwal, Himabindu Lakkaraju
cs.CL
Large Language Models (LLMs) are increasingly used as powerful tools for several high-stakes natural language processing (NLP) applications. Recent prompting works claim to elicit intermediate reasoning steps and key tokens that serve as proxy explanations for LLM predictions. However, there is no certainty whether the...
http://arxiv.org/pdf/2311.03533v1
2023-11-06T21:14:40
2023-11-06T21:14:40
{ "title": [ -0.04574585, -0.010765076, -0.039855957, -0.077697754, -0.002790451, 0.010673523, -0.071899414, 0.0030651093, 0.003753662, 0.0053367615, -0.022003174, 0.0095825195, -0.009552002, -0.018859863, -0.027236938, -0.0007882118, -0.012992859, ...
http://arxiv.org/abs/2311.03510v1
Spoken Dialogue System for Medical Prescription Acquisition on Smartphone: Development, Corpus and Evaluation
Ali Can Kocabiyikoglu, François Portet, Jean-Marc Babouchkine, Prudence Gibert, Hervé Blanchon, Gaëtan Gavazzi
cs.CL
Hospital information systems (HIS) have become an essential part of healthcare institutions and now incorporate prescribing support software. Prescription support software allows for structured information capture, which improves the safety, appropriateness and efficiency of prescriptions and reduces the number of adve...
http://arxiv.org/pdf/2311.03510v1
2023-11-06T20:36:55
2023-11-06T20:36:55
{ "title": [ -0.022644043, -0.012992859, -0.048034668, -0.06866455, 0.0002207756, 0.001704216, -0.050628662, 0.027694702, -0.02583313, 0.03918457, -0.02684021, 0.014305115, -0.032928467, -0.004272461, -0.053710938, -0.022201538, -0.014953613, 0.0...
http://arxiv.org/abs/2311.03498v1
In-Context Exemplars as Clues to Retrieving from Large Associative Memory
Jiachen Zhao
cs.CL, cs.LG
Recently, large language models (LLMs) have made remarkable progress in natural language processing. The most representative ability of LLMs is in-context learning (ICL), which enables LLMs to learn patterns from in-context exemplars without training. The performance of ICL greatly depends on the exemplars used. Howeve...
http://arxiv.org/pdf/2311.03498v1
2023-11-06T20:13:29
2023-11-06T20:13:29
{ "title": [ 0.026000977, -0.0059661865, -0.0072669983, -0.041381836, 0.00014913082, -0.020599365, 0.0069236755, -0.02394104, 0.001657486, 0.038330078, -0.032714844, 0.0035114288, 0.0020637512, -0.021087646, -0.030441284, -0.0020561218, 0.004234314, ...
http://arxiv.org/abs/2311.03348v1
Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation
Rusheb Shah, Quentin Feuillade--Montixi, Soroush Pour, Arush Tagade, Stephen Casper, Javier Rando
cs.CL, cs.AI, cs.LG
Despite efforts to align large language models to produce harmless responses, they are still vulnerable to jailbreak prompts that elicit unrestricted behaviour. In this work, we investigate persona modulation as a black-box jailbreaking method to steer a target model to take on personalities that are willing to comply ...
http://arxiv.org/pdf/2311.03348v1
2023-11-06T18:55:18
2023-11-06T18:55:18
{ "title": [ -0.01637268, -0.012687683, -0.0345459, -0.05392456, -0.003490448, 0.0026664734, -0.054229736, -0.0025138855, -0.03253174, 0.04043579, 0.014595032, 0.008987427, 0.02746582, -0.026000977, -0.022903442, 0.019851685, -0.027801514, 0.0390...
http://arxiv.org/abs/2311.03320v1
Tackling Concept Shift in Text Classification using Entailment-style Modeling
Sumegh Roychowdhury, Karan Gupta, Siva Rajesh Kasa, Prasanna Srinivasa Murthy, Alok Chandra
cs.CL
Pre-trained language models (PLMs) have seen tremendous success in text classification (TC) problems in the context of Natural Language Processing (NLP). In many real-world text classification tasks, the class definitions being learned do not remain constant but rather change with time - this is known as Concept Shift....
http://arxiv.org/pdf/2311.03320v1
2023-11-06T18:15:36
2023-11-06T18:15:36
{ "title": [ -0.008033752, 0.0045776367, -0.042144775, -0.03994751, 0.0077209473, -0.011436462, -0.017578125, -0.025970459, 0.015220642, 0.012565613, -0.011871338, 0.01348114, -0.013671875, -0.010627747, -0.034698486, -0.0046691895, -0.03717041, ...
http://arxiv.org/abs/2311.03319v1
DAIL: Data Augmentation for In-Context Learning via Self-Paraphrase
Dawei Li, Yaxuan Li, Dheeraj Mekala, Shuyao Li, Yulin wang, Xueqi Wang, William Hogan, Jingbo Shang
cs.CL, cs.AI
In-Context Learning (ICL) combined with pre-trained large language models has achieved promising results on various NLP tasks. However, ICL requires high-quality annotated demonstrations which might not be available in real-world scenarios. To overcome this limitation, we propose \textbf{D}ata \textbf{A}ugmentation for...
http://arxiv.org/pdf/2311.03319v1
2023-11-06T18:12:55
2023-11-06T18:12:55
{ "title": [ 0.0042915344, -0.006881714, -0.055267334, -0.061798096, 0.009895325, -0.01725769, -0.018814087, 0.0064468384, -0.018554688, 0.042236328, -0.027282715, 0.015899658, -0.0054740906, -0.00642395, -0.04434204, -0.0015497208, -0.00945282, ...
http://arxiv.org/abs/2311.03311v1
Unraveling Downstream Gender Bias from Large Language Models: A Study on AI Educational Writing Assistance
Thiemo Wambsganss, Xiaotian Su, Vinitra Swamy, Seyed Parsa Neshaei, Roman Rietsche, Tanja Käser
cs.CL, cs.CY
Large Language Models (LLMs) are increasingly utilized in educational tasks such as providing writing suggestions to students. Despite their potential, LLMs are known to harbor inherent biases which may negatively impact learners. Previous studies have investigated bias in models and data representations separately, ne...
http://arxiv.org/pdf/2311.03311v1
2023-11-06T18:01:34
2023-11-06T18:01:34
{ "title": [ -0.053344727, -0.009963989, -0.05230713, 0.0000538826, -0.021652222, -0.02142334, -0.016937256, -0.0061569214, -0.008560181, 0.04449463, -0.030319214, 0.0059890747, 0.019134521, 0.010040283, -0.047088623, -0.019744873, -0.03781128, 0...
http://arxiv.org/abs/2311.03301v1
Ziya2: Data-centric Learning is All LLMs Need
Ruyi Gan, Ziwei Wu, Renliang Sun, Junyu Lu, Xiaojun Wu, Dixiang Zhang, Kunhao Pan, Ping Yang, Qi Yang, Jiaxing Zhang, Yan Song
cs.CL
Various large language models (LLMs) have been proposed in recent years, including closed- and open-source ones, continually setting new records on multiple benchmarks. However, the development of LLMs still faces several issues, such as high cost of training models from scratch, and continual pre-training leading to c...
http://arxiv.org/pdf/2311.03301v1
2023-11-06T17:49:34
2023-11-06T17:49:34
{ "title": [ -0.048065186, -0.016159058, -0.0758667, -0.013504028, -0.052124023, -0.034606934, -0.049682617, 0.028381348, -0.046936035, 0.017089844, -0.041625977, 0.0095825195, -0.008338928, -0.03427124, 0.0146102905, -0.041412354, 0.007335663, 0...
http://arxiv.org/abs/2311.03287v2
Holistic Analysis of Hallucination in GPT-4V(ision): Bias and Interference Challenges
Chenhang Cui, Yiyang Zhou, Xinyu Yang, Shirley Wu, Linjun Zhang, James Zou, Huaxiu Yao
cs.LG, cs.CL, cs.CV
While GPT-4V(ision) impressively models both visual and textual information simultaneously, it's hallucination behavior has not been systematically assessed. To bridge this gap, we introduce a new benchmark, namely, the Bias and Interference Challenges in Visual Language Models (Bingo). This benchmark is designed to ev...
http://arxiv.org/pdf/2311.03287v2
2023-11-07T02:18:48
2023-11-06T17:26:59
{ "title": [ 0.019836426, 0.028900146, -0.01247406, -0.021530151, 0.0071792603, -0.020523071, 0.0012989044, -0.04650879, -0.022125244, 0.033935547, 0.0009860992, 0.011100769, 0.0046539307, -0.02758789, -0.020935059, 0.011680603, -0.023666382, 0.0...
http://arxiv.org/abs/2311.03253v1
Coherent Entity Disambiguation via Modeling Topic and Categorical Dependency
Zilin Xiao, Linjun Shou, Xingyao Zhang, Jie Wu, Ming Gong, Jian Pei, Daxin Jiang
cs.CL, cs.AI
Previous entity disambiguation (ED) methods adopt a discriminative paradigm, where prediction is made based on matching scores between mention context and candidate entities using length-limited encoders. However, these methods often struggle to capture explicit discourse-level dependencies, resulting in incoherent pre...
http://arxiv.org/pdf/2311.03253v1
2023-11-06T16:40:13
2023-11-06T16:40:13
{ "title": [ -0.020401001, 0.01171875, -0.056884766, -0.09020996, 0.030685425, 0.011566162, -0.05718994, 0.0018005371, 0.020584106, 0.03527832, -0.006137848, 0.010574341, 0.032409668, -0.017196655, -0.038146973, 0.02279663, -0.017211914, 0.016662...
http://arxiv.org/abs/2311.03250v1
Instructed Language Models with Retrievers Are Powerful Entity Linkers
Zilin Xiao, Ming Gong, Jie Wu, Xingyao Zhang, Linjun Shou, Jian Pei, Daxin Jiang
cs.CL, cs.AI
Generative approaches powered by large language models (LLMs) have demonstrated emergent abilities in tasks that require complex reasoning abilities. Yet the generative nature still makes the generated content suffer from hallucinations, thus unsuitable for entity-centric tasks like entity linking (EL) requiring precis...
http://arxiv.org/pdf/2311.03250v1
2023-11-06T16:38:51
2023-11-06T16:38:51
{ "title": [ 0.05230713, -0.0051994324, -0.025680542, -0.08392334, -0.008560181, -0.015174866, -0.038909912, 0.006752014, -0.034454346, 0.05255127, 0.026824951, -0.010643005, 0.06262207, -0.058685303, 0.028213501, 0.016113281, -0.010894775, 0.041...
http://arxiv.org/abs/2311.03243v1
Safurai-Csharp: Harnessing Synthetic Data to improve language-specific Code LLM
Davide Cifarelli, Leonardo Boiardi, Alessandro Puppo, Leon Jovanovic
cs.CL
This paper introduces Safurai-Csharp, an open-source model designed to specialize in the generation, completion, and debugging of C# code. Safurai-Csharp is built upon the novel CodeLlama 34B model and leverages the EvolInstruct technique, creating a refined and expanded dataset for its fine-tuning process. The results...
http://arxiv.org/pdf/2311.03243v1
2023-11-06T16:31:48
2023-11-06T16:31:48
{ "title": [ -0.06604004, -0.035125732, -0.05215454, -0.04626465, -0.0446167, 0.00843811, -0.078186035, 0.028030396, 0.012771606, 0.01096344, -0.037506104, 0.048858643, -0.0013990402, -0.020751953, 0.020874023, 0.009643555, -0.02734375, 0.0254058...
http://arxiv.org/abs/2311.03235v1
p-Laplacian Transformer
Tuan Nguyen, Tam Nguyen, Vinh Nguyen, Tan M. Nguyen
cs.LG, cs.CL, stat.ML
$p$-Laplacian regularization, rooted in graph and image signal processing, introduces a parameter $p$ to control the regularization effect on these data. Smaller values of $p$ promote sparsity and interpretability, while larger values encourage smoother solutions. In this paper, we first show that the self-attention me...
http://arxiv.org/pdf/2311.03235v1
2023-11-06T16:25:56
2023-11-06T16:25:56
{ "title": [ 0.011802673, 0.031158447, -0.03894043, -0.017959595, -0.021942139, -0.029144287, -0.036346436, -0.039733887, -0.024429321, 0.036193848, 0.000029027462, -0.029251099, 0.03857422, -0.055511475, 0.023864746, 0.00422287, 0.017364502, 0.0...
http://arxiv.org/abs/2311.07584v1
Performance Prediction of Data-Driven Knowledge summarization of High Entropy Alloys (HEAs) literature implementing Natural Language Processing algorithms
Akshansh Mishra, Vijaykumar S Jatti, Vaishnavi More, Anish Dasgupta, Devarrishi Dixit, Eyob Messele Sefene
cs.CL, cs.AI, cs.IR, cs.IT, cs.LG, math.IT
The ability to interpret spoken language is connected to natural language processing. It involves teaching the AI how words relate to one another, how they are meant to be used, and in what settings. The goal of natural language processing (NLP) is to get a machine intelligence to process words the same way a human bra...
http://arxiv.org/pdf/2311.07584v1
2023-11-06T16:22:32
2023-11-06T16:22:32
{ "title": [ -0.026748657, -0.004962921, -0.06982422, -0.049224854, -0.02909851, -0.024765015, -0.054260254, 0.012306213, 0.011047363, 0.079589844, -0.012077332, 0.008361816, 0.07122803, -0.019836426, -0.026397705, 0.03112793, 0.028717041, 0.0246...
http://arxiv.org/abs/2311.03228v1
An Efficient Self-Supervised Cross-View Training For Sentence Embedding
Peerat Limkonchotiwat, Wuttikorn Ponwitayarat, Lalita Lowphansirikul, Can Udomcharoenchaikit, Ekapol Chuangsuwanich, Sarana Nutanong
cs.CL, cs.AI
Self-supervised sentence representation learning is the task of constructing an embedding space for sentences without relying on human annotation efforts. One straightforward approach is to finetune a pretrained language model (PLM) with a representation learning method such as contrastive learning. While this approach...
http://arxiv.org/pdf/2311.03228v1
2023-11-06T16:12:25
2023-11-06T16:12:25
{ "title": [ -0.0018825531, 0.008041382, -0.018692017, -0.052093506, 0.0023860931, -0.0002450943, -0.03366089, -0.012672424, 0.03173828, 0.01007843, -0.031463623, 0.023773193, -0.0154418945, -0.009002686, -0.03161621, 0.0109939575, -0.015197754, ...
http://arxiv.org/abs/2311.03220v1
ALYMPICS: Language Agents Meet Game Theory
Shaoguang Mao, Yuzhe Cai, Yan Xia, Wenshan Wu, Xun Wang, Fengyi Wang, Tao Ge, Furu Wei
cs.CL, cs.AI, cs.GT
This paper introduces Alympics, a platform that leverages Large Language Model (LLM) agents to facilitate investigations in game theory. By employing LLMs and autonomous agents to simulate human behavior and enable multi-agent collaborations, we can construct realistic and dynamic models of human interactions for game ...
http://arxiv.org/pdf/2311.03220v1
2023-11-06T16:03:46
2023-11-06T16:03:46
{ "title": [ -0.010559082, -0.015838623, -0.0028533936, -0.008842468, -0.026809692, -0.006500244, 0.00057315826, -0.019546509, -0.037872314, 0.031341553, -0.001745224, 0.002401352, -0.035308838, -0.026321411, -0.029205322, -0.024307251, -0.0060272217...
http://arxiv.org/abs/2311.03216v1
Mini Minds: Exploring Bebeshka and Zlata Baby Models
Irina Proskurina, Guillaume Metzler, Julien Velcin
cs.CL, cs.AI, cs.LG
In this paper, we describe the University of Lyon 2 submission to the Strict-Small track of the BabyLM competition. The shared task is created with an emphasis on small-scale language modelling from scratch on limited-size data and human language acquisition. Dataset released for the Strict-Small track has 10M words, w...
http://arxiv.org/pdf/2311.03216v1
2023-11-06T16:01:10
2023-11-06T16:01:10
{ "title": [ 0.0012950897, -0.022872925, -0.049316406, 0.016921997, -0.046447754, -0.035858154, -0.0016183853, -0.015136719, -0.05480957, 0.047668457, -0.0256958, -0.015731812, 0.018707275, -0.033691406, 0.013389587, -0.035064697, 0.026397705, -0...
http://arxiv.org/abs/2311.03196v1
Pseudo-Labeling for Domain-Agnostic Bangla Automatic Speech Recognition
Rabindra Nath Nandi, Mehadi Hasan Menon, Tareq Al Muntasir, Sagor Sarker, Quazi Sarwar Muhtaseem, Md. Tariqul Islam, Shammur Absar Chowdhury, Firoj Alam
cs.CL, cs.AI, 68T50, F.2.2; I.2.7
One of the major challenges for developing automatic speech recognition (ASR) for low-resource languages is the limited access to labeled data with domain-specific variations. In this study, we propose a pseudo-labeling approach to develop a large-scale domain-agnostic ASR dataset. With the proposed methodology, we dev...
http://arxiv.org/pdf/2311.03196v1
2023-11-06T15:37:14
2023-11-06T15:37:14
{ "title": [ -0.013679504, 0.004600525, -0.04196167, -0.06402588, -0.008094788, 0.004573822, -0.04373169, 0.027755737, -0.017333984, 0.05050659, 0.0046195984, -0.024154663, 0.00995636, -0.043121338, -0.05117798, -0.0138549805, -0.0076828003, 0.04...
http://arxiv.org/abs/2311.03186v1
Model-based Counterfactual Generator for Gender Bias Mitigation
Ewoenam Kwaku Tokpo, Toon Calders
cs.CL
Counterfactual Data Augmentation (CDA) has been one of the preferred techniques for mitigating gender bias in natural language models. CDA techniques have mostly employed word substitution based on dictionaries. Although such dictionary-based CDA techniques have been shown to significantly improve the mitigation of gen...
http://arxiv.org/pdf/2311.03186v1
2023-11-06T15:25:30
2023-11-06T15:25:30
{ "title": [ 0.016174316, 0.00554657, -0.044952393, -0.033477783, 0.005706787, -0.02798462, -0.013061523, -0.022521973, 0.003353119, 0.04925537, -0.016647339, -0.01398468, 0.011161804, -0.015777588, -0.066711426, -0.007133484, -0.017150879, 0.038...
http://arxiv.org/abs/2311.03184v1
Nexus at ArAIEval Shared Task: Fine-Tuning Arabic Language Models for Propaganda and Disinformation Detection
Yunze Xiao, Firoj Alam
cs.CL, cs.AI, cs.SI, 68T50, F.2.2; I.2.7
The spread of disinformation and propagandistic content poses a threat to societal harmony, undermining informed decision-making and trust in reliable sources. Online platforms often serve as breeding grounds for such content, and malicious actors exploit the vulnerabilities of audiences to shape public opinion. Althou...
http://arxiv.org/pdf/2311.03184v1
2023-11-06T15:24:18
2023-11-06T15:24:18
{ "title": [ 0.013786316, -0.0065460205, -0.016052246, -0.039276123, -0.0056991577, -0.015380859, -0.020904541, -0.025909424, -0.039611816, 0.023208618, 0.0004386902, -0.013893127, 0.0019416809, -0.053100586, -0.0048828125, -0.022247314, -0.017944336...
http://arxiv.org/abs/2311.03179v1
ArAIEval Shared Task: Persuasion Techniques and Disinformation Detection in Arabic Text
Maram Hasanain, Firoj Alam, Hamdy Mubarak, Samir Abdaljalil, Wajdi Zaghouani, Preslav Nakov, Giovanni Da San Martino, Abed Alhakim Freihat
cs.CL, cs.AI, 68T50, F.2.2; I.2.7
We present an overview of the ArAIEval shared task, organized as part of the first ArabicNLP 2023 conference co-located with EMNLP 2023. ArAIEval offers two tasks over Arabic text: (i) persuasion technique detection, focusing on identifying persuasion techniques in tweets and news articles, and (ii) disinformation dete...
http://arxiv.org/pdf/2311.03179v1
2023-11-06T15:21:19
2023-11-06T15:21:19
{ "title": [ 0.025222778, 0.0016851425, -0.0129852295, -0.026138306, -0.0047950745, -0.016036987, 0.0015010834, -0.028503418, -0.05630493, 0.0048713684, -0.009391785, 0.018066406, -0.011489868, -0.06286621, 0.032806396, -0.04058838, -0.021759033, ...
http://arxiv.org/abs/2311.03153v1
Architectural Sweet Spots for Modeling Human Label Variation by the Example of Argument Quality: It's Best to Relate Perspectives!
Philipp Heinisch, Matthias Orlikowski, Julia Romberg, Philipp Cimiano
cs.CL
Many annotation tasks in natural language processing are highly subjective in that there can be different valid and justified perspectives on what is a proper label for a given example. This also applies to the judgment of argument quality, where the assignment of a single ground truth is often questionable. At the sam...
http://arxiv.org/pdf/2311.03153v1
2023-11-06T14:47:48
2023-11-06T14:47:48
{ "title": [ 0.024719238, -0.020843506, 0.008766174, -0.026351929, -0.032928467, -0.0025539398, -0.044036865, -0.02394104, -0.021911621, 0.0027580261, 0.0064468384, 0.020523071, -0.039154053, -0.055236816, 0.0043792725, -0.027648926, -0.018600464, ...
http://arxiv.org/abs/2311.03127v1
Findings of the WMT 2023 Shared Task on Discourse-Level Literary Translation: A Fresh Orb in the Cosmos of LLMs
Longyue Wang, Zhaopeng Tu, Yan Gu, Siyou Liu, Dian Yu, Qingsong Ma, Chenyang Lyu, Liting Zhou, Chao-Hong Liu, Yufeng Ma, Weiyu Chen, Yvette Graham, Bonnie Webber, Philipp Koehn, Andy Way, Yulin Yuan, Shuming Shi
cs.CL, cs.AI
Translating literary works has perennially stood as an elusive dream in machine translation (MT), a journey steeped in intricate challenges. To foster progress in this domain, we hold a new shared task at WMT 2023, the first edition of the Discourse-Level Literary Translation. First, we (Tencent AI Lab and China Litera...
http://arxiv.org/pdf/2311.03127v1
2023-11-06T14:23:49
2023-11-06T14:23:49
{ "title": [ -0.029785156, -0.028411865, -0.009315491, -0.009399414, -0.03994751, -0.010314941, -0.028625488, 0.0073509216, -0.008605957, 0.03643799, 0.02748108, 0.024978638, 0.05529785, -0.053985596, 0.021148682, 0.042785645, 0.007827759, 0.0204...
http://arxiv.org/abs/2311.03420v1
Text Augmentations with R-drop for Classification of Tweets Self Reporting Covid-19
Sumam Francis, Marie-Francine Moens
cs.CL, cs.IR, cs.LG
This paper presents models created for the Social Media Mining for Health 2023 shared task. Our team addressed the first task, classifying tweets that self-report Covid-19 diagnosis. Our approach involves a classification model that incorporates diverse textual augmentations and utilizes R-drop to augment data and miti...
http://arxiv.org/pdf/2311.03420v1
2023-11-06T14:18:16
2023-11-06T14:18:16
{ "title": [ -0.0006828308, 0.013305664, -0.047698975, -0.03515625, -0.017578125, -0.005973816, -0.031173706, 0.012367249, -0.03781128, 0.045288086, -0.034454346, 0.03326416, 0.024246216, -0.01259613, 0.011276245, -0.022628784, 0.0022735596, 0.01...
http://arxiv.org/abs/2311.03113v1
Injecting Categorical Labels and Syntactic Information into Biomedical NER
Sumam Francis, Marie-Francine Moens
cs.CL, cs.IR, cs.LG
We present a simple approach to improve biomedical named entity recognition (NER) by injecting categorical labels and Part-of-speech (POS) information into the model. We use two approaches, in the first approach, we first train a sequence-level classifier to classify the sentences into categories to obtain the sentence...
http://arxiv.org/pdf/2311.03113v1
2023-11-06T14:03:59
2023-11-06T14:03:59
{ "title": [ -0.005710602, -0.0063667297, -0.015808105, -0.05996704, -0.031829834, 0.021896362, -0.045562744, -0.027435303, -0.04888916, 0.015853882, 0.0007138252, 0.0062713623, -0.013473511, -0.056427002, 0.009681702, -0.005706787, -0.0032653809, ...
http://arxiv.org/abs/2311.03099v1
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Le Yu, Bowen Yu, Haiyang Yu, Fei Huang, Yongbin Li
cs.CL, cs.LG
In this paper, we uncover that Language Models (LMs), either encoder- or decoder-based, can obtain new capabilities by assimilating the parameters of homologous models without retraining or GPUs. Typically, new abilities of LMs can be imparted by Supervised Fine-Tuning (SFT), reflected in the disparity between fine-tun...
http://arxiv.org/pdf/2311.03099v1
2023-11-06T13:43:07
2023-11-06T13:43:07
{ "title": [ -0.007686615, -0.0005068779, -0.026260376, -0.07550049, -0.010009766, 0.005329132, -0.05859375, 0.00010514259, -0.011413574, 0.032806396, -0.0024662018, 0.0010938644, 0.019454956, -0.032836914, -0.031051636, 0.022033691, -0.00374794, ...
http://arxiv.org/abs/2311.03084v2
A Simple yet Efficient Ensemble Approach for AI-generated Text Detection
Harika Abburi, Kalyani Roy, Michael Suesserman, Nirmala Pudota, Balaji Veeramani, Edward Bowen, Sanmitra Bhattacharya
cs.CL, cs.AI
Recent Large Language Models (LLMs) have demonstrated remarkable capabilities in generating text that closely resembles human writing across wide range of styles and genres. However, such capabilities are prone to potential abuse, such as fake news generation, spam email creation, and misuse in academic assignments. He...
http://arxiv.org/pdf/2311.03084v2
2023-11-08T04:28:57
2023-11-06T13:11:02
{ "title": [ -0.025115967, 0.008285522, -0.02923584, -0.023254395, 0.014289856, -0.010635376, -0.0061149597, -0.020324707, 0.01083374, 0.036376953, -0.035614014, 0.011734009, -0.0033359528, 0.0015287399, -0.05899048, -0.0019292831, -0.019439697, ...
http://arxiv.org/abs/2311.03078v1
BanLemma: A Word Formation Dependent Rule and Dictionary Based Bangla Lemmatizer
Sadia Afrin, Md. Shahad Mahmud Chowdhury, Md. Ekramul Islam, Faisal Ahamed Khan, Labib Imam Chowdhury, MD. Motahar Mahtab, Nazifa Nuha Chowdhury, Massud Forkan, Neelima Kundu, Hakim Arif, Mohammad Mamun Or Rashid, Mohammad Ruhul Amin, Nabeel Mohammed
cs.CL
Lemmatization holds significance in both natural language processing (NLP) and linguistics, as it effectively decreases data density and aids in comprehending contextual meaning. However, due to the highly inflected nature and morphological richness, lemmatization in Bangla text poses a complex challenge. In this study...
http://arxiv.org/pdf/2311.03078v1
2023-11-06T13:02:07
2023-11-06T13:02:07
{ "title": [ -0.025024414, -0.0027942657, -0.055511475, -0.060394287, -0.0047454834, 0.0037288666, -0.061187744, 0.018722534, -0.025344849, 0.02003479, -0.021347046, 0.0049438477, -0.0016441345, -0.024597168, -0.015205383, -0.010726929, -0.0236969, ...
http://arxiv.org/abs/2311.03058v1
Zero-shot Bilingual App Reviews Mining with Large Language Models
Jialiang Wei, Anne-Lise Courbis, Thomas Lambolais, Binbin Xu, Pierre Louis Bernard, Gérard Dray
cs.CL
App reviews from app stores are crucial for improving software requirements. A large number of valuable reviews are continually being posted, describing software problems and expected features. Effectively utilizing user reviews necessitates the extraction of relevant information, as well as their subsequent summarizat...
http://arxiv.org/pdf/2311.03058v1
2023-11-06T12:36:46
2023-11-06T12:36:46
{ "title": [ -0.04043579, -0.039764404, -0.0077438354, -0.013687134, -0.034820557, 0.014381409, -0.037597656, 0.0038604736, 0.008125305, 0.050201416, -0.025497437, 0.006591797, -0.011260986, -0.0022392273, -0.05807495, 0.0040359497, -0.02406311, ...
http://arxiv.org/abs/2311.03026v1
Detecting Agreement in Multi-party Conversational AI
Laura Schauer, Jason Sweeney, Charlie Lyttle, Zein Said, Aron Szeles, Cale Clark, Katie McAskill, Xander Wickham, Tom Byars, Daniel Hernández Garcia, Nancie Gunson, Angus Addlesee, Oliver Lemon
cs.CL
Today, conversational systems are expected to handle conversations in multi-party settings, especially within Socially Assistive Robots (SARs). However, practical usability remains difficult as there are additional challenges to overcome, such as speaker recognition, addressee recognition, and complex turn-taking. In t...
http://arxiv.org/pdf/2311.03026v1
2023-11-06T11:04:39
2023-11-06T11:04:39
{ "title": [ -0.027816772, 0.0071868896, -0.06488037, -0.048828125, 0.009819031, -0.023605347, -0.048950195, -0.004219055, 0.028121948, 0.0287323, -0.025512695, -0.001666069, 0.0049324036, -0.010910034, -0.05834961, 0.0018606186, -0.028320312, 0....
http://arxiv.org/abs/2311.03021v1
Detecting agreement in multi-party dialogue: evaluating speaker diarisation versus a procedural baseline to enhance user engagement
Angus Addlesee, Daniel Denley, Andy Edmondson, Nancie Gunson, Daniel Hernández Garcia, Alexandre Kha, Oliver Lemon, James Ndubuisi, Neil O'Reilly, Lia Perochaud, Raphaël Valeri, Miebaka Worika
cs.CL
Conversational agents participating in multi-party interactions face significant challenges in dialogue state tracking, since the identity of the speaker adds significant contextual meaning. It is common to utilise diarisation models to identify the speaker. However, it is not clear if these are accurate enough to corr...
http://arxiv.org/pdf/2311.03021v1
2023-11-06T11:00:44
2023-11-06T11:00:44
{ "title": [ -0.0011739731, 0.010169983, -0.062683105, -0.057800293, 0.01008606, -0.029785156, -0.04598999, -0.003692627, 0.005470276, 0.03250122, 0.010025024, -0.0006866455, -0.009162903, -0.029006958, -0.060577393, -0.0077667236, -0.02696228, 0...
http://arxiv.org/abs/2311.02985v2
Towards a Transformer-Based Reverse Dictionary Model for Quality Estimation of Definitions
Julien Guité-Vinet, Alexandre Blondin Massé, Fatiha Sadat
cs.CL
In the last years, several variants of transformers have emerged. In this paper, we compare different transformer-based models for solving the reverse dictionary task and explore their use in the context of a serious game called The Dictionary Game.
http://arxiv.org/pdf/2311.02985v2
2023-11-08T15:00:19
2023-11-06T09:42:44
{ "title": [ -0.046539307, 0.008354187, -0.01423645, -0.010231018, -0.012619019, 0.017044067, -0.055236816, -0.051116943, 0.0072898865, 0.019332886, -0.03845215, -0.008972168, 0.018859863, -0.017456055, -0.03366089, -0.005050659, -0.025360107, 0....
http://arxiv.org/abs/2311.02962v1
Retrieval-Augmented Code Generation for Universal Information Extraction
Yucan Guo, Zixuan Li, Xiaolong Jin, Yantao Liu, Yutao Zeng, Wenxuan Liu, Xiang Li, Pan Yang, Long Bai, Jiafeng Guo, Xueqi Cheng
cs.AI, cs.CL, cs.IR
Information Extraction (IE) aims to extract structural knowledge (e.g., entities, relations, events) from natural language texts, which brings challenges to existing methods due to task-specific schemas and complex text expressions. Code, as a typical kind of formalized language, is capable of describing structural kno...
http://arxiv.org/pdf/2311.02962v1
2023-11-06T09:03:21
2023-11-06T09:03:21
{ "title": [ -0.028213501, -0.012939453, -0.039215088, -0.0657959, -0.00919342, 0.012298584, -0.052215576, 0.014587402, 0.0055770874, 0.04547119, -0.016967773, 0.029922485, 0.0015220642, -0.001127243, -0.043518066, 0.015327454, -0.017242432, 0.03...
http://arxiv.org/abs/2311.02961v1
Adapting Pre-trained Generative Models for Extractive Question Answering
Prabir Mallick, Tapas Nayak, Indrajit Bhattacharya
cs.CL
Pre-trained Generative models such as BART, T5, etc. have gained prominence as a preferred method for text generation in various natural language processing tasks, including abstractive long-form question answering (QA) and summarization. However, the potential of generative models in extractive QA tasks, where discrim...
http://arxiv.org/pdf/2311.02961v1
2023-11-06T09:01:02
2023-11-06T09:01:02
{ "title": [ -0.014511108, -0.008407593, -0.020523071, -0.060577393, -0.014152527, 0.0032291412, -0.04067993, 0.0006432533, 0.02331543, 0.02104187, -0.019210815, -0.002588272, -0.027252197, -0.0027770996, -0.06640625, -0.0019569397, 0.0021762848, ...
http://arxiv.org/abs/2311.02956v1
In-Context Learning for Knowledge Base Question Answering for Unmanned Systems based on Large Language Models
Yunlong Chen, Yaming Zhang, Jianfei Yu, Li Yang, Rui Xia
cs.CL, cs.AI, I.2.7
Knowledge Base Question Answering (KBQA) aims to answer factoid questions based on knowledge bases. However, generating the most appropriate knowledge base query code based on Natural Language Questions (NLQ) poses a significant challenge in KBQA. In this work, we focus on the CCKS2023 Competition of Question Answering...
http://arxiv.org/pdf/2311.02956v1
2023-11-06T08:52:11
2023-11-06T08:52:11
{ "title": [ -0.009506226, 0.0005726814, -0.027694702, -0.04006958, 0.0057868958, -0.012756348, -0.024520874, -0.0038547516, -0.003376007, 0.03289795, -0.051208496, -0.006626129, -0.010124207, -0.0007710457, -0.038116455, -0.026382446, 0.0068893433, ...
http://arxiv.org/abs/2311.04235v1
Can LLMs Follow Simple Rules?
Norman Mu, Sarah Chen, Zifan Wang, Sizhe Chen, David Karamardian, Lulwa Aljeraisy, Dan Hendrycks, David Wagner
cs.AI, cs.CL, cs.LG
As Large Language Models (LLMs) are deployed with increasing real-world responsibilities, it is important to be able to specify and constrain the behavior of these systems in a reliable manner. Model developers may wish to set explicit rules for the model, such as "do not generate abusive content", but these may be cir...
http://arxiv.org/pdf/2311.04235v1
2023-11-06T08:50:29
2023-11-06T08:50:29
{ "title": [ -0.019943237, 0.01576233, -0.044799805, -0.0340271, -0.020980835, -0.0029182434, -0.06286621, -0.018203735, -0.026901245, 0.02696228, -0.017684937, 0.013244629, -0.03213501, -0.020126343, 0.0112838745, -0.035369873, 0.0129852295, -0....
http://arxiv.org/abs/2311.02945v1
PhoGPT: Generative Pre-training for Vietnamese
Dat Quoc Nguyen, Linh The Nguyen, Chi Tran, Dung Ngoc Nguyen, Nhung Nguyen, Thien Huu Nguyen, Dinh Phung, Hung Bui
cs.CL
We open-source a state-of-the-art 7.5B-parameter generative model series named PhoGPT for Vietnamese, which includes the base pre-trained monolingual model PhoGPT-7B5 and its instruction-following variant, PhoGPT-7B5-Instruct. In addition, we also demonstrate its superior performance compared to previous open-source mo...
http://arxiv.org/pdf/2311.02945v1
2023-11-06T08:26:14
2023-11-06T08:26:14
{ "title": [ -0.014205933, -0.030029297, -0.03265381, -0.047851562, -0.036987305, 0.0030403137, -0.04711914, -0.0008416176, -0.027877808, 0.020141602, -0.021972656, -0.001540184, 0.00042676926, -0.040222168, -0.0049057007, -0.022766113, -0.0022125244...
http://arxiv.org/abs/2311.02883v1
SQLPrompt: In-Context Text-to-SQL with Minimal Labeled Data
Ruoxi Sun, Sercan Ö. Arik, Rajarishi Sinha, Hootan Nakhost, Hanjun Dai, Pengcheng Yin, Tomas Pfister
cs.CL
Text-to-SQL aims to automate the process of generating SQL queries on a database from natural language text. In this work, we propose "SQLPrompt", tailored to improve the few-shot prompting capabilities of Text-to-SQL for Large Language Models (LLMs). Our methods include innovative prompt design, execution-based consis...
http://arxiv.org/pdf/2311.02883v1
2023-11-06T05:24:06
2023-11-06T05:24:06
{ "title": [ 0.004295349, -0.013221741, -0.08203125, -0.082092285, 0.011405945, -0.016448975, -0.030929565, 0.026763916, 0.019805908, 0.038604736, -0.06036377, 0.0035419464, 0.026748657, -0.024291992, -0.019989014, -0.016326904, 0.00592041, 0.023...
http://arxiv.org/abs/2311.02861v1
Less than One-shot: Named Entity Recognition via Extremely Weak Supervision
Letian Peng, Zihan Wang, Jingbo Shang
cs.CL
We study the named entity recognition (NER) problem under the extremely weak supervision (XWS) setting, where only one example entity per type is given in a context-free way. While one can see that XWS is lighter than one-shot in terms of the amount of supervision, we propose a novel method X-NER that can outperform th...
http://arxiv.org/pdf/2311.02861v1
2023-11-06T04:20:42
2023-11-06T04:20:42
{ "title": [ -0.0053634644, -0.0047302246, -0.03189087, -0.07147217, 0.01373291, -0.00016963482, -0.03024292, 0.005634308, 0.0070991516, 0.016082764, -0.046661377, 0.018722534, 0.00025701523, 0.0042419434, -0.025802612, -0.005680084, -0.0077171326, ...
http://arxiv.org/abs/2311.02851v1
Improving Machine Translation with Large Language Models: A Preliminary Study with Cooperative Decoding
Jiali Zeng, Fandong Meng, Yongjing Yin, Jie Zhou
cs.CL
Contemporary translation engines built upon the encoder-decoder framework have reached a high level of development, while the emergence of Large Language Models (LLMs) has disrupted their position by offering the potential for achieving superior translation quality. Therefore, it is crucial to understand in which scena...
http://arxiv.org/pdf/2311.02851v1
2023-11-06T03:41:57
2023-11-06T03:41:57
{ "title": [ -0.049194336, -0.010406494, -0.012168884, -0.041381836, -0.022064209, 0.016281128, -0.061035156, -0.0005393028, -0.00068855286, 0.064697266, -0.0069618225, 0.0018978119, 0.036865234, -0.013679504, -0.06225586, 0.025985718, -0.010284424, ...
http://arxiv.org/abs/2311.02849v2
Co-training and Co-distillation for Quality Improvement and Compression of Language Models
Hayeon Lee, Rui Hou, Jongpil Kim, Davis Liang, Hongbo Zhang, Sung Ju Hwang, Alexander Min
cs.CL, cs.AI
Knowledge Distillation (KD) compresses computationally expensive pre-trained language models (PLMs) by transferring their knowledge to smaller models, allowing their use in resource-constrained or real-time settings. However, most smaller models fail to surpass the performance of the original larger model, resulting in...
http://arxiv.org/pdf/2311.02849v2
2023-11-07T18:41:55
2023-11-06T03:29:00
{ "title": [ -0.039367676, 0.0035686493, -0.05142212, -0.07696533, 0.0009937286, 0.020874023, -0.08538818, 0.012176514, 0.0011692047, 0.045135498, -0.0051879883, 0.0037021637, 0.025543213, -0.025527954, -0.031951904, 0.016403198, -0.012481689, 0....
http://arxiv.org/abs/2311.02807v1
QualEval: Qualitative Evaluation for Model Improvement
Vishvak Murahari, Ameet Deshpande, Peter Clark, Tanmay Rajpurohit, Ashish Sabharwal, Karthik Narasimhan, Ashwin Kalyan
cs.LG, cs.AI, cs.CL
Quantitative evaluation metrics have traditionally been pivotal in gauging the advancements of artificial intelligence systems, including large language models (LLMs). However, these metrics have inherent limitations. Given the intricate nature of real-world tasks, a single scalar to quantify and compare is insufficien...
http://arxiv.org/pdf/2311.02807v1
2023-11-06T00:21:44
2023-11-06T00:21:44
{ "title": [ -0.022872925, 0.011154175, -0.041107178, -0.044708252, 0.030670166, 0.0039863586, -0.033966064, -0.05984497, -0.0018949509, 0.018218994, -0.00868988, 0.0058135986, -0.03149414, -0.029769897, -0.054138184, 0.00055742264, -0.017547607, ...
http://arxiv.org/abs/2311.02805v1
Tailoring Self-Rationalizers with Multi-Reward Distillation
Sahana Ramnath, Brihi Joshi, Skyler Hallinan, Ximing Lu, Liunian Harold Li, Aaron Chan, Jack Hessel, Yejin Choi, Xiang Ren
cs.CL
Large language models (LMs) are capable of generating free-text rationales to aid question answering. However, prior work 1) suggests that useful self-rationalization is emergent only at significant scales (e.g., 175B parameter GPT-3); and 2) focuses largely on downstream performance, ignoring the semantics of the rati...
http://arxiv.org/pdf/2311.02805v1
2023-11-06T00:20:11
2023-11-06T00:20:11
{ "title": [ 0.0335083, 0.024307251, -0.027374268, -0.07385254, 0.019744873, -0.009941101, -0.026870728, -0.024459839, -0.03149414, 0.022323608, 0.000021874905, 0.0033226013, -0.05480957, -0.038848877, -0.01637268, -0.03466797, 0.0064964294, 0.03...
http://arxiv.org/abs/2311.02802v1
Incorporating Worker Perspectives into MTurk Annotation Practices for NLP
Olivia Huang, Eve Fleisig, Dan Klein
cs.CL, cs.AI
Current practices regarding data collection for natural language processing on Amazon Mechanical Turk (MTurk) often rely on a combination of studies on data quality and heuristics shared among NLP researchers. However, without considering the perspectives of MTurk workers, these approaches are susceptible to issues reg...
http://arxiv.org/pdf/2311.02802v1
2023-11-06T00:06:11
2023-11-06T00:06:11
{ "title": [ 0.0104522705, -0.0030078888, -0.023773193, -0.06488037, -0.0075416565, -0.02394104, -0.054260254, 0.0028152466, -0.018173218, 0.028656006, -0.01940918, 0.002784729, -0.0056152344, -0.015533447, -0.023910522, -0.005432129, -0.0036029816, ...
http://arxiv.org/abs/2311.02790v1
CausalCite: A Causal Formulation of Paper Citations
Ishan Kumar, Zhijing Jin, Ehsan Mokhtarian, Siyuan Guo, Yuen Chen, Negar Kiyavash, Mrinmaya Sachan, Bernhard Schoelkopf
cs.CL, cs.AI, cs.CY, cs.IR, cs.LG
Evaluating the significance of a paper is pivotal yet challenging for the scientific community. While the citation count is the most commonly used proxy for this purpose, they are widely criticized for failing to accurately reflect a paper's true impact. In this work, we propose a causal inference method, TextMatch, wh...
http://arxiv.org/pdf/2311.02790v1
2023-11-05T23:09:39
2023-11-05T23:09:39
{ "title": [ -0.016937256, -0.011795044, -0.04711914, -0.0592041, 0.0044822693, -0.0063171387, -0.026153564, -0.022094727, -0.021743774, 0.013404846, -0.00957489, -0.0049362183, 0.008338928, -0.033843994, -0.021911621, -0.0069770813, -0.015823364, ...
http://arxiv.org/abs/2311.04930v1
Large language models implicitly learn to straighten neural sentence trajectories to construct a predictive representation of natural language
Eghbal A. Hosseini, Evelina Fedorenko
cs.CL, cs.AI
Predicting upcoming events is critical to our ability to interact with our environment. Transformer models, trained on next-word prediction, appear to construct representations of linguistic input that can support diverse downstream tasks. But how does a predictive objective shape such representations? Inspired by rece...
http://arxiv.org/pdf/2311.04930v1
2023-11-05T22:16:21
2023-11-05T22:16:21
{ "title": [ -0.0107421875, -0.013961792, 0.0015573502, -0.010269165, -0.055023193, -0.008148193, -0.053863525, -0.009017944, 0.013893127, 0.029571533, -0.033447266, -0.017440796, 0.017501831, -0.024002075, -0.00027871132, -0.015098572, 0.019821167, ...
http://arxiv.org/abs/2311.02777v1
Robust Generalization Strategies for Morpheme Glossing in an Endangered Language Documentation Context
Michael Ginn, Alexis Palmer
cs.CL, cs.LG
Generalization is of particular importance in resource-constrained settings, where the available training data may represent only a small fraction of the distribution of possible texts. We investigate the ability of morpheme labeling models to generalize by evaluating their performance on unseen genres of text, and we ...
http://arxiv.org/pdf/2311.02777v1
2023-11-05T21:45:57
2023-11-05T21:45:57
{ "title": [ -0.014411926, -0.010978699, -0.020645142, -0.041534424, -0.02357483, 0.0068244934, -0.0340271, 0.0059776306, -0.0067100525, 0.044891357, -0.034606934, 0.019073486, 0.0032043457, -0.005420685, -0.035095215, -0.02571106, -0.030563354, ...
http://arxiv.org/abs/2311.02775v2
ChaTA: Towards an Intelligent Question-Answer Teaching Assistant using Open-Source LLMs
Yann Hicke, Anmol Agarwal, Qianou Ma, Paul Denny
cs.LG, cs.AI, cs.CL
Responding to the thousands of student questions on online QA platforms each semester has a considerable human cost, particularly in computing courses with rapidly growing enrollments. To address the challenges of scalable and intelligent question-answering (QA), we introduce an innovative solution that leverages open-...
http://arxiv.org/pdf/2311.02775v2
2023-11-13T16:03:15
2023-11-05T21:43:02
{ "title": [ -0.05911255, -0.020614624, -0.029190063, -0.016616821, -0.02609253, 0.013328552, -0.03491211, 0.00061941147, 0.017730713, 0.040130615, -0.046539307, 0.014831543, 0.0023860931, -0.000119805336, -0.05331421, -0.01373291, -0.01687622, 0...
http://arxiv.org/abs/2311.02772v1
Attention or Convolution: Transformer Encoders in Audio Language Models for Inference Efficiency
Sungho Jeon, Ching-Feng Yeh, Hakan Inan, Wei-Ning Hsu, Rashi Rungta, Yashar Mehdad, Daniel Bikel
cs.SD, cs.CL, eess.AS
In this paper, we show that a simple self-supervised pre-trained audio model can achieve comparable inference efficiency to more complicated pre-trained models with speech transformer encoders. These speech transformers rely on mixing convolutional modules with self-attention modules. They achieve state-of-the-art perf...
http://arxiv.org/pdf/2311.02772v1
2023-11-05T21:30:10
2023-11-05T21:30:10
{ "title": [ -0.011383057, -0.0043907166, -0.04663086, -0.05670166, -0.00365448, -0.019866943, -0.04788208, 0.0010061264, 0.008911133, 0.043273926, -0.032836914, -0.011581421, 0.067993164, -0.016571045, -0.022476196, 0.014007568, -0.010192871, 0....
http://arxiv.org/abs/2311.02765v1
Rule Learning as Machine Translation using the Atomic Knowledge Bank
Kristoffer Æsøy, Ana Ozaki
cs.CL, cs.AI
Machine learning models, and in particular language models, are being applied to various tasks that require reasoning. While such models are good at capturing patterns their ability to reason in a trustable and controlled manner is frequently questioned. On the other hand, logic-based rule systems allow for controlled ...
http://arxiv.org/pdf/2311.02765v1
2023-11-05T20:48:54
2023-11-05T20:48:54
{ "title": [ -0.05718994, -0.02418518, -0.05621338, -0.055633545, -0.021270752, 0.011116028, -0.0690918, 0.02229309, -0.027038574, 0.04269409, -0.0051651, 0.012275696, 0.008956909, -0.020568848, -0.03643799, 0.00060129166, -0.001906395, 0.0370178...
http://arxiv.org/abs/2311.02748v1
Pyclipse, a library for deidentification of free-text clinical notes
Callandra Moore, Jonathan Ranisau, Walter Nelson, Jeremy Petch, Alistair Johnson
cs.CL
Automated deidentification of clinical text data is crucial due to the high cost of manual deidentification, which has been a barrier to sharing clinical text and the advancement of clinical natural language processing. However, creating effective automated deidentification tools faces several challenges, including iss...
http://arxiv.org/pdf/2311.02748v1
2023-11-05T19:56:58
2023-11-05T19:56:58
{ "title": [ -0.015396118, -0.018249512, -0.051452637, -0.045654297, -0.04034424, 0.00014913082, -0.09906006, -0.006416321, -0.0050697327, 0.023498535, -0.022109985, -0.0110321045, 0.043182373, -0.049713135, 0.020385742, -0.018997192, 0.016586304, ...
http://arxiv.org/abs/2311.02702v1
Extraction of Atypical Aspects from Customer Reviews: Datasets and Experiments with Language Models
Smita Nannaware, Erfan Al-Hossami, Razvan Bunescu
cs.CL, cs.AI
A restaurant dinner may become a memorable experience due to an unexpected aspect enjoyed by the customer, such as an origami-making station in the waiting area. If aspects that are atypical for a restaurant experience were known in advance, they could be leveraged to make recommendations that have the potential to eng...
http://arxiv.org/pdf/2311.02702v1
2023-11-05T16:15:50
2023-11-05T16:15:50
{ "title": [ -0.019577026, -0.008331299, -0.021316528, -0.027389526, -0.005958557, 0.0012159348, -0.051116943, -0.025985718, 0.0056762695, 0.042510986, -0.014404297, 0.011909485, 0.008644104, 0.004020691, -0.06506348, 0.008529663, -0.0154953, 0.0...
http://arxiv.org/abs/2311.02699v1
Nepali Video Captioning using CNN-RNN Architecture
Bipesh Subedi, Saugat Singh, Bal Krishna Bal
cs.CV, cs.CL, cs.LG, I.2.7; I.2.10
This article presents a study on Nepali video captioning using deep neural networks. Through the integration of pre-trained CNNs and RNNs, the research focuses on generating precise and contextually relevant captions for Nepali videos. The approach involves dataset collection, data preprocessing, model implementation, ...
http://arxiv.org/pdf/2311.02699v1
2023-11-05T16:09:40
2023-11-05T16:09:40
{ "title": [ -0.030731201, 0.0043411255, -0.019683838, -0.029220581, -0.021820068, 0.00063991547, -0.023040771, -0.00737381, -0.024291992, 0.05255127, -0.016296387, -0.004699707, 0.020126343, -0.025802612, -0.0385437, 0.001663208, -0.008010864, 0...
http://arxiv.org/abs/2311.02660v1
LLM-enhanced Self-training for Cross-domain Constituency Parsing
Jianling Li, Meishan Zhang, Peiming Guo, Min Zhang, Yue Zhang
cs.CL
Self-training has proven to be an effective approach for cross-domain tasks, and in this study, we explore its application to cross-domain constituency parsing. Traditional self-training methods rely on limited and potentially low-quality raw corpora. To overcome this limitation, we propose enhancing self-training with...
http://arxiv.org/pdf/2311.02660v1
2023-11-05T14:13:29
2023-11-05T14:13:29
{ "title": [ -0.004776001, 0.01121521, -0.035186768, -0.07330322, 0.0140686035, -0.0017518997, -0.0524292, 0.004589081, -0.037078857, 0.015731812, -0.008468628, 0.041229248, -0.033721924, -0.023712158, 0.0040397644, 0.008758545, -0.011192322, 0.0...
http://arxiv.org/abs/2311.07583v1
Cross-Dialect Sentence Transformation: A Comparative Analysis of Language Models for Adapting Sentences to British English
Shruti Dutta, Shashwat Mookherjee
cs.CL, cs.AI
This study explores linguistic distinctions among American, Indian, and Irish English dialects and assesses various Language Models (LLMs) in their ability to generate British English translations from these dialects. Using cosine similarity analysis, the study measures the linguistic proximity between original British...
http://arxiv.org/pdf/2311.07583v1
2023-11-05T12:56:28
2023-11-05T12:56:28
{ "title": [ -0.033203125, 0.0018825531, -0.02104187, -0.025497437, 0.0052108765, 0.008460999, -0.040863037, -0.016082764, 0.0017127991, 0.040802002, 0.006893158, -0.0028381348, 0.031677246, -0.02482605, -0.057159424, 0.022155762, -0.015975952, 0...
http://arxiv.org/abs/2311.02616v1
Divide & Conquer for Entailment-aware Multi-hop Evidence Retrieval
Fan Luo, Mihai Surdeanu
cs.CL, cs.IR
Lexical and semantic matches are commonly used as relevance measurements for information retrieval. Together they estimate the semantic equivalence between the query and the candidates. However, semantic equivalence is not the only relevance signal that needs to be considered when retrieving evidences for multi-hop que...
http://arxiv.org/pdf/2311.02616v1
2023-11-05T10:31:40
2023-11-05T10:31:40
{ "title": [ 0.010818481, -0.02645874, -0.032958984, -0.078186035, -0.0026283264, 0.001499176, -0.013290405, 0.035858154, 0.021331787, 0.031433105, -0.04373169, 0.026947021, 0.0057754517, 0.0029563904, -0.024246216, 0.0075683594, -0.012237549, -0...
http://arxiv.org/abs/2311.02602v1
Get the Ball Rolling: Alerting Autonomous Robots When to Help to Close the Healthcare Loop
Jiaxin Shen, Yanyao Liu, Ziming Wang, Ziyuan Jiao, Yufeng Chen, Wenjuan Han
cs.RO, cs.AI, cs.CL
To facilitate the advancement of research in healthcare robots without human intervention or commands, we introduce the Autonomous Helping Challenge, along with a crowd-sourcing large-scale dataset. The goal is to create healthcare robots that possess the ability to determine when assistance is necessary, generate usef...
http://arxiv.org/pdf/2311.02602v1
2023-11-05T08:57:59
2023-11-05T08:57:59
{ "title": [ -0.006576538, 0.012199402, -0.03817749, -0.031707764, 0.031341553, -0.020614624, -0.014282227, -0.023529053, -0.058258057, 0.010749817, -0.035369873, 0.030029297, -0.031585693, -0.018737793, -0.0048599243, -0.04989624, 0.031341553, 0...
http://arxiv.org/abs/2311.02597v1
FloodBrain: Flood Disaster Reporting by Web-based Retrieval Augmented Generation with an LLM
Grace Colverd, Paul Darm, Leonard Silverberg, Noah Kasmanoff
cs.AI, cs.CL, I.2.7
Fast disaster impact reporting is crucial in planning humanitarian assistance. Large Language Models (LLMs) are well known for their ability to write coherent text and fulfill a variety of tasks relevant to impact reporting, such as question answering or text summarization. However, LLMs are constrained by the knowledg...
http://arxiv.org/pdf/2311.02597v1
2023-11-05T08:34:26
2023-11-05T08:34:26
{ "title": [ -0.023254395, -0.0050735474, -0.07171631, -0.052368164, -0.010864258, -0.017730713, -0.031982422, 0.01927185, 0.0032653809, 0.041870117, -0.029464722, 0.0209198, 0.045440674, -0.039855957, 0.0022296906, -0.0010147095, -0.012924194, -...
http://arxiv.org/abs/2311.02579v1
mahaNLP: A Marathi Natural Language Processing Library
Vidula Magdum, Omkar Dhekane, Sharayu Hiwarkhedkar, Saloni Mittal, Raviraj Joshi
cs.CL, cs.LG
We present mahaNLP, an open-source natural language processing (NLP) library specifically built for the Marathi language. It aims to enhance the support for the low-resource Indian language Marathi in the field of NLP. It is an easy-to-use, extensible, and modular toolkit for Marathi text analysis built on state-of-the...
http://arxiv.org/pdf/2311.02579v1
2023-11-05T06:59:59
2023-11-05T06:59:59
{ "title": [ -0.02746582, -0.024230957, -0.036193848, -0.074279785, -0.010116577, 0.013534546, -0.07751465, 0.0021190643, 0.0028514862, 0.004470825, -0.03918457, 0.01676941, 0.012794495, -0.022964478, -0.017486572, -0.00001335144, -0.0098724365, ...
http://arxiv.org/abs/2311.02578v1
Temporal Sequencing of Documents
Michael Gervers, Gelila Tilahun
cs.CL
We outline an unsupervised method for temporal rank ordering of sets of historical documents, namely American State of the Union Addresses and DEEDS, a corpus of medieval English property transfer documents. Our method relies upon effectively capturing the gradual change in word usage via a bandwidth estimate for the n...
http://arxiv.org/pdf/2311.02578v1
2023-11-05T06:51:04
2023-11-05T06:51:04
{ "title": [ 0.007911682, -0.008903503, -0.04788208, -0.08862305, -0.03881836, -0.02494812, -0.07458496, 0.026733398, -0.0034713745, 0.07409668, -0.039489746, -0.016998291, 0.04324341, -0.019836426, -0.016677856, -0.007850647, 0.027389526, 0.0373...
http://arxiv.org/abs/2311.02570v1
BanMANI: A Dataset to Identify Manipulated Social Media News in Bangla
Mahammed Kamruzzaman, Md. Minul Islam Shovon, Gene Louis Kim
cs.CL
Initial work has been done to address fake news detection and misrepresentation of news in the Bengali language. However, no work in Bengali yet addresses the identification of specific claims in social media news that falsely manipulates a related news article. At this point, this problem has been tackled in English a...
http://arxiv.org/pdf/2311.02570v1
2023-11-05T05:49:57
2023-11-05T05:49:57
{ "title": [ -0.009307861, 0.002149582, -0.041290283, -0.034606934, 0.019958496, -0.022628784, 0.004306793, -0.0014076233, -0.036224365, 0.010559082, -0.016616821, 0.012413025, 0.0082473755, -0.03488159, -0.012832642, -0.03186035, -0.031829834, 0...
http://arxiv.org/abs/2311.02566v1
Topic model based on co-occurrence word networks for unbalanced short text datasets
Chengjie Ma, Junping Du, Meiyu Liang, Zeli Guan
cs.CL
We propose a straightforward solution for detecting scarce topics in unbalanced short-text datasets. Our approach, named CWUTM (Topic model based on co-occurrence word networks for unbalanced short text datasets), Our approach addresses the challenge of sparse and unbalanced short text topics by mitigating the effects ...
http://arxiv.org/pdf/2311.02566v1
2023-11-05T04:44:23
2023-11-05T04:44:23
{ "title": [ -0.039031982, 0.016082764, -0.041870117, -0.009521484, -0.0087509155, -0.0035934448, -0.044708252, -0.04421997, 0.022659302, 0.033355713, -0.02204895, 0.038360596, -0.0037517548, -0.005344391, -0.030029297, 0.009857178, -0.022857666, ...
http://arxiv.org/abs/2311.02564v1
Relation Extraction Model Based on Semantic Enhancement Mechanism
Peiyu Liu, Junping Du, Yingxia Shao, Zeli Guan
cs.CL
Relational extraction is one of the basic tasks related to information extraction in the field of natural language processing, and is an important link and core task in the fields of information extraction, natural language understanding, and information retrieval. None of the existing relation extraction methods can e...
http://arxiv.org/pdf/2311.02564v1
2023-11-05T04:40:39
2023-11-05T04:40:39
{ "title": [ -0.005580902, 0.004634857, -0.06384277, -0.07696533, 0.02355957, 0.0013141632, -0.03845215, 0.0036411285, 0.022140503, 0.026184082, -0.0058898926, -0.0020427704, -0.0039901733, -0.011787415, -0.050872803, 0.013450623, -0.018798828, 0...
http://arxiv.org/abs/2311.07582v1
Evaluating the Potential of Leading Large Language Models in Reasoning Biology Questions
Xinyu Gong, Jason Holmes, Yiwei Li, Zhengliang Liu, Qi Gan, Zihao Wu, Jianli Zhang, Yusong Zou, Yuxi Teng, Tian Jiang, Hongtu Zhu, Wei Liu, Tianming Liu, Yajun Yan
cs.CL, cs.AI
Recent advances in Large Language Models (LLMs) have presented new opportunities for integrating Artificial General Intelligence (AGI) into biological research and education. This study evaluated the capabilities of leading LLMs, including GPT-4, GPT-3.5, PaLM2, Claude2, and SenseNova, in answering conceptual biology q...
http://arxiv.org/pdf/2311.07582v1
2023-11-05T03:34:17
2023-11-05T03:34:17
{ "title": [ -0.036865234, -0.0317688, -0.037231445, -0.045288086, -0.026733398, -0.00020956993, -0.051330566, 0.0113220215, 0.011619568, 0.0007805824, -0.018920898, -0.026062012, -0.007843018, -0.03665161, -0.026016235, -0.0068359375, -0.010810852, ...
http://arxiv.org/abs/2311.02499v1
Can Chat GPT solve a Linguistics Exam?
Patricia Ronan, Gerold Schneider
cs.CL, J.5; K.3
The present study asks if ChatGPT4, the version of ChatGPT which uses the language model GPT4, can successfully solve introductory linguistic exams. Previous exam questions of an Introduction to Linguistics course at a German university are used to test this. The exam questions were fed into ChatGPT4 with only minimal ...
http://arxiv.org/pdf/2311.02499v1
2023-11-04T20:02:57
2023-11-04T20:02:57
{ "title": [ -0.0039043427, -0.022338867, 0.043151855, 0.0021896362, -0.05899048, 0.030029297, -0.039489746, -0.03289795, -0.0011196136, 0.00970459, -0.011001587, -0.028198242, -0.009674072, -0.03161621, 0.0035800934, -0.02519226, 0.0029773712, 0...
http://arxiv.org/abs/2311.02408v3
Citance-Contextualized Summarization of Scientific Papers
Shahbaz Syed, Ahmad Dawar Hakimi, Khalid Al-Khatib, Martin Potthast
cs.CL
Current approaches to automatic summarization of scientific papers generate informative summaries in the form of abstracts. However, abstracts are not intended to show the relationship between a paper and the references cited in it. We propose a new contextualized summarization approach that can generate an informative...
http://arxiv.org/pdf/2311.02408v3
2023-11-13T08:40:04
2023-11-04T14:08:15
{ "title": [ 0.0020008087, 0.003665924, -0.024536133, -0.08148193, -0.0065956116, 0.0021972656, -0.044036865, -0.012969971, -0.0043563843, 0.060638428, -0.023986816, 0.005428314, 0.029449463, -0.009117126, -0.03930664, 0.010498047, -0.009376526, ...
http://arxiv.org/abs/2311.02355v1
TreeSwap: Data Augmentation for Machine Translation via Dependency Subtree Swapping
Attila Nagy, Dorina Lakatos, Botond Barta, Judit Ács
cs.CL
Data augmentation methods for neural machine translation are particularly useful when limited amount of training data is available, which is often the case when dealing with low-resource languages. We introduce a novel augmentation method, which generates new sentences by swapping objects and subjects across bisentence...
http://arxiv.org/pdf/2311.02355v1
2023-11-04T09:27:40
2023-11-04T09:27:40
{ "title": [ -0.041900635, -0.02885437, -0.043945312, -0.07989502, -0.011985779, 0.017227173, -0.053527832, 0.036621094, 0.0049552917, 0.03503418, -0.025344849, 0.0032787323, 0.013633728, -0.009681702, -0.052703857, 0.0008940697, -0.0009870529, 0...
http://arxiv.org/abs/2311.02345v1
Perturbation-based Active Learning for Question Answering
Fan Luo, Mihai Surdeanu
cs.CL, cs.AI, cs.LG
Building a question answering (QA) model with less annotation costs can be achieved by utilizing active learning (AL) training strategy. It selects the most informative unlabeled training data to update the model effectively. Acquisition functions for AL are used to determine how informative each training example is, s...
http://arxiv.org/pdf/2311.02345v1
2023-11-04T08:07:23
2023-11-04T08:07:23
{ "title": [ 0.0070114136, 0.014198303, 0.0026016235, -0.05255127, 0.01953125, -0.005077362, -0.0015325546, -0.006591797, 0.0335083, 0.014587402, -0.022659302, 0.0028915405, -0.023605347, -0.020690918, -0.03857422, 0.0014009476, 0.013160706, 0.02...
http://arxiv.org/abs/2311.02344v1
You Only Forward Once: Prediction and Rationalization in A Single Forward Pass
Han Jiang, Junwen Duan, Zhe Qu, Jianxin Wang
cs.CL, cs.AI
Unsupervised rationale extraction aims to extract concise and contiguous text snippets to support model predictions without any annotated rationale. Previous studies have used a two-phase framework known as the Rationalizing Neural Prediction (RNP) framework, which follows a generate-then-predict paradigm. They assumed...
http://arxiv.org/pdf/2311.02344v1
2023-11-04T08:04:28
2023-11-04T08:04:28
{ "title": [ 0.008590698, -0.021392822, -0.010566711, -0.06137085, -0.009529114, -0.007675171, -0.017547607, -0.028640747, -0.037231445, 0.02848816, -0.023284912, 0.029891968, -0.027145386, -0.015602112, -0.03414917, -0.01991272, 0.016860962, 0.0...
http://arxiv.org/abs/2311.02341v1
Enhancing English Writing Proficiency in China's Polytechnic Students An In-Depth Literature Review on the Application of the Input Hypothesis
Wei Zhou
cs.CL
Having good English writing skills is extremely important for students in polytechnic institutions. However, a lot of students in technical schools have difficulties in reaching high levels of skill. The Input Hypothesis, created by Stephen Krashen, suggests that people learn languages well when they receive informatio...
http://arxiv.org/pdf/2311.02341v1
2023-11-04T07:41:12
2023-11-04T07:41:12
{ "title": [ 0.024932861, -0.0006747246, -0.014472961, 0.039978027, -0.0033683777, -0.037628174, 0.04345703, -0.03186035, -0.068725586, 0.025878906, 0.0049438477, 0.034362793, -0.019546509, -0.037872314, 0.0021419525, -0.019897461, -0.052520752, ...
http://arxiv.org/abs/2311.02321v1
Identifying Context-Dependent Translations for Evaluation Set Production
Rachel Wicks, Matt Post
cs.CL
A major impediment to the transition to context-aware machine translation is the absence of good evaluation metrics and test sets. Sentences that require context to be translated correctly are rare in test sets, reducing the utility of standard corpus-level metrics such as COMET or BLEU. On the other hand, datasets tha...
http://arxiv.org/pdf/2311.02321v1
2023-11-04T04:29:08
2023-11-04T04:29:08
{ "title": [ -0.04925537, 0.013206482, -0.05355835, -0.03591919, 0.012199402, 0.0022068024, -0.05718994, -0.010787964, -0.026275635, 0.04876709, -0.0017766953, 0.019546509, -0.03591919, -0.019424438, -0.043670654, -0.012550354, -0.009483337, 0.03...
http://arxiv.org/abs/2311.02310v1
Narrowing the Gap between Zero- and Few-shot Machine Translation by Matching Styles
Weiting Tan, Haoran Xu, Lingfeng Shen, Shuyue Stella Li, Kenton Murray, Philipp Koehn, Benjamin Van Durme, Yunmo Chen
cs.CL
Large language models trained primarily in a monolingual setting have demonstrated their ability to generalize to machine translation using zero- and few-shot examples with in-context learning. However, even though zero-shot translations are relatively good, there remains a discernible gap comparing their performance w...
http://arxiv.org/pdf/2311.02310v1
2023-11-04T03:18:45
2023-11-04T03:18:45
{ "title": [ -0.03643799, -0.021606445, -0.01386261, -0.035705566, -0.029144287, 0.009033203, -0.04751587, 0.0042648315, -0.00223732, 0.05493164, 0.0009226799, 0.011657715, 0.035186768, -0.025558472, -0.03704834, 0.023086548, 0.01322937, 0.005867...
http://arxiv.org/abs/2311.02294v1
LLMs grasp morality in concept
Mark Pock, Andre Ye, Jared Moore
cs.CL, cs.CY
Work in AI ethics and fairness has made much progress in regulating LLMs to reflect certain values, such as fairness, truth, and diversity. However, it has taken the problem of how LLMs might 'mean' anything at all for granted. Without addressing this, it is not clear what imbuing LLMs with such values even means. In r...
http://arxiv.org/pdf/2311.02294v1
2023-11-04T01:37:41
2023-11-04T01:37:41
{ "title": [ 0.0256958, -0.0019512177, 0.032806396, -0.0056648254, -0.054840088, -0.008026123, -0.01789856, -0.039611816, -0.00037765503, 0.026947021, -0.034088135, -0.013114929, -0.031982422, -0.016159058, -0.0054779053, -0.034698486, -0.009994507, ...
http://arxiv.org/abs/2311.02271v2
FaMeSumm: Investigating and Improving Faithfulness of Medical Summarization
Nan Zhang, Yusen Zhang, Wu Guo, Prasenjit Mitra, Rui Zhang
cs.CL, cs.AI
Summaries of medical text shall be faithful by being consistent and factual with source inputs, which is an important but understudied topic for safety and efficiency in healthcare. In this paper, we investigate and improve faithfulness in summarization on a broad range of medical summarization tasks. Our investigation...
http://arxiv.org/pdf/2311.02271v2
2023-11-08T22:54:33
2023-11-03T23:25:53
{ "title": [ -0.0029678345, 0.000104248524, -0.029647827, -0.08935547, 0.0044670105, 0.013710022, -0.0552063, -0.01322937, -0.005634308, 0.057769775, 0.0046463013, 0.019241333, 0.022277832, -0.006389618, -0.051239014, 0.023406982, 0.0015745163, 0...
http://arxiv.org/abs/2311.02265v2
Not all layers are equally as important: Every Layer Counts BERT
Lucas Georges Gabriel Charpentier, David Samuel
cs.CL
This paper introduces a novel modification of the transformer architecture, tailored for the data-efficient pretraining of language models. This aspect is evaluated by participating in the BabyLM challenge, where our solution won both the strict and strict-small tracks. Our approach allows each transformer layer to sel...
http://arxiv.org/pdf/2311.02265v2
2023-11-07T21:36:11
2023-11-03T23:08:50
{ "title": [ 0.027618408, -0.008079529, -0.00806427, -0.061401367, -0.01071167, -0.0012598038, -0.017669678, 0.0019607544, -0.036315918, 0.023208618, -0.015602112, 0.0010490417, 0.01979065, -0.04714966, 0.022521973, -0.02708435, -0.008956909, 0.0...
http://arxiv.org/abs/2311.02262v1
Tell Your Model Where to Attend: Post-hoc Attention Steering for LLMs
Qingru Zhang, Chandan Singh, Liyuan Liu, Xiaodong Liu, Bin Yu, Jianfeng Gao, Tuo Zhao
cs.CL, cs.LG
In human-written articles, we often leverage the subtleties of text style, such as bold and italics, to guide the attention of readers. These textual emphases are vital for the readers to grasp the conveyed information. When interacting with large language models (LLMs), we have a similar need - steering the model to p...
http://arxiv.org/pdf/2311.02262v1
2023-11-03T22:56:43
2023-11-03T22:56:43
{ "title": [ -0.009376526, 0.004142761, -0.03665161, -0.033447266, -0.018173218, -0.009849548, -0.050689697, -0.044158936, -0.012496948, 0.012611389, -0.03955078, 0.021240234, -0.008644104, -0.029205322, 0.0023460388, -0.022583008, 0.0008869171, ...
http://arxiv.org/abs/2311.02248v1
COSMIC: Data Efficient Instruction-tuning For Speech In-Context Learning
Jing Pan, Jian Wu, Yashesh Gaur, Sunit Sivasankaran, Zhuo Chen, Shujie Liu, Jinyu Li
cs.CL, cs.AI, eess.AS
We present a data and cost efficient way of incorporating the speech modality into a large language model (LLM). The resulting multi-modal LLM is a COntextual Speech Model with Instruction-following/in-context-learning Capabilities - COSMIC. Speech comprehension test question-answer (SQA) pairs are generated using GPT-...
http://arxiv.org/pdf/2311.02248v1
2023-11-03T21:47:03
2023-11-03T21:47:03
{ "title": [ -0.01524353, -0.022827148, -0.04135132, -0.07208252, -0.0072402954, -0.015945435, -0.034606934, 0.033477783, -0.02027893, 0.037994385, -0.04812622, 0.027008057, 0.009124756, -0.0019416809, -0.02519226, -0.0141067505, -0.012542725, 0....
http://arxiv.org/abs/2311.02236v1
Robust Fine-Tuning of Vision-Language Models for Domain Generalization
Kevin Vogt-Lowell, Noah Lee, Theodoros Tsiligkaridis, Marc Vaillant
cs.CV, cs.AI, cs.CL, cs.LG
Transfer learning enables the sharing of common knowledge among models for a variety of downstream tasks, but traditional methods suffer in limited training data settings and produce narrow models incapable of effectively generalizing under distribution shifts. Foundation models have recently demonstrated impressive ze...
http://arxiv.org/pdf/2311.02236v1
2023-11-03T20:50:40
2023-11-03T20:50:40
{ "title": [ -0.015670776, 0.0048217773, -0.033416748, -0.05429077, 0.0032634735, -0.014587402, -0.04663086, -0.011581421, 0.01777649, 0.04095459, -0.03111267, 0.00724411, 0.002948761, -0.0008625984, -0.058929443, -0.0023155212, 0.005180359, 0.03...