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/2307.15290v1
ChatHome: Development and Evaluation of a Domain-Specific Language Model for Home Renovation
Cheng Wen, Xianghui Sun, Shuaijiang Zhao, Xiaoquan Fang, Liangyu Chen, Wei Zou
cs.CL
This paper presents the development and evaluation of ChatHome, a domain-specific language model (DSLM) designed for the intricate field of home renovation. Considering the proven competencies of large language models (LLMs) like GPT-4 and the escalating fascination with home renovation, this study endeavors to reconci...
http://arxiv.org/pdf/2307.15290v1
2023-07-28T04:04:43
2023-07-28T04:04:43
{ "title": [ -0.04812622, -0.042541504, -0.068481445, -0.034332275, -0.018478394, -0.017929077, -0.056762695, 0.01828003, -0.02571106, 0.042877197, -0.0011692047, 0.018325806, 0.018035889, -0.026000977, -0.017333984, 0.00077056885, -0.01398468, 0...
http://arxiv.org/abs/2307.15286v1
Multilingual Lexical Simplification via Paraphrase Generation
Kang Liu, Jipeng Qiang, Yun Li, Yunhao Yuan, Yi Zhu, Kaixun Hua
cs.CL
Lexical simplification (LS) methods based on pretrained language models have made remarkable progress, generating potential substitutes for a complex word through analysis of its contextual surroundings. However, these methods require separate pretrained models for different languages and disregard the preservation of ...
http://arxiv.org/pdf/2307.15286v1
2023-07-28T03:47:44
2023-07-28T03:47:44
{ "title": [ -0.029541016, -0.020767212, -0.014350891, -0.07611084, -0.017074585, 0.03781128, -0.08605957, 0.0073509216, -0.0066986084, 0.06512451, 0.0052986145, 0.016525269, 0.017105103, -0.008758545, -0.046325684, 0.020309448, -0.017089844, 0.0...
http://arxiv.org/abs/2308.00085v2
Reasoning before Responding: Integrating Commonsense-based Causality Explanation for Empathetic Response Generation
Yahui Fu, Koji Inoue, Chenhui Chu, Tatsuya Kawahara
cs.CL, cs.AI
Recent approaches to empathetic response generation try to incorporate commonsense knowledge or reasoning about the causes of emotions to better understand the user's experiences and feelings. However, these approaches mainly focus on understanding the causalities of context from the user's perspective, ignoring the sy...
http://arxiv.org/pdf/2308.00085v2
2023-09-05T05:45:30
2023-07-28T01:52:16
{ "title": [ 0.023956299, -0.026855469, -0.025177002, -0.06732178, -0.012367249, -0.019241333, -0.024017334, 0.014923096, -0.0056114197, -0.010932922, 0.010787964, 0.007843018, -0.012229919, -0.050628662, 0.0016727448, -0.0093307495, -0.0118255615, ...
http://arxiv.org/abs/2308.00002v2
An Overview Of Temporal Commonsense Reasoning and Acquisition
Georg Wenzel, Adam Jatowt
cs.AI, cs.CL, cs.LG, 68T50, I.2.7
Temporal commonsense reasoning refers to the ability to understand the typical temporal context of phrases, actions, and events, and use it to reason over problems requiring such knowledge. This trait is essential in temporal natural language processing tasks, with possible applications such as timeline summarization, ...
http://arxiv.org/pdf/2308.00002v2
2023-09-29T16:54:37
2023-07-28T01:30:15
{ "title": [ 0.017974854, -0.02217102, -0.036499023, -0.07745361, 0.010269165, -0.021957397, 0.0049362183, 0.021987915, 0.009101868, 0.00762558, -0.026489258, 0.0032520294, -0.029525757, -0.024261475, -0.03692627, -0.007663727, -0.017333984, 0.03...
http://arxiv.org/abs/2307.15217v2
Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Wang, Samuel Marks, Charbel-Raphaël Segerie, Micah Carroll, Andi Peng, Phillip Christoffersen, Mehul Damani, Stewart Slocum, Usman Anwar, Anand Siththaranja...
cs.AI, cs.CL, cs.LG
Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In thi...
http://arxiv.org/pdf/2307.15217v2
2023-09-11T17:25:24
2023-07-27T22:29:25
{ "title": [ 0.0024433136, 0.007785797, 0.0020256042, -0.0031433105, 0.011650085, -0.021865845, 0.032073975, -0.0473938, -0.031677246, 0.02017212, -0.04421997, 0.004169464, -0.044769287, -0.0026454926, -0.039916992, -0.033233643, 0.009292603, 0.0...
http://arxiv.org/abs/2307.15199v2
PromptStyler: Prompt-driven Style Generation for Source-free Domain Generalization
Junhyeong Cho, Gilhyun Nam, Sungyeon Kim, Hunmin Yang, Suha Kwak
cs.CV, cs.AI, cs.CL, cs.LG
In a joint vision-language space, a text feature (e.g., from "a photo of a dog") could effectively represent its relevant image features (e.g., from dog photos). Also, a recent study has demonstrated the cross-modal transferability phenomenon of this joint space. From these observations, we propose PromptStyler which s...
http://arxiv.org/pdf/2307.15199v2
2023-08-15T08:30:45
2023-07-27T21:14:46
{ "title": [ -0.0066947937, -0.010620117, -0.059326172, -0.06866455, 0.0021266937, -0.011634827, -0.046813965, 0.010353088, 0.002046585, 0.033721924, 0.003622055, 0.028640747, 0.013320923, -0.021484375, -0.009483337, 0.019378662, -0.0009264946, 0...
http://arxiv.org/abs/2307.15190v1
f-Divergence Minimization for Sequence-Level Knowledge Distillation
Yuqiao Wen, Zichao Li, Wenyu Du, Lili Mou
cs.CL, cs.LG, I.2.7; I.2.6; I.2.m; I.5.1; I.7.m
Knowledge distillation (KD) is the process of transferring knowledge from a large model to a small one. It has gained increasing attention in the natural language processing community, driven by the demands of compressing ever-growing language models. In this work, we propose an f-DISTILL framework, which formulates se...
http://arxiv.org/pdf/2307.15190v1
2023-07-27T20:39:06
2023-07-27T20:39:06
{ "title": [ -0.020187378, 0.006263733, -0.09460449, -0.09991455, 0.018249512, -0.006755829, -0.07232666, 0.032226562, 0.032440186, 0.03173828, -0.052856445, -0.0010080338, 0.0029792786, 0.0056762695, -0.051574707, -0.018081665, 0.0126953125, 0.0...
http://arxiv.org/abs/2307.15176v1
RCT Rejection Sampling for Causal Estimation Evaluation
Katherine A. Keith, Sergey Feldman, David Jurgens, Jonathan Bragg, Rohit Bhattacharya
cs.AI, cs.CL, cs.LG, stat.ME
Confounding is a significant obstacle to unbiased estimation of causal effects from observational data. For settings with high-dimensional covariates -- such as text data, genomics, or the behavioral social sciences -- researchers have proposed methods to adjust for confounding by adapting machine learning methods to t...
http://arxiv.org/pdf/2307.15176v1
2023-07-27T20:11:07
2023-07-27T20:11:07
{ "title": [ -0.008529663, -0.017288208, 0.0020885468, -0.042510986, -0.018951416, 0.0010843277, -0.02128601, -0.024765015, -0.025390625, -0.0015001297, -0.015136719, 0.0124435425, -0.02507019, -0.053009033, 0.006706238, -0.011711121, -0.026824951, ...
http://arxiv.org/abs/2307.15164v1
VISU at WASSA 2023 Shared Task: Detecting Emotions in Reaction to News Stories Leveraging BERT and Stacked Embeddings
Vivek Kumar, Sushmita Singh, Prayag Tiwari
cs.CL, cs.AI, cs.LG
Our system, VISU, participated in the WASSA 2023 Shared Task (3) of Emotion Classification from essays written in reaction to news articles. Emotion detection from complex dialogues is challenging and often requires context/domain understanding. Therefore in this research, we have focused on developing deep learning (D...
http://arxiv.org/pdf/2307.15164v1
2023-07-27T19:42:22
2023-07-27T19:42:22
{ "title": [ 0.0065727234, -0.021133423, -0.03515625, 0.006706238, -0.030639648, -0.039001465, 0.0000042915344, -0.000995636, 0.027435303, 0.018630981, -0.0385437, -0.02192688, 0.035888672, -0.021957397, -0.0057258606, -0.010986328, 0.009994507, ...
http://arxiv.org/abs/2308.07931v1
Distilled Feature Fields Enable Few-Shot Language-Guided Manipulation
William Shen, Ge Yang, Alan Yu, Jansen Wong, Leslie Pack Kaelbling, Phillip Isola
cs.CV, cs.AI, cs.CL, cs.LG, cs.RO
Self-supervised and language-supervised image models contain rich knowledge of the world that is important for generalization. Many robotic tasks, however, require a detailed understanding of 3D geometry, which is often lacking in 2D image features. This work bridges this 2D-to-3D gap for robotic manipulation by levera...
http://arxiv.org/pdf/2308.07931v1
2023-07-27T17:59:14
2023-07-27T17:59:14
{ "title": [ 0.023239136, -0.047302246, -0.02293396, -0.06652832, -0.06161499, -0.009292603, -0.06512451, 0.0068092346, 0.008361816, -0.0050621033, -0.054992676, 0.0014886856, -0.0062713623, -0.039123535, 0.013954163, -0.050811768, 0.0075569153, ...
http://arxiv.org/abs/2307.15054v2
A Geometric Notion of Causal Probing
Clément Guerner, Anej Svete, Tianyu Liu, Alexander Warstadt, Ryan Cotterell
cs.CL
Large language models rely on real-valued representations of text to make their predictions. These representations contain information learned from the data that the model has trained on, including knowledge of linguistic properties and forms of demographic bias, e.g., based on gender. A growing body of work has consid...
http://arxiv.org/pdf/2307.15054v2
2023-07-30T14:22:07
2023-07-27T17:57:57
{ "title": [ 0.0064315796, -0.0124053955, -0.007801056, -0.06604004, -0.017532349, -0.008735657, -0.010864258, 0.0040359497, -0.020599365, 0.0062675476, 0.005607605, -0.027770996, -0.01146698, -0.054840088, -0.0038700104, -0.0062026978, 0.015296936, ...
http://arxiv.org/abs/2307.15051v2
Matching Patients to Clinical Trials with Large Language Models
Qiao Jin, Zifeng Wang, Charalampos S. Floudas, Jimeng Sun, Zhiyong Lu
cs.CL, cs.AI
Clinical trials are vital in advancing drug development and evidence-based medicine, but their success is often hindered by challenges in patient recruitment. In this work, we investigate the potential of large language models (LLMs) to assist individual patients and referral physicians in identifying suitable clinical...
http://arxiv.org/pdf/2307.15051v2
2023-07-28T14:45:50
2023-07-27T17:56:56
{ "title": [ -0.024642944, 0.0062179565, -0.036895752, -0.049957275, -0.00015628338, 0.009262085, -0.05319214, -0.024963379, -0.005622864, 0.03253174, -0.021408081, 0.008483887, -0.017303467, -0.012771606, -0.02986145, -0.0016956329, -0.012916565, ...
http://arxiv.org/abs/2307.15043v1
Universal and Transferable Adversarial Attacks on Aligned Language Models
Andy Zou, Zifan Wang, J. Zico Kolter, Matt Fredrikson
cs.CL, cs.AI, cs.CR, cs.LG
Because "out-of-the-box" large language models are capable of generating a great deal of objectionable content, recent work has focused on aligning these models in an attempt to prevent undesirable generation. While there has been some success at circumventing these measures -- so-called "jailbreaks" against LLMs -- th...
http://arxiv.org/pdf/2307.15043v1
2023-07-27T17:49:12
2023-07-27T17:49:12
{ "title": [ -0.02017212, -0.010238647, -0.031402588, -0.053833008, 0.0041923523, 0.0052337646, -0.040771484, 0.0052719116, -0.033416748, 0.0440979, 0.018493652, 0.022079468, -0.012084961, -0.020889282, -0.03540039, 0.007789612, -0.011238098, 0.0...
http://arxiv.org/abs/2307.15020v1
SuperCLUE: A Comprehensive Chinese Large Language Model Benchmark
Liang Xu, Anqi Li, Lei Zhu, Hang Xue, Changtai Zhu, Kangkang Zhao, Haonan He, Xuanwei Zhang, Qiyue Kang, Zhenzhong Lan
cs.CL, cs.AI
Large language models (LLMs) have shown the potential to be integrated into human daily lives. Therefore, user preference is the most critical criterion for assessing LLMs' performance in real-world scenarios. However, existing benchmarks mainly focus on measuring models' accuracy using multi-choice questions, which li...
http://arxiv.org/pdf/2307.15020v1
2023-07-27T17:24:09
2023-07-27T17:24:09
{ "title": [ -0.039642334, -0.012878418, -0.03527832, -0.07269287, 0.0065231323, 0.012588501, -0.06335449, 0.010978699, 0.008811951, 0.03552246, -0.0057373047, -0.0066070557, 0.018600464, -0.02772522, -0.047424316, 0.016677856, -0.01687622, 0.023...
http://arxiv.org/abs/2307.15002v5
Gzip versus bag-of-words for text classification
Juri Opitz
cs.CL
The effectiveness of compression in text classification ('gzip') has recently garnered lots of attention. In this note we show that `bag-of-words' approaches can achieve similar or better results, and are more efficient.
http://arxiv.org/pdf/2307.15002v5
2023-08-08T17:39:57
2023-07-27T16:57:32
{ "title": [ -0.023651123, 0.0023536682, -0.039215088, -0.040283203, 0.002626419, -0.0017204285, -0.015289307, -0.010002136, -0.005115509, 0.024780273, -0.041381836, 0.007789612, -0.016937256, -0.007484436, -0.04373169, -0.014846802, -0.00623703, ...
http://arxiv.org/abs/2307.14995v1
Scaling TransNormer to 175 Billion Parameters
Zhen Qin, Dong Li, Weigao Sun, Weixuan Sun, Xuyang Shen, Xiaodong Han, Yunshen Wei, Baohong Lv, Fei Yuan, Xiao Luo, Yu Qiao, Yiran Zhong
cs.CL
We present TransNormerLLM, the first linear attention-based Large Language Model (LLM) that outperforms conventional softmax attention-based models in terms of both accuracy and efficiency. TransNormerLLM evolves from the previous linear attention architecture TransNormer by making advanced modifications that include p...
http://arxiv.org/pdf/2307.14995v1
2023-07-27T16:45:33
2023-07-27T16:45:33
{ "title": [ -0.00018680096, -0.010757446, -0.05303955, -0.04537964, -0.02407837, -0.015213013, -0.057159424, 0.00724411, -0.043823242, 0.028076172, -0.01838684, -0.010864258, -0.00756073, -0.050872803, 0.026611328, -0.021652222, 0.0236969, 0.006...
http://arxiv.org/abs/2307.14988v1
Incrementally-Computable Neural Networks: Efficient Inference for Dynamic Inputs
Or Sharir, Anima Anandkumar
cs.LG, cs.CL, stat.ML
Deep learning often faces the challenge of efficiently processing dynamic inputs, such as sensor data or user inputs. For example, an AI writing assistant is required to update its suggestions in real time as a document is edited. Re-running the model each time is expensive, even with compression techniques like knowle...
http://arxiv.org/pdf/2307.14988v1
2023-07-27T16:30:27
2023-07-27T16:30:27
{ "title": [ -0.009223938, -0.007724762, -0.041046143, -0.07659912, 0.023666382, -0.0034065247, -0.025726318, -0.0010471344, 0.014968872, 0.00095558167, -0.034088135, 0.004512787, -0.047180176, -0.009277344, -0.04156494, -0.0054779053, 0.007827759, ...
http://arxiv.org/abs/2307.14936v1
PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback
Bo Shen, Jiaxin Zhang, Taihong Chen, Daoguang Zan, Bing Geng, An Fu, Muhan Zeng, Ailun Yu, Jichuan Ji, Jingyang Zhao, Yuenan Guo, Qianxiang Wang
cs.CL, cs.AI, cs.LG, cs.PL, cs.SE
Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed to boost the code generation performance of pre-trained Code LLMs, such as supervised fine-tuning, in...
http://arxiv.org/pdf/2307.14936v1
2023-07-27T15:28:29
2023-07-27T15:28:29
{ "title": [ -0.039398193, -0.009994507, -0.040100098, -0.060943604, 0.013458252, 0.011138916, -0.07183838, -0.0038642883, 0.009536743, 0.02180481, -0.012611389, 0.004447937, 0.0075531006, -0.018234253, -0.032958984, 0.013122559, -0.020629883, 0....
http://arxiv.org/abs/2307.14913v1
ARC-NLP at PAN 2023: Transition-Focused Natural Language Inference for Writing Style Detection
Izzet Emre Kucukkaya, Umitcan Sahin, Cagri Toraman
cs.CL
The task of multi-author writing style detection aims at finding any positions of writing style change in a given text document. We formulate the task as a natural language inference problem where two consecutive paragraphs are paired. Our approach focuses on transitions between paragraphs while truncating input tokens...
http://arxiv.org/pdf/2307.14913v1
2023-07-27T14:56:06
2023-07-27T14:56:06
{ "title": [ -0.05331421, -0.016967773, -0.067871094, -0.01802063, -0.016357422, -0.012954712, -0.04776001, 0.00043535233, -0.0015907288, 0.029403687, -0.02168274, 0.013542175, 0.020584106, -0.0068092346, -0.013710022, 0.0121536255, -0.018508911, ...
http://arxiv.org/abs/2307.14912v1
ARC-NLP at PAN 2023: Hierarchical Long Text Classification for Trigger Detection
Umitcan Sahin, Izzet Emre Kucukkaya, Cagri Toraman
cs.CL, cs.SI
Fanfiction, a popular form of creative writing set within established fictional universes, has gained a substantial online following. However, ensuring the well-being and safety of participants has become a critical concern in this community. The detection of triggering content, material that may cause emotional distre...
http://arxiv.org/pdf/2307.14912v1
2023-07-27T14:55:10
2023-07-27T14:55:10
{ "title": [ 0.008644104, -0.0006785393, -0.043060303, -0.089782715, 0.018203735, -0.009239197, -0.047058105, 0.00248909, 0.0009431839, 0.010292053, -0.018463135, 0.008651733, 0.014343262, -0.024795532, -0.015945435, 0.009460449, -0.02482605, 0.0...
http://arxiv.org/abs/2307.14899v2
Retrieval-based Text Selection for Addressing Class-Imbalanced Data in Classification
Sareh Ahmadi, Aditya Shah, Edward Fox
cs.CL
This paper addresses the problem of selecting of a set of texts for annotation in text classification using retrieval methods when there are limits on the number of annotations due to constraints on human resources. An additional challenge addressed is dealing with binary categories that have a small number of positive...
http://arxiv.org/pdf/2307.14899v2
2023-11-09T19:39:23
2023-07-27T14:42:16
{ "title": [ -0.014549255, 0.008331299, -0.03262329, -0.01939392, 0.016693115, -0.013923645, -0.0018863678, -0.036346436, 0.012489319, 0.02999878, -0.044830322, 0.01927185, -0.005634308, -0.003162384, -0.056732178, -0.019088745, -0.032043457, 0.0...
http://arxiv.org/abs/2307.14878v1
MESED: A Multi-modal Entity Set Expansion Dataset with Fine-grained Semantic Classes and Hard Negative Entities
Yangning Li, Tingwei Lu, Yinghui Li, Tianyu Yu, Shulin Huang, Hai-Tao Zheng, Rui Zhang, Jun Yuan
cs.CL
The Entity Set Expansion (ESE) task aims to expand a handful of seed entities with new entities belonging to the same semantic class. Conventional ESE methods are based on mono-modality (i.e., literal modality), which struggle to deal with complex entities in the real world such as: (1) Negative entities with fine-grai...
http://arxiv.org/pdf/2307.14878v1
2023-07-27T14:09:59
2023-07-27T14:09:59
{ "title": [ 0.012451172, -0.001616478, -0.06088257, -0.055755615, 0.027053833, -0.035705566, -0.026901245, -0.015319824, 0.040863037, 0.020309448, -0.044647217, 0.02960205, 0.015823364, -0.009590149, -0.027954102, 0.017791748, -0.0054779053, 0.0...
http://arxiv.org/abs/2307.15097v1
Cascaded Cross-Modal Transformer for Request and Complaint Detection
Nicolae-Catalin Ristea, Radu Tudor Ionescu
cs.CL, cs.LG, cs.MM, eess.AS
We propose a novel cascaded cross-modal transformer (CCMT) that combines speech and text transcripts to detect customer requests and complaints in phone conversations. Our approach leverages a multimodal paradigm by transcribing the speech using automatic speech recognition (ASR) models and translating the transcripts ...
http://arxiv.org/pdf/2307.15097v1
2023-07-27T13:45:42
2023-07-27T13:45:42
{ "title": [ 0.027893066, 0.024749756, -0.039001465, -0.051971436, 0.025726318, -0.018508911, -0.031051636, -0.03555298, 0.017028809, 0.039093018, -0.039520264, -0.01285553, 0.055541992, -0.027786255, -0.024810791, 0.008918762, -0.01751709, 0.075...
http://arxiv.org/abs/2307.14856v1
Exploiting the Potential of Seq2Seq Models as Robust Few-Shot Learners
Jihyeon Lee, Dain Kim, Doohae Jung, Boseop Kim, Kyoung-Woon On
cs.CL, cs.AI
In-context learning, which offers substantial advantages over fine-tuning, is predominantly observed in decoder-only models, while encoder-decoder (i.e., seq2seq) models excel in methods that rely on weight updates. Recently, a few studies have demonstrated the feasibility of few-shot learning with seq2seq models; howe...
http://arxiv.org/pdf/2307.14856v1
2023-07-27T13:37:06
2023-07-27T13:37:06
{ "title": [ 0.014045715, 0.00081539154, -0.02809143, -0.015068054, 0.00085544586, -0.025283813, -0.012893677, -0.036071777, 0.0012865067, 0.011016846, -0.023010254, 0.0014667511, -0.03161621, -0.03439331, -0.041931152, -0.023406982, 0.004447937, ...
http://arxiv.org/abs/2307.14852v1
ArcGPT: A Large Language Model Tailored for Real-world Archival Applications
Shitou Zhang, Jingrui Hou, Siyuan Peng, Zuchao Li, Qibiao Hu, Ping Wang
cs.CL
Archives play a crucial role in preserving information and knowledge, and the exponential growth of such data necessitates efficient and automated tools for managing and utilizing archive information resources. Archival applications involve managing massive data that are challenging to process and analyze. Although LLM...
http://arxiv.org/pdf/2307.14852v1
2023-07-27T13:31:45
2023-07-27T13:31:45
{ "title": [ -0.034576416, -0.014190674, -0.0513916, -0.06616211, 0.00932312, 0.00077295303, -0.055755615, 0.014129639, 0.034576416, 0.034088135, -0.0020771027, -0.013572693, 0.03173828, -0.023132324, -0.044158936, 0.03265381, -0.008087158, 0.041...
http://arxiv.org/abs/2307.14850v4
Turkish Native Language Identification
Ahmet Yavuz Uluslu, Gerold Schneider
cs.CL
In this paper, we present the first application of Native Language Identification (NLI) for the Turkish language. NLI involves predicting the writer's first language by analysing their writing in different languages. While most NLI research has focused on English, our study extends its scope to Turkish. We used the rec...
http://arxiv.org/pdf/2307.14850v4
2023-11-04T11:23:55
2023-07-27T13:28:31
{ "title": [ 0.0368042, 0.03466797, -0.023452759, -0.005065918, -0.026123047, -0.02760315, -0.012191772, -0.025390625, -0.052368164, 0.04714966, 0.010757446, -0.021621704, 0.018951416, -0.052093506, 0.025817871, -0.014434814, 0.023727417, 0.03320...
http://arxiv.org/abs/2307.14818v1
What Makes a Good Paraphrase: Do Automated Evaluations Work?
Anna Moskvina, Bhushan Kotnis, Chris Catacata, Michael Janz, Nasrin Saef
cs.CL
Paraphrasing is the task of expressing an essential idea or meaning in different words. But how different should the words be in order to be considered an acceptable paraphrase? And can we exclusively use automated metrics to evaluate the quality of a paraphrase? We attempt to answer these questions by conducting exper...
http://arxiv.org/pdf/2307.14818v1
2023-07-27T12:51:16
2023-07-27T12:51:16
{ "title": [ 0.0061073303, -0.00010532141, 0.026321411, -0.014846802, -0.032562256, 0.02104187, -0.020645142, -0.048706055, -0.041625977, 0.059265137, -0.014137268, 0.006511688, -0.016433716, -0.0087890625, -0.017288208, -0.026489258, -0.019607544, ...
http://arxiv.org/abs/2307.14817v1
Models of reference production: How do they withstand the test of time?
Fahime Same, Guanyi Chen, Kees van Deemter
cs.CL
In recent years, many NLP studies have focused solely on performance improvement. In this work, we focus on the linguistic and scientific aspects of NLP. We use the task of generating referring expressions in context (REG-in-context) as a case study and start our analysis from GREC, a comprehensive set of shared tasks ...
http://arxiv.org/pdf/2307.14817v1
2023-07-27T12:46:38
2023-07-27T12:46:38
{ "title": [ -0.027740479, 0.016235352, -0.06427002, -0.06677246, 0.014915466, -0.0096588135, -0.062683105, -0.013031006, -0.022735596, 0.04714966, 0.0002746582, -0.011383057, -0.0029563904, -0.028686523, -0.048675537, -0.0007157326, 0.0002579689, ...
http://arxiv.org/abs/2307.14785v1
Improving Aspect-Based Sentiment with End-to-End Semantic Role Labeling Model
Pavel Přibáň, Ondřej Pražák
cs.CL
This paper presents a series of approaches aimed at enhancing the performance of Aspect-Based Sentiment Analysis (ABSA) by utilizing extracted semantic information from a Semantic Role Labeling (SRL) model. We propose a novel end-to-end Semantic Role Labeling model that effectively captures most of the structured seman...
http://arxiv.org/pdf/2307.14785v1
2023-07-27T11:28:16
2023-07-27T11:28:16
{ "title": [ 0.013298035, -0.0066108704, -0.026504517, -0.089538574, 0.0043907166, 0.016418457, -0.043701172, 0.015419006, 0.005367279, 0.044952393, 0.0072898865, -0.008720398, -0.004585266, -0.013244629, -0.068115234, 0.0052108765, -0.018814087, ...
http://arxiv.org/abs/2307.14783v1
Emotion4MIDI: a Lyrics-based Emotion-Labeled Symbolic Music Dataset
Serkan Sulun, Pedro Oliveira, Paula Viana
eess.AS, cs.AI, cs.CL, cs.LG, cs.MM
We present a new large-scale emotion-labeled symbolic music dataset consisting of 12k MIDI songs. To create this dataset, we first trained emotion classification models on the GoEmotions dataset, achieving state-of-the-art results with a model half the size of the baseline. We then applied these models to lyrics from t...
http://arxiv.org/pdf/2307.14783v1
2023-07-27T11:24:47
2023-07-27T11:24:47
{ "title": [ -0.008598328, -0.019454956, -0.040130615, -0.045318604, -0.0066947937, -0.023422241, -0.0068092346, 0.013786316, 0.0044822693, 0.029281616, -0.031982422, 0.0058517456, -0.0064811707, -0.012237549, -0.047943115, -0.0064735413, -0.00019454...
http://arxiv.org/abs/2307.16732v1
LLMediator: GPT-4 Assisted Online Dispute Resolution
Hannes Westermann, Jaromir Savelka, Karim Benyekhlef
cs.CL, cs.AI, cs.CY
In this article, we introduce LLMediator, an experimental platform designed to enhance online dispute resolution (ODR) by utilizing capabilities of state-of-the-art large language models (LLMs) such as GPT-4. In the context of high-volume, low-intensity legal disputes, alternative dispute resolution methods such as neg...
http://arxiv.org/pdf/2307.16732v1
2023-07-27T10:25:29
2023-07-27T10:25:29
{ "title": [ 0.027786255, 0.0078048706, -0.021087646, -0.047454834, -0.0126571655, -0.0116119385, -0.015159607, 0.00096321106, -0.023635864, 0.019119263, 0.012054443, 0.00340271, 0.04360962, -0.06616211, 0.05279541, -0.0063171387, -0.004764557, 0...
http://arxiv.org/abs/2307.14743v2
Turning Whisper into Real-Time Transcription System
Dominik Macháček, Raj Dabre, Ondřej Bojar
cs.CL
Whisper is one of the recent state-of-the-art multilingual speech recognition and translation models, however, it is not designed for real time transcription. In this paper, we build on top of Whisper and create Whisper-Streaming, an implementation of real-time speech transcription and translation of Whisper-like model...
http://arxiv.org/pdf/2307.14743v2
2023-09-21T09:41:17
2023-07-27T10:00:05
{ "title": [ -0.009857178, -0.044128418, -0.014808655, -0.054382324, -0.08258057, -0.0053520203, -0.064208984, 0.0184021, 0.0044555664, 0.043548584, -0.04611206, -0.023468018, 0.0154418945, -0.01713562, -0.01789856, -0.03213501, 0.016845703, 0.00...
http://arxiv.org/abs/2307.14712v1
Evaluating Generative Models for Graph-to-Text Generation
Shuzhou Yuan, Michael Färber
cs.CL, cs.AI
Large language models (LLMs) have been widely employed for graph-to-text generation tasks. However, the process of finetuning LLMs requires significant training resources and annotation work. In this paper, we explore the capability of generative models to generate descriptive text from graph data in a zero-shot settin...
http://arxiv.org/pdf/2307.14712v1
2023-07-27T09:03:05
2023-07-27T09:03:05
{ "title": [ -0.038024902, -0.017059326, -0.030441284, -0.046966553, -0.0209198, 0.019638062, -0.07006836, -0.0023822784, 0.011383057, 0.03390503, -0.0129776, 0.026931763, -0.013557434, -0.013244629, -0.047058105, 0.008865356, -0.015914917, 0.043...
http://arxiv.org/abs/2307.14666v1
Improving Natural Language Inference in Arabic using Transformer Models and Linguistically Informed Pre-Training
Mohammad Majd Saad Al Deen, Maren Pielka, Jörn Hees, Bouthaina Soulef Abdou, Rafet Sifa
cs.CL
This paper addresses the classification of Arabic text data in the field of Natural Language Processing (NLP), with a particular focus on Natural Language Inference (NLI) and Contradiction Detection (CD). Arabic is considered a resource-poor language, meaning that there are few data sets available, which leads to limit...
http://arxiv.org/pdf/2307.14666v1
2023-07-27T07:40:11
2023-07-27T07:40:11
{ "title": [ -0.018951416, -0.022460938, -0.020751953, -0.07305908, -0.020828247, 0.011070251, -0.062805176, 0.0113220215, -0.015808105, -0.0020542145, -0.025817871, -0.0023422241, 0.005065918, -0.03665161, 0.00049495697, -0.0129470825, 0.00089645386...
http://arxiv.org/abs/2307.14632v1
Metric-Based In-context Learning: A Case Study in Text Simplification
Subha Vadlamannati, Gözde Gül Şahin
cs.CL, cs.AI
In-context learning (ICL) for large language models has proven to be a powerful approach for many natural language processing tasks. However, determining the best method to select examples for ICL is nontrivial as the results can vary greatly depending on the quality, quantity, and order of examples used. In this paper...
http://arxiv.org/pdf/2307.14632v1
2023-07-27T05:45:35
2023-07-27T05:45:35
{ "title": [ -0.023345947, -0.016693115, -0.052581787, -0.04171753, -0.0037498474, -0.01499176, -0.028152466, 0.012931824, -0.0032653809, 0.07556152, -0.013198853, 0.03201294, 0.018722534, 0.00025057793, -0.049926758, -0.009918213, 0.0116119385, ...
http://arxiv.org/abs/2307.14544v1
Speed Reading Tool Powered by Artificial Intelligence for Students with ADHD, Dyslexia, or Short Attention Span
Megat Irfan Zackry Bin Ismail Ahmad Nazran bin Yusri Muhammad Hafizzul Bin Abdul Manap Muhammad Muizzuddin Bin Kamarozaman
cs.CL, cs.AI
This paper presents a novel approach to assist students with dyslexia, ADHD, and short attention span in digesting any text-based information more efficiently. The proposed solution utilizes the Multilayer Perceptron (MLP) algorithm for complex text processing and summarization tasks. The tool leverages the T5 (Text-to...
http://arxiv.org/pdf/2307.14544v1
2023-07-26T23:47:14
2023-07-26T23:47:14
{ "title": [ -0.018875122, -0.013328552, -0.019302368, -0.0050086975, -0.019317627, -0.0211792, 0.005783081, 0.0008597374, 0.009666443, 0.026809692, -0.026748657, -0.010192871, 0.008651733, -0.025405884, -0.0025863647, -0.03137207, 0.007347107, 0...
http://arxiv.org/abs/2307.14539v2
Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models
Erfan Shayegani, Yue Dong, Nael Abu-Ghazaleh
cs.CR, cs.CL
We introduce new jailbreak attacks on vision language models (VLMs), which use aligned LLMs and are resilient to text-only jailbreak attacks. Specifically, we develop cross-modality attacks on alignment where we pair adversarial images going through the vision encoder with textual prompts to break the alignment of the ...
http://arxiv.org/pdf/2307.14539v2
2023-10-10T22:17:17
2023-07-26T23:11:15
{ "title": [ -0.011024475, -0.0040779114, -0.031799316, -0.07110596, 0.0129852295, 0.0076560974, -0.06732178, -0.00016736984, -0.04260254, 0.03933716, 0.031921387, 0.017120361, -0.00198555, -0.034118652, -0.035125732, 0.00089263916, -0.022079468, ...
http://arxiv.org/abs/2307.14522v2
CliniDigest: A Case Study in Large Language Model Based Large-Scale Summarization of Clinical Trial Descriptions
Renee D. White, Tristan Peng, Pann Sripitak, Alexander Rosenberg Johansen, Michael Snyder
cs.CL
A clinical trial is a study that evaluates new biomedical interventions. To design new trials, researchers draw inspiration from those current and completed. In 2022, there were on average more than 100 clinical trials submitted to ClinicalTrials.gov every day, with each trial having a mean of approximately 1500 words ...
http://arxiv.org/pdf/2307.14522v2
2023-07-31T19:00:05
2023-07-26T21:49:14
{ "title": [ -0.036102295, -0.014526367, -0.05340576, -0.08227539, -0.023010254, 0.017089844, -0.10437012, -0.001282692, -0.008079529, 0.048034668, 0.00033521652, 0.015106201, -0.0011024475, -0.01096344, -0.051513672, 0.012893677, -0.008323669, 0...
http://arxiv.org/abs/2307.14511v1
Words That Stick: Predicting Decision Making and Synonym Engagement Using Cognitive Biases and Computational Linguistics
Nimrod Dvir, Elaine Friedman, Suraj Commuri, Fan Yang, Jennifer Romano
cs.HC, cs.AI, cs.CL, 03B65, H.5; I.7
This research draws upon cognitive psychology and information systems studies to anticipate user engagement and decision-making on digital platforms. By employing natural language processing (NLP) techniques and insights from cognitive bias research, we delve into user interactions with synonyms within digital content....
http://arxiv.org/pdf/2307.14511v1
2023-07-26T21:20:03
2023-07-26T21:20:03
{ "title": [ -0.011672974, -0.010505676, -0.036132812, -0.015579224, -0.0060806274, -0.023284912, 0.0023899078, -0.015274048, -0.0007786751, 0.036987305, -0.018066406, 0.00415802, 0.022384644, -0.014221191, -0.03869629, 0.0027160645, -0.018936157, ...
http://arxiv.org/abs/2307.14500v1
A Predictive Model of Digital Information Engagement: Forecasting User Engagement With English Words by Incorporating Cognitive Biases, Computational Linguistics and Natural Language Processing
Nimrod Dvir, Elaine Friedman, Suraj Commuri, Fan yang, Jennifer Romano
cs.HC, cs.CL, cs.LG, 68U15, H.5; H.5.1; H.5.2; D.3.2
This study introduces and empirically tests a novel predictive model for digital information engagement (IE) - the READ model, an acronym for the four pivotal attributes of engaging information: Representativeness, Ease-of-use, Affect, and Distribution. Conceptualized within the theoretical framework of Cumulative Pros...
http://arxiv.org/pdf/2307.14500v1
2023-07-26T20:58:47
2023-07-26T20:58:47
{ "title": [ 0.022888184, 0.017120361, -0.015991211, -0.01499176, -0.023132324, -0.028625488, -0.03048706, -0.02722168, 0.027832031, 0.05142212, -0.027328491, -0.03717041, 0.034973145, -0.029281616, -0.05947876, -0.01033783, -0.013999939, 0.05435...
http://arxiv.org/abs/2308.00109v1
A Sentence is Worth a Thousand Pictures: Can Large Language Models Understand Human Language?
Gary Marcus, Evelina Leivada, Elliot Murphy
cs.CL
Artificial Intelligence applications show great potential for language-related tasks that rely on next-word prediction. The current generation of large language models have been linked to claims about human-like linguistic performance and their applications are hailed both as a key step towards Artificial General Intel...
http://arxiv.org/pdf/2308.00109v1
2023-07-26T18:58:53
2023-07-26T18:58:53
{ "title": [ -0.0064201355, 0.0032482147, -0.028045654, -0.0317688, -0.018798828, -0.022842407, -0.03765869, -0.021011353, -0.0026721954, 0.015151978, -0.03427124, -0.010917664, 0.014213562, -0.029922485, -0.032592773, -0.02003479, -0.0003848076, ...
http://arxiv.org/abs/2307.14440v1
Controllable Generation of Dialogue Acts for Dialogue Systems via Few-Shot Response Generation and Ranking
Angela Ramirez, Karik Agarwal, Juraj Juraska, Utkarsh Garg, Marilyn A. Walker
cs.CL
Dialogue systems need to produce responses that realize multiple types of dialogue acts (DAs) with high semantic fidelity. In the past, natural language generators (NLGs) for dialogue were trained on large parallel corpora that map from a domain-specific DA and its semantic attributes to an output utterance. Recent wor...
http://arxiv.org/pdf/2307.14440v1
2023-07-26T18:16:45
2023-07-26T18:16:45
{ "title": [ 0.0002361536, -0.01878357, -0.028564453, -0.036315918, -0.0052833557, -0.0005507469, -0.019165039, 0.008224487, -0.0027885437, 0.019073486, -0.009796143, 0.008918762, -0.040100098, -0.015434265, -0.06555176, -0.02319336, -0.0055160522, ...
http://arxiv.org/abs/2307.14430v1
Skill-it! A Data-Driven Skills Framework for Understanding and Training Language Models
Mayee F. Chen, Nicholas Roberts, Kush Bhatia, Jue Wang, Ce Zhang, Frederic Sala, Christopher Ré
cs.CL, cs.LG
The quality of training data impacts the performance of pre-trained large language models (LMs). Given a fixed budget of tokens, we study how to best select data that leads to good downstream model performance across tasks. We develop a new framework based on a simple hypothesis: just as humans acquire interdependent s...
http://arxiv.org/pdf/2307.14430v1
2023-07-26T18:01:49
2023-07-26T18:01:49
{ "title": [ -0.022247314, -0.013305664, -0.049957275, -0.05621338, -0.011230469, -0.01586914, -0.04244995, 0.006046295, -0.010597229, 0.01828003, -0.008140564, 0.0040397644, 0.014419556, -0.028274536, -0.025939941, -0.0039596558, -0.012359619, 0...
http://arxiv.org/abs/2308.00113v2
Three Bricks to Consolidate Watermarks for Large Language Models
Pierre Fernandez, Antoine Chaffin, Karim Tit, Vivien Chappelier, Teddy Furon
cs.CL, cs.AI
The task of discerning between generated and natural texts is increasingly challenging. In this context, watermarking emerges as a promising technique for ascribing generated text to a specific model. It alters the sampling generation process so as to leave an invisible trace in the generated output, facilitating later...
http://arxiv.org/pdf/2308.00113v2
2023-11-08T18:56:19
2023-07-26T17:56:36
{ "title": [ 0.008224487, -0.005508423, -0.0736084, -0.08355713, -0.02558899, -0.024124146, -0.06982422, 0.036865234, -0.029815674, 0.061340332, -0.0099105835, 0.0211792, 0.047821045, -0.03012085, 0.006275177, 0.01020813, 0.018234253, 0.00894165,...
http://arxiv.org/abs/2307.14334v1
Towards Generalist Biomedical AI
Tao Tu, Shekoofeh Azizi, Danny Driess, Mike Schaekermann, Mohamed Amin, Pi-Chuan Chang, Andrew Carroll, Chuck Lau, Ryutaro Tanno, Ira Ktena, Basil Mustafa, Aakanksha Chowdhery, Yun Liu, Simon Kornblith, David Fleet, Philip Mansfield, Sushant Prakash, Renee Wong, Sunny Virmani, Christopher Semturs, S Sara Mahdavi, Bradl...
cs.CL, cs.CV
Medicine is inherently multimodal, with rich data modalities spanning text, imaging, genomics, and more. Generalist biomedical artificial intelligence (AI) systems that flexibly encode, integrate, and interpret this data at scale can potentially enable impactful applications ranging from scientific discovery to care de...
http://arxiv.org/pdf/2307.14334v1
2023-07-26T17:52:22
2023-07-26T17:52:22
{ "title": [ -0.017028809, -0.004589081, -0.020812988, -0.04248047, 0.0031909943, 0.0012226105, -0.01725769, -0.010116577, 0.010612488, 0.029342651, -0.011665344, -0.005065918, -0.008155823, -0.031463623, -0.041809082, 0.0021858215, 0.009086609, ...
http://arxiv.org/abs/2307.15717v1
Utilizing Large Language Models for Natural Interface to Pharmacology Databases
Hong Lu, Chuan Li, Yinheng Li, Jie Zhao
cs.CL, cs.AI, cs.DB
The drug development process necessitates that pharmacologists undertake various tasks, such as reviewing literature, formulating hypotheses, designing experiments, and interpreting results. Each stage requires accessing and querying vast amounts of information. In this abstract, we introduce a Large Language Model (LL...
http://arxiv.org/pdf/2307.15717v1
2023-07-26T17:50:11
2023-07-26T17:50:11
{ "title": [ -0.023635864, -0.021911621, -0.029556274, -0.052703857, -0.016357422, 0.004878998, -0.054748535, 0.0063552856, 0.017105103, 0.021102905, -0.008659363, -0.031463623, -0.010314941, -0.03414917, -0.05807495, 0.0028762817, -0.0031719208, ...
http://arxiv.org/abs/2307.14324v1
Evaluating the Moral Beliefs Encoded in LLMs
Nino Scherrer, Claudia Shi, Amir Feder, David M. Blei
cs.CL, cs.AI, cs.CY, cs.LG
This paper presents a case study on the design, administration, post-processing, and evaluation of surveys on large language models (LLMs). It comprises two components: (1) A statistical method for eliciting beliefs encoded in LLMs. We introduce statistical measures and evaluation metrics that quantify the probability ...
http://arxiv.org/pdf/2307.14324v1
2023-07-26T17:42:43
2023-07-26T17:42:43
{ "title": [ 0.004108429, 0.006164551, -0.014282227, 0.015403748, -0.04663086, -0.02999878, 0.0005402565, -0.019958496, -0.03970337, 0.015731812, -0.021133423, -0.025924683, -0.011779785, -0.055267334, 0.018722534, -0.039398193, -0.0059051514, 0....
http://arxiv.org/abs/2307.14311v1
Comparative Analysis of Libraries for the Sentimental Analysis
Wendy Ccoya, Edson Pinto
cs.CL
This study is main goal is to provide a comparative comparison of libraries using machine learning methods. Experts in natural language processing (NLP) are becoming more and more interested in sentiment analysis (SA) of text changes. The objective of employing NLP text analysis techniques is to recognize and categoriz...
http://arxiv.org/pdf/2307.14311v1
2023-07-26T17:21:53
2023-07-26T17:21:53
{ "title": [ 0.0043563843, 0.005302429, -0.018463135, -0.002023697, -0.025344849, -0.023254395, 0.010795593, -0.028686523, -0.049591064, 0.059020996, -0.0009918213, 0.007865906, 0.029632568, -0.043273926, 0.0066833496, -0.025756836, 0.014350891, ...
http://arxiv.org/abs/2307.14305v1
Automatically Evaluating Opinion Prevalence in Opinion Summarization
Christopher Malon
cs.CL
When faced with a large number of product reviews, it is not clear that a human can remember all of them and weight opinions representatively to write a good reference summary. We propose an automatic metric to test the prevalence of the opinions that a summary expresses, based on counting the number of reviews that ar...
http://arxiv.org/pdf/2307.14305v1
2023-07-26T17:13:00
2023-07-26T17:13:00
{ "title": [ 0.017089844, 0.014823914, -0.00995636, -0.059051514, 0.0044937134, -0.000016510487, -0.021865845, -0.036956787, 0.021392822, 0.04660034, -0.044525146, -0.0002939701, -0.015487671, 0.013725281, -0.08935547, -0.007045746, -0.020690918, ...
http://arxiv.org/abs/2307.14298v1
ChatGPT and Persuasive Technologies for the Management and Delivery of Personalized Recommendations in Hotel Hospitality
Manolis Remountakis, Konstantinos Kotis, Babis Kourtzis, George E. Tsekouras
cs.IR, cs.AI, cs.CL, cs.LG, 68T01, I.2.1
Recommender systems have become indispensable tools in the hotel hospitality industry, enabling personalized and tailored experiences for guests. Recent advancements in large language models (LLMs), such as ChatGPT, and persuasive technologies, have opened new avenues for enhancing the effectiveness of those systems. T...
http://arxiv.org/pdf/2307.14298v1
2023-07-26T16:58:10
2023-07-26T16:58:10
{ "title": [ 0.05206299, -0.0060920715, -0.009086609, -0.0068740845, -0.010887146, -0.027908325, 0.01651001, -0.024551392, -0.04876709, 0.02885437, -0.027709961, 0.035980225, -0.0102005005, -0.02532959, 0.0051879883, -0.02243042, -0.013916016, 0....
http://arxiv.org/abs/2307.14291v2
Founding a mathematical diffusion model in linguistics. The case study of German syntactic features in the North-Eastern Italian dialects
I. Lazzizzera
cs.CL
The initial motivation for this work was the linguistic case of the spread of Germanic syntactic features into Romance dialects of North-Eastern Italy, which occurred after the immigration of German people to Tyrol during the High Middle Ages. To obtain a representation of the data over the territory suitable for a mat...
http://arxiv.org/pdf/2307.14291v2
2023-09-16T19:19:01
2023-07-26T16:49:11
{ "title": [ -0.019897461, -0.0043525696, -0.0013990402, 0.06524658, -0.038146973, -0.008911133, 0.0030136108, -0.053863525, 0.030731201, 0.025314331, 0.02571106, -0.01033783, 0.028305054, -0.054382324, -0.009941101, 0.01424408, -0.004512787, 0.0...
http://arxiv.org/abs/2307.16819v1
The flow of ideas in word embeddings
Debayan Dasgupta
cs.CL, cs.AI
The flow of ideas has been extensively studied by physicists, psychologists, and machine learning engineers. This paper adopts specific tools from microrheology to investigate the similarity-based flow of ideas. We introduce a random walker in word embeddings and study its behavior. Such similarity-mediated random walk...
http://arxiv.org/pdf/2307.16819v1
2023-07-26T15:51:31
2023-07-26T15:51:31
{ "title": [ 0.0013504028, -0.014015198, -0.001332283, -0.03665161, -0.018630981, -0.006652832, -0.028900146, -0.024673462, 0.010177612, 0.03741455, -0.03439331, 0.018356323, 0.00038146973, -0.0020751953, -0.032714844, 0.013046265, 0.0035629272, ...
http://arxiv.org/abs/2307.14236v1
UnScientify: Detecting Scientific Uncertainty in Scholarly Full Text
Panggih Kusuma Ningrum, Philipp Mayr, Iana Atanassova
cs.CL, cs.AI, cs.DL
This demo paper presents UnScientify, an interactive system designed to detect scientific uncertainty in scholarly full text. The system utilizes a weakly supervised technique that employs a fine-grained annotation scheme to identify verbally formulated uncertainty at the sentence level in scientific texts. The pipelin...
http://arxiv.org/pdf/2307.14236v1
2023-07-26T15:04:24
2023-07-26T15:04:24
{ "title": [ -0.04547119, 0.012481689, -0.052124023, -0.03137207, 0.010169983, -0.0074310303, -0.02645874, -0.03125, 0.0046539307, 0.008422852, -0.0064430237, 0.0149002075, -0.026412964, -0.021255493, -0.05001831, -0.0071640015, 0.00059747696, 0....
http://arxiv.org/abs/2307.14142v1
LOIS: Looking Out of Instance Semantics for Visual Question Answering
Siyu Zhang, Yeming Chen, Yaoru Sun, Fang Wang, Haibo Shi, Haoran Wang
cs.CV, cs.AI, cs.CL
Visual question answering (VQA) has been intensively studied as a multimodal task that requires effort in bridging vision and language to infer answers correctly. Recent attempts have developed various attention-based modules for solving VQA tasks. However, the performance of model inference is largely bottlenecked by ...
http://arxiv.org/pdf/2307.14142v1
2023-07-26T12:13:00
2023-07-26T12:13:00
{ "title": [ 0.004173279, -0.02758789, 0.0014419556, -0.042388916, -0.019699097, -0.005050659, -0.020004272, -0.015167236, 0.038391113, 0.0015001297, -0.038116455, -0.0005683899, -0.019378662, -0.012054443, -0.048309326, -0.012435913, 0.02331543, ...
http://arxiv.org/abs/2307.14134v1
Developing and Evaluating Tiny to Medium-Sized Turkish BERT Models
Himmet Toprak Kesgin, Muzaffer Kaan Yuce, Mehmet Fatih Amasyali
cs.CL, cs.AI, cs.LG
This study introduces and evaluates tiny, mini, small, and medium-sized uncased Turkish BERT models, aiming to bridge the research gap in less-resourced languages. We trained these models on a diverse dataset encompassing over 75GB of text from multiple sources and tested them on several tasks, including mask predictio...
http://arxiv.org/pdf/2307.14134v1
2023-07-26T12:02:30
2023-07-26T12:02:30
{ "title": [ -0.0004963875, -0.024536133, -0.017059326, -0.030227661, -0.035095215, -0.0041618347, -0.055664062, -0.0018043518, -0.034057617, 0.054595947, 0.0015707016, -0.02229309, 0.033477783, -0.04324341, -0.029159546, 0.003080368, 0.0032787323, ...
http://arxiv.org/abs/2307.14132v2
Say Goodbye to RNN-T Loss: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition
Tian-Hao Zhang, Dinghao Zhou, Guiping Zhong, Baoxiang Li
cs.SD, cs.CL, eess.AS
RNN-T models are widely used in ASR, which rely on the RNN-T loss to achieve length alignment between input audio and target sequence. However, the implementation complexity and the alignment-based optimization target of RNN-T loss lead to computational redundancy and a reduced role for predictor network, respectively....
http://arxiv.org/pdf/2307.14132v2
2023-07-27T00:41:36
2023-07-26T11:59:14
{ "title": [ -0.028015137, -0.014060974, -0.033935547, -0.057617188, -0.018493652, -0.0048294067, -0.03842163, 0.030807495, -0.0036258698, 0.05810547, -0.031204224, 0.018936157, -0.0038452148, 0.0063667297, -0.06774902, -0.010437012, -0.009292603, ...
http://arxiv.org/abs/2307.14117v1
Leveraging Implicit Feedback from Deployment Data in Dialogue
Richard Yuanzhe Pang, Stephen Roller, Kyunghyun Cho, He He, Jason Weston
cs.CL
We study improving social conversational agents by learning from natural dialogue between users and a deployed model, without extra annotations. To implicitly measure the quality of a machine-generated utterance, we leverage signals like user response length, sentiment and reaction of the future human utterances in the...
http://arxiv.org/pdf/2307.14117v1
2023-07-26T11:34:53
2023-07-26T11:34:53
{ "title": [ -0.01687622, -0.0042800903, -0.027359009, -0.053131104, 0.008262634, 0.0015916824, -0.023223877, -0.008628845, 0.014694214, 0.012786865, -0.031402588, 0.00041604042, -0.045074463, -0.014770508, -0.036987305, -0.038391113, -0.0030231476, ...
http://arxiv.org/abs/2307.14107v2
Decoding ChatGPT: A Taxonomy of Existing Research, Current Challenges, and Possible Future Directions
Shahab Saquib Sohail, Faiza Farhat, Yassine Himeur, Mohammad Nadeem, Dag Øivind Madsen, Yashbir Singh, Shadi Atalla, Wathiq Mansoor
cs.CL, cs.CY
Chat Generative Pre-trained Transformer (ChatGPT) has gained significant interest and attention since its launch in November 2022. It has shown impressive performance in various domains, including passing exams and creative writing. However, challenges and concerns related to biases and trust persist. In this work, we ...
http://arxiv.org/pdf/2307.14107v2
2023-08-25T09:00:08
2023-07-26T11:10:04
{ "title": [ -0.017074585, -0.011390686, -0.0059394836, -0.016540527, -0.0034065247, 0.0143585205, -0.023925781, -0.02684021, -0.019927979, 0.02633667, 0.0025730133, 0.030212402, -0.0016117096, -0.019866943, -0.02243042, 0.021697998, -0.018554688, ...
http://arxiv.org/abs/2307.14031v1
Multi3WOZ: A Multilingual, Multi-Domain, Multi-Parallel Dataset for Training and Evaluating Culturally Adapted Task-Oriented Dialog Systems
Songbo Hu, Han Zhou, Mete Hergul, Milan Gritta, Guchun Zhang, Ignacio Iacobacci, Ivan Vulić, Anna Korhonen
cs.CL
Creating high-quality annotated data for task-oriented dialog (ToD) is known to be notoriously difficult, and the challenges are amplified when the goal is to create equitable, culturally adapted, and large-scale ToD datasets for multiple languages. Therefore, the current datasets are still very scarce and suffer from ...
http://arxiv.org/pdf/2307.14031v1
2023-07-26T08:29:42
2023-07-26T08:29:42
{ "title": [ -0.009521484, -0.022079468, -0.04159546, -0.042144775, -0.015205383, -0.024536133, -0.040100098, 0.021636963, -0.0018606186, 0.062805176, 0.005622864, 0.010375977, 0.011100769, -0.008201599, -0.047180176, 0.0070610046, -0.0038719177, ...
http://arxiv.org/abs/2307.14005v1
Unsupervised extraction of local and global keywords from a single text
Lida Aleksanyan, Armen E. Allahverdyan
cs.CL, cs.DL
We propose an unsupervised, corpus-independent method to extract keywords from a single text. It is based on the spatial distribution of words and the response of this distribution to a random permutation of words. As compared to existing methods (such as e.g. YAKE) our method has three advantages. First, it is signifi...
http://arxiv.org/pdf/2307.14005v1
2023-07-26T07:36:25
2023-07-26T07:36:25
{ "title": [ -0.024276733, -0.0014019012, -0.04925537, -0.050476074, 0.0044288635, -0.0025520325, -0.023757935, -0.0041503906, 0.01398468, 0.035980225, -0.01651001, 0.014854431, -0.00242424, -0.00920105, -0.035949707, 0.006549835, -0.0004916191, ...
http://arxiv.org/abs/2307.14004v1
Affective Natural Language Generation of Event Descriptions through Fine-grained Appraisal Conditions
Yarik Menchaca Resendiz, Roman Klinger
cs.CL
Models for affective text generation have shown a remarkable progress, but they commonly rely only on basic emotion theories or valance/arousal values as conditions. This is appropriate when the goal is to create explicit emotion statements ("The kid is happy."). Emotions are, however, commonly communicated implicitly....
http://arxiv.org/pdf/2307.14004v1
2023-07-26T07:34:19
2023-07-26T07:34:19
{ "title": [ 0.008087158, -0.015388489, -0.023406982, -0.07946777, -0.0072517395, 0.0049095154, -0.04348755, -0.0014228821, 0.033721924, 0.02168274, -0.016525269, -0.019180298, -0.0051078796, -0.020690918, -0.07171631, -0.004776001, -0.0069351196, ...
http://arxiv.org/abs/2308.00108v1
DPBERT: Efficient Inference for BERT based on Dynamic Planning
Weixin Wu, Hankz Hankui Zhuo
cs.CL, cs.AI
Large-scale pre-trained language models such as BERT have contributed significantly to the development of NLP. However, those models require large computational resources, making it difficult to be applied to mobile devices where computing power is limited. In this paper we aim to address the weakness of existing input...
http://arxiv.org/pdf/2308.00108v1
2023-07-26T07:18:50
2023-07-26T07:18:50
{ "title": [ -0.017807007, -0.026535034, -0.024673462, -0.09539795, -0.010757446, 0.022994995, -0.06945801, 0.009857178, -0.009544373, 0.008171082, -0.030029297, 0.018798828, -0.0009083748, -0.022003174, 0.0028266907, -0.0033779144, -0.0026741028, ...
http://arxiv.org/abs/2307.14389v1
Diff-E: Diffusion-based Learning for Decoding Imagined Speech EEG
Soowon Kim, Young-Eun Lee, Seo-Hyun Lee, Seong-Whan Lee
eess.AS, cs.CL, cs.HC, cs.LG, 68T10
Decoding EEG signals for imagined speech is a challenging task due to the high-dimensional nature of the data and low signal-to-noise ratio. In recent years, denoising diffusion probabilistic models (DDPMs) have emerged as promising approaches for representation learning in various domains. Our study proposes a novel m...
http://arxiv.org/pdf/2307.14389v1
2023-07-26T07:12:39
2023-07-26T07:12:39
{ "title": [ -0.013137817, 0.000035107136, -0.013916016, -0.0015850067, -0.0028152466, -0.0060043335, 0.007865906, -0.028030396, 0.013710022, 0.045013428, -0.02267456, -0.003917694, 0.046325684, -0.035614014, -0.04559326, 0.007091522, -0.013038635, ...
http://arxiv.org/abs/2308.00072v1
How User Language Affects Conflict Fatality Estimates in ChatGPT
Daniel Kazenwadel, Christoph V. Steinert
cs.CL, cs.AI, cs.HC
OpenAI's ChatGPT language model has gained popularity as a powerful tool for complex problem-solving and information retrieval. However, concerns arise about the reproduction of biases present in the language-specific training data. In this study, we address this issue in the context of the Israeli-Palestinian and Turk...
http://arxiv.org/pdf/2308.00072v1
2023-07-26T07:07:44
2023-07-26T07:07:44
{ "title": [ -0.0061798096, -0.015701294, -0.016799927, -0.04220581, -0.014411926, -0.008102417, -0.03060913, 0.0024642944, -0.03540039, 0.031677246, -0.011978149, 0.040863037, 0.008522034, -0.016326904, -0.011100769, 0.0072746277, -0.014503479, ...
http://arxiv.org/abs/2307.13989v1
This is not correct! Negation-aware Evaluation of Language Generation Systems
Miriam Anschütz, Diego Miguel Lozano, Georg Groh
cs.CL, cs.LG
Large language models underestimate the impact of negations on how much they change the meaning of a sentence. Therefore, learned evaluation metrics based on these models are insensitive to negations. In this paper, we propose NegBLEURT, a negation-aware version of the BLEURT evaluation metric. For that, we designed a ...
http://arxiv.org/pdf/2307.13989v1
2023-07-26T06:54:31
2023-07-26T06:54:31
{ "title": [ -0.02848816, -0.011184692, -0.022018433, -0.04751587, -0.0064582825, 0.004207611, -0.039733887, -0.004524231, -0.005493164, 0.024780273, -0.039764404, 0.0009536743, 0.00434494, -0.013435364, -0.04824829, -0.016998291, -0.024765015, 0...
http://arxiv.org/abs/2307.14385v3
Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text Data
Xuhai Xu, Bingsheng Yao, Yuanzhe Dong, Saadia Gabriel, Hong Yu, James Hendler, Marzyeh Ghassemi, Anind K. Dey, Dakuo Wang
cs.CL, 68U35, H.5.2; I.2.m
Advances in large language models (LLMs) have empowered a variety of applications. However, there is still a significant gap in research when it comes to understanding and enhancing the capabilities of LLMs in the field of mental health. In this work, we present the first comprehensive evaluation of multiple LLMs, incl...
http://arxiv.org/pdf/2307.14385v3
2023-09-16T00:52:41
2023-07-26T06:00:50
{ "title": [ -0.007904053, -0.00774765, -0.03894043, -0.07672119, 0.015449524, 0.008293152, -0.05734253, -0.0010738373, 0.00869751, 0.024124146, -0.011932373, 0.002960205, 0.013748169, -0.023757935, -0.035980225, 0.009132385, -0.0009794235, 0.033...
http://arxiv.org/abs/2307.13949v1
How Does Diffusion Influence Pretrained Language Models on Out-of-Distribution Data?
Huazheng Wang, Daixuan Cheng, Haifeng Sun, Jingyu Wang, Qi Qi, Jianxin Liao, Jing Wang, Cong Liu
cs.CL, cs.AI
Transformer-based pretrained language models (PLMs) have achieved great success in modern NLP. An important advantage of PLMs is good out-of-distribution (OOD) robustness. Recently, diffusion models have attracted a lot of work to apply diffusion to PLMs. It remains under-explored how diffusion influences PLMs on OOD d...
http://arxiv.org/pdf/2307.13949v1
2023-07-26T04:03:25
2023-07-26T04:03:25
{ "title": [ 0.0284729, 0.0013570786, -0.02470398, 0.00390625, -0.013038635, -0.034301758, 0.009162903, -0.011131287, -0.020645142, 0.0126953125, -0.027496338, 0.008110046, -0.014167786, -0.03111267, 0.0129776, -0.026901245, -0.010612488, 0.06695...
http://arxiv.org/abs/2307.13923v2
GrammarGPT: Exploring Open-Source LLMs for Native Chinese Grammatical Error Correction with Supervised Fine-Tuning
Yaxin Fan, Feng Jiang, Peifeng Li, Haizhou Li
cs.CL
Grammatical error correction aims to correct ungrammatical sentences automatically. Recently, some work has demonstrated the excellent capabilities of closed-source Large Language Models (LLMs, e.g., ChatGPT) in grammatical error correction. However, the potential of open-source LLMs remains unexplored. In this paper, ...
http://arxiv.org/pdf/2307.13923v2
2023-08-17T19:58:42
2023-07-26T02:45:38
{ "title": [ -0.02999878, -0.0007953644, -0.019821167, -0.0413208, -0.024734497, 0.011039734, -0.056488037, -0.0063934326, -0.023406982, 0.025894165, -0.044708252, 0.029281616, 0.009223938, -0.022003174, -0.0088272095, -0.010902405, -0.035736084, ...
http://arxiv.org/abs/2307.16833v2
Data Augmentation for Neural Machine Translation using Generative Language Model
Seokjin Oh, Su Ah Lee, Woohwan Jung
cs.CL, cs.AI
Despite the rapid growth in model architecture, the scarcity of large parallel corpora remains the main bottleneck in Neural Machine Translation. Data augmentation is a technique that enhances the performance of data-hungry models by generating synthetic data instead of collecting new ones. We explore prompt-based data...
http://arxiv.org/pdf/2307.16833v2
2023-11-13T13:17:03
2023-07-26T02:12:58
{ "title": [ -0.016830444, -0.0066719055, -0.037506104, -0.05697632, -0.0010166168, -0.00039196014, -0.041259766, 0.0056877136, 0.0034542084, 0.04434204, -0.015365601, 0.0070152283, 0.021759033, -0.017990112, -0.04510498, 0.026290894, -0.0048599243, ...
http://arxiv.org/abs/2307.13900v1
FinTree: Financial Dataset Pretrain Transformer Encoder for Relation Extraction
Hyunjong Ok
cs.CL, cs.AI
We present FinTree, Financial Dataset Pretrain Transformer Encoder for Relation Extraction. Utilizing an encoder language model, we further pretrain FinTree on the financial dataset, adapting the model in financial domain tasks. FinTree stands out with its novel structure that predicts a masked token instead of the con...
http://arxiv.org/pdf/2307.13900v1
2023-07-26T01:48:52
2023-07-26T01:48:52
{ "title": [ 0.0074272156, 0.010856628, -0.082092285, -0.06304932, 0.018875122, -0.03326416, -0.024475098, -0.0036144257, 0.023986816, 0.038024902, -0.04034424, -0.0054244995, 0.030380249, -0.007255554, -0.049835205, -0.009109497, -0.008773804, 0...
http://arxiv.org/abs/2307.13854v3
WebArena: A Realistic Web Environment for Building Autonomous Agents
Shuyan Zhou, Frank F. Xu, Hao Zhu, Xuhui Zhou, Robert Lo, Abishek Sridhar, Xianyi Cheng, Tianyue Ou, Yonatan Bisk, Daniel Fried, Uri Alon, Graham Neubig
cs.AI, cs.CL, cs.LG
With advances in generative AI, there is now potential for autonomous agents to manage daily tasks via natural language commands. However, current agents are primarily created and tested in simplified synthetic environments, leading to a disconnect with real-world scenarios. In this paper, we build an environment for l...
http://arxiv.org/pdf/2307.13854v3
2023-10-25T01:56:14
2023-07-25T22:59:32
{ "title": [ -0.024017334, -0.0038795471, -0.04977417, -0.04611206, 0.034698486, -0.020996094, 0.0045433044, -0.015930176, -0.01146698, 0.02571106, -0.020202637, -0.001074791, -0.0013341904, -0.019500732, -0.040405273, -0.0071372986, -0.013954163, ...
http://arxiv.org/abs/2308.00073v1
Trustworthiness of Children Stories Generated by Large Language Models
Prabin Bhandari, Hannah Marie Brennan
cs.CL
Large Language Models (LLMs) have shown a tremendous capacity for generating literary text. However, their effectiveness in generating children's stories has yet to be thoroughly examined. In this study, we evaluate the trustworthiness of children's stories generated by LLMs using various measures, and we compare and c...
http://arxiv.org/pdf/2308.00073v1
2023-07-25T22:55:51
2023-07-25T22:55:51
{ "title": [ -0.038970947, -0.0057144165, -0.05407715, -0.044799805, -0.016601562, -0.00982666, -0.027526855, 0.005355835, -0.03289795, 0.0345459, -0.007896423, 0.0033454895, -0.0016841888, -0.01802063, -0.028320312, -0.017044067, -0.040893555, 0...
http://arxiv.org/abs/2307.13829v1
ARC-NLP at Multimodal Hate Speech Event Detection 2023: Multimodal Methods Boosted by Ensemble Learning, Syntactical and Entity Features
Umitcan Sahin, Izzet Emre Kucukkaya, Oguzhan Ozcelik, Cagri Toraman
cs.CL, cs.SI
Text-embedded images can serve as a means of spreading hate speech, propaganda, and extremist beliefs. Throughout the Russia-Ukraine war, both opposing factions heavily relied on text-embedded images as a vehicle for spreading propaganda and hate speech. Ensuring the effective detection of hate speech and propaganda is...
http://arxiv.org/pdf/2307.13829v1
2023-07-25T21:56:14
2023-07-25T21:56:14
{ "title": [ 0.017410278, 0.018325806, -0.058685303, -0.035369873, 0.0019760132, -0.039215088, -0.038726807, 0.013725281, 0.011543274, 0.02494812, 0.0050849915, 0.006839752, 0.043548584, -0.04675293, 0.025878906, 0.024993896, -0.014953613, 0.0645...
http://arxiv.org/abs/2308.00129v1
Speech representation learning: Learning bidirectional encoders with single-view, multi-view, and multi-task methods
Qingming Tang
eess.AS, cs.AI, cs.CL
This thesis focuses on representation learning for sequence data over time or space, aiming to improve downstream sequence prediction tasks by using the learned representations. Supervised learning has been the most dominant approach for training deep neural networks for learning good sequential representations. Howeve...
http://arxiv.org/pdf/2308.00129v1
2023-07-25T20:38:55
2023-07-25T20:38:55
{ "title": [ -0.0037078857, 0.012229919, -0.058776855, -0.07171631, 0.01159668, -0.012008667, -0.04824829, 0.013084412, -0.007499695, 0.053253174, -0.023071289, 0.012863159, 0.017150879, -0.027008057, -0.019622803, 0.009933472, -0.012832642, 0.04...
http://arxiv.org/abs/2307.13808v1
Watermarking Conditional Text Generation for AI Detection: Unveiling Challenges and a Semantic-Aware Watermark Remedy
Yu Fu, Deyi Xiong, Yue Dong
cs.CL, cs.CR
To mitigate potential risks associated with language models, recent AI detection research proposes incorporating watermarks into machine-generated text through random vocabulary restrictions and utilizing this information for detection. While these watermarks only induce a slight deterioration in perplexity, our empiri...
http://arxiv.org/pdf/2307.13808v1
2023-07-25T20:24:22
2023-07-25T20:24:22
{ "title": [ -0.042816162, 0.015083313, -0.0395813, -0.024307251, 0.0132751465, -0.0027637482, -0.015182495, -0.01651001, -0.0236969, 0.024871826, -0.039215088, 0.025741577, 0.0022792816, -0.009757996, -0.030090332, -0.013420105, -0.01512146, 0.0...
http://arxiv.org/abs/2307.13779v1
Is GPT a Computational Model of Emotion? Detailed Analysis
Ala N. Tak, Jonathan Gratch
cs.CL, cs.AI, cs.CY, cs.HC
This paper investigates the emotional reasoning abilities of the GPT family of large language models via a component perspective. The paper first examines how the model reasons about autobiographical memories. Second, it systematically varies aspects of situations to impact emotion intensity and coping tendencies. Even...
http://arxiv.org/pdf/2307.13779v1
2023-07-25T19:34:44
2023-07-25T19:34:44
{ "title": [ 0.002571106, 0.012893677, -0.034362793, -0.017868042, -0.0119018555, -0.037628174, -0.019760132, -0.040649414, -0.0059776306, 0.023956299, -0.020507812, -0.025344849, 0.028442383, -0.04232788, -0.019256592, -0.0030384064, 0.002670288, ...
http://arxiv.org/abs/2307.13776v1
Combating the Curse of Multilinguality in Cross-Lingual WSD by Aligning Sparse Contextualized Word Representations
Gábor Berend
cs.CL, cs.AI
In this paper, we advocate for using large pre-trained monolingual language models in cross lingual zero-shot word sense disambiguation (WSD) coupled with a contextualized mapping mechanism. We also report rigorous experiments that illustrate the effectiveness of employing sparse contextualized word representations obt...
http://arxiv.org/pdf/2307.13776v1
2023-07-25T19:20:50
2023-07-25T19:20:50
{ "title": [ -0.011680603, -0.015548706, -0.061431885, -0.0574646, -0.011924744, -0.013504028, -0.07366943, 0.008468628, 0.016067505, 0.047424316, 0.00018000603, 0.0004553795, 0.020446777, -0.013687134, -0.029800415, 0.011299133, 0.014076233, 0.0...
http://arxiv.org/abs/2307.13693v2
Evaluating Large Language Models for Radiology Natural Language Processing
Zhengliang Liu, Tianyang Zhong, Yiwei Li, Yutong Zhang, Yi Pan, Zihao Zhao, Peixin Dong, Chao Cao, Yuxiao Liu, Peng Shu, Yaonai Wei, Zihao Wu, Chong Ma, Jiaqi Wang, Sheng Wang, Mengyue Zhou, Zuowei Jiang, Chunlin Li, Jason Holmes, Shaochen Xu, Lu Zhang, Haixing Dai, Kai Zhang, Lin Zhao, Yuanhao Chen, Xu Liu, Peilong Wa...
cs.CL
The rise of large language models (LLMs) has marked a pivotal shift in the field of natural language processing (NLP). LLMs have revolutionized a multitude of domains, and they have made a significant impact in the medical field. Large language models are now more abundant than ever, and many of these models exhibit bi...
http://arxiv.org/pdf/2307.13693v2
2023-07-27T12:58:59
2023-07-25T17:57:18
{ "title": [ -0.02609253, 0.00034093857, -0.027618408, -0.042053223, 0.0006093979, 0.005302429, -0.035583496, -0.01133728, 0.0005130768, 0.02281189, -0.024475098, -0.01977539, -0.011306763, -0.023513794, -0.042755127, -0.005290985, 0.0019989014, ...
http://arxiv.org/abs/2307.13692v2
ARB: Advanced Reasoning Benchmark for Large Language Models
Tomohiro Sawada, Daniel Paleka, Alexander Havrilla, Pranav Tadepalli, Paula Vidas, Alexander Kranias, John J. Nay, Kshitij Gupta, Aran Komatsuzaki
cs.CL, cs.LG
Large Language Models (LLMs) have demonstrated remarkable performance on various quantitative reasoning and knowledge benchmarks. However, many of these benchmarks are losing utility as LLMs get increasingly high scores, despite not yet reaching expert performance in these domains. We introduce ARB, a novel benchmark c...
http://arxiv.org/pdf/2307.13692v2
2023-07-28T03:31:08
2023-07-25T17:55:19
{ "title": [ -0.039886475, -0.031463623, -0.031402588, -0.09124756, 0.0005502701, 0.020736694, -0.078308105, 0.020584106, 0.010475159, 0.023391724, -0.010902405, 0.0071029663, 0.008171082, -0.02255249, -0.029434204, 0.021484375, -0.013557434, 0.0...
http://arxiv.org/abs/2307.14377v1
How Can Large Language Models Help Humans in Design and Manufacturing?
Liane Makatura, Michael Foshey, Bohan Wang, Felix HähnLein, Pingchuan Ma, Bolei Deng, Megan Tjandrasuwita, Andrew Spielberg, Crystal Elaine Owens, Peter Yichen Chen, Allan Zhao, Amy Zhu, Wil J Norton, Edward Gu, Joshua Jacob, Yifei Li, Adriana Schulz, Wojciech Matusik
cs.CL, cs.AI
The advancement of Large Language Models (LLMs), including GPT-4, provides exciting new opportunities for generative design. We investigate the application of this tool across the entire design and manufacturing workflow. Specifically, we scrutinize the utility of LLMs in tasks such as: converting a text-based prompt i...
http://arxiv.org/pdf/2307.14377v1
2023-07-25T17:30:38
2023-07-25T17:30:38
{ "title": [ -0.039886475, -0.007659912, -0.043518066, -0.013702393, -0.04434204, -0.016403198, -0.064697266, 0.00048804283, -0.029510498, 0.02796936, -0.009223938, -0.0035877228, -0.0005030632, -0.03366089, 0.021499634, -0.007873535, 0.01335907, ...
http://arxiv.org/abs/2307.13655v1
A Comprehensive Evaluation and Analysis Study for Chinese Spelling Check
Xunjian Yin, Xiaojun Wan
cs.CL
With the development of pre-trained models and the incorporation of phonetic and graphic information, neural models have achieved high scores in Chinese Spelling Check (CSC). However, it does not provide a comprehensive reflection of the models' capability due to the limited test sets. In this study, we abstract the re...
http://arxiv.org/pdf/2307.13655v1
2023-07-25T17:02:38
2023-07-25T17:02:38
{ "title": [ -0.010765076, 0.0014972687, -0.02166748, -0.014503479, -0.007972717, -0.019104004, 0.012939453, -0.011268616, 0.0076522827, 0.057647705, -0.026733398, -0.0110321045, 0.05126953, -0.017822266, -0.029251099, 0.015731812, -0.028198242, ...
http://arxiv.org/abs/2307.13631v1
Contributions to the Improvement of Question Answering Systems in the Biomedical Domain
Mourad Sarrouti
cs.CL
This thesis work falls within the framework of question answering (QA) in the biomedical domain where several specific challenges are addressed, such as specialized lexicons and terminologies, the types of treated questions, and the characteristics of targeted documents. We are particularly interested in studying and i...
http://arxiv.org/pdf/2307.13631v1
2023-07-25T16:31:20
2023-07-25T16:31:20
{ "title": [ 0.0050354004, -0.0069618225, -0.025283813, -0.05206299, 0.005130768, -0.0026187897, -0.026016235, -0.0028018951, 0.0065574646, 0.028564453, -0.012718201, -0.011131287, 0.010620117, -0.025497437, -0.05581665, 0.009422302, 0.0065231323, ...
http://arxiv.org/abs/2307.13617v2
GPT-3 Models are Few-Shot Financial Reasoners
Raul Salles de Padua, Imran Qureshi, Mustafa U. Karakaplan
cs.CL, cs.AI
Financial analysis is an important tool for evaluating company performance. Practitioners work to answer financial questions to make profitable investment decisions, and use advanced quantitative analyses to do so. As a result, Financial Question Answering (QA) is a question answering task that requires deep reasoning ...
http://arxiv.org/pdf/2307.13617v2
2023-07-26T16:14:24
2023-07-25T16:21:07
{ "title": [ 0.042541504, -0.014694214, 0.00024580956, -0.022705078, -0.06311035, -0.021224976, -0.032714844, -0.0206604, -0.016098022, 0.02027893, -0.06274414, -0.009811401, 0.0079193115, -0.02444458, -0.00036001205, -0.037200928, 0.012794495, 0...
http://arxiv.org/abs/2307.13714v1
Diversity and Language Technology: How Techno-Linguistic Bias Can Cause Epistemic Injustice
Paula Helm, Gábor Bella, Gertraud Koch, Fausto Giunchiglia
cs.CY, cs.CL
It is well known that AI-based language technology -- large language models, machine translation systems, multilingual dictionaries, and corpora -- is currently limited to 2 to 3 percent of the world's most widely spoken and/or financially and politically best supported languages. In response, recent research efforts h...
http://arxiv.org/pdf/2307.13714v1
2023-07-25T16:08:27
2023-07-25T16:08:27
{ "title": [ -0.02670288, 0.002796173, -0.039978027, -0.02670288, 0.008918762, -0.020507812, 0.015556335, -0.0006546974, -0.012145996, 0.0385437, -0.018341064, 0.0025234222, 0.020614624, -0.016860962, -0.04421997, -0.0024375916, -0.020629883, 0.0...
http://arxiv.org/abs/2307.13560v2
XDLM: Cross-lingual Diffusion Language Model for Machine Translation
Linyao Chen, Aosong Feng, Boming Yang, Zihui Li
cs.CL
Recently, diffusion models have excelled in image generation tasks and have also been applied to neural language processing (NLP) for controllable text generation. However, the application of diffusion models in a cross-lingual setting is less unexplored. Additionally, while pretraining with diffusion models has been s...
http://arxiv.org/pdf/2307.13560v2
2023-07-31T01:29:31
2023-07-25T15:08:34
{ "title": [ -0.04336548, -0.016113281, -0.027420044, -0.03149414, -0.0039520264, 0.008796692, -0.03353882, 0.0012960434, 0.0057640076, 0.041290283, -0.0047073364, 0.022155762, 0.021774292, -0.019592285, -0.036315918, 0.027297974, -0.029312134, 0...
http://arxiv.org/abs/2307.13528v2
FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
I-Chun Chern, Steffi Chern, Shiqi Chen, Weizhe Yuan, Kehua Feng, Chunting Zhou, Junxian He, Graham Neubig, Pengfei Liu
cs.CL, cs.AI
The emergence of generative pre-trained models has facilitated the synthesis of high-quality text, but it has also posed challenges in identifying factual errors in the generated text. In particular: (1) A wider range of tasks now face an increasing risk of containing factual errors when handled by generative models. (...
http://arxiv.org/pdf/2307.13528v2
2023-07-26T15:17:49
2023-07-25T14:20:51
{ "title": [ 0.0004069805, 0.004348755, -0.056549072, -0.0519104, -0.009994507, -0.025222778, -0.04071045, -0.016586304, 0.006504059, 0.015838623, -0.003326416, 0.013656616, 0.042663574, -0.050933838, 0.025878906, 0.015029907, -0.03717041, 0.0566...
http://arxiv.org/abs/2307.13497v1
Zshot: An Open-source Framework for Zero-Shot Named Entity Recognition and Relation Extraction
Gabriele Picco, Marcos Martínez Galindo, Alberto Purpura, Leopold Fuchs, Vanessa López, Hoang Thanh Lam
cs.CL, cs.AI, cs.LG
The Zero-Shot Learning (ZSL) task pertains to the identification of entities or relations in texts that were not seen during training. ZSL has emerged as a critical research area due to the scarcity of labeled data in specific domains, and its applications have grown significantly in recent years. With the advent of la...
http://arxiv.org/pdf/2307.13497v1
2023-07-25T13:46:36
2023-07-25T13:46:36
{ "title": [ -0.007255554, -0.013267517, -0.066711426, -0.0859375, 0.018814087, -0.0024662018, -0.03805542, 0.013618469, 0.023422241, 0.024108887, -0.012367249, 0.016616821, 0.011924744, -0.009223938, -0.036468506, 0.015571594, -0.0057029724, 0.0...
http://arxiv.org/abs/2307.13424v1
Holistic Exploration on Universal Decompositional Semantic Parsing: Architecture, Data Augmentation, and LLM Paradigm
Hexuan Deng, Xin Zhang, Meishan Zhang, Xuebo Liu, Min Zhang
cs.CL
In this paper, we conduct a holistic exploration of the Universal Decompositional Semantic (UDS) Parsing. We first introduce a cascade model for UDS parsing that decomposes the complex parsing task into semantically appropriate subtasks. Our approach outperforms the prior models, while significantly reducing inference ...
http://arxiv.org/pdf/2307.13424v1
2023-07-25T11:44:28
2023-07-25T11:44:28
{ "title": [ -0.037719727, 0.0005092621, -0.05810547, -0.09246826, -0.002286911, 0.010719299, -0.090026855, 0.020965576, -0.014053345, 0.04006958, -0.0124435425, 0.041229248, 0.0072631836, -0.009063721, -0.01121521, 0.022079468, -0.01663208, 0.06...
http://arxiv.org/abs/2307.13417v1
Towards Resolving Word Ambiguity with Word Embeddings
Matthias Thurnbauer, Johannes Reisinger, Christoph Goller, Andreas Fischer
cs.CL
Ambiguity is ubiquitous in natural language. Resolving ambiguous meanings is especially important in information retrieval tasks. While word embeddings carry semantic information, they fail to handle ambiguity well. Transformer models have been shown to handle word ambiguity for complex queries, but they cannot be used...
http://arxiv.org/pdf/2307.13417v1
2023-07-25T11:29:55
2023-07-25T11:29:55
{ "title": [ -0.031173706, -0.0051956177, -0.03579712, -0.043701172, 0.0024719238, 0.01335907, -0.051940918, -0.007949829, 0.020141602, 0.021636963, -0.019561768, -0.0026359558, 0.008148193, -0.008666992, -0.044647217, 0.0066375732, -0.017959595, ...
http://arxiv.org/abs/2307.13405v1
Towards Bridging the Digital Language Divide
Gábor Bella, Paula Helm, Gertraud Koch, Fausto Giunchiglia
cs.CL, cs.AI, I.2.7; K.4.2
It is a well-known fact that current AI-based language technology -- language models, machine translation systems, multilingual dictionaries and corpora -- focuses on the world's 2-3% most widely spoken languages. Recent research efforts have attempted to expand the coverage of AI technology to `under-resourced languag...
http://arxiv.org/pdf/2307.13405v1
2023-07-25T10:53:20
2023-07-25T10:53:20
{ "title": [ -0.023895264, -0.0050239563, -0.035858154, -0.019973755, -0.02218628, -0.019927979, -0.018753052, 0.0030231476, -0.0146865845, 0.049926758, -0.008453369, -0.013130188, 0.022277832, -0.038848877, -0.028945923, 0.012901306, -0.0068130493, ...
http://arxiv.org/abs/2307.13377v1
Embedding Models for Supervised Automatic Extraction and Classification of Named Entities in Scientific Acknowledgements
Nina Smirnova, Philipp Mayr
cs.DL, cs.CL, cs.IR, J.4; J.5; I.5.1; H.3.3; I.2.7
Acknowledgments in scientific papers may give an insight into aspects of the scientific community, such as reward systems, collaboration patterns, and hidden research trends. The aim of the paper is to evaluate the performance of different embedding models for the task of automatic extraction and classification of ackn...
http://arxiv.org/pdf/2307.13377v1
2023-07-25T09:51:17
2023-07-25T09:51:17
{ "title": [ -0.011619568, 0.014961243, -0.058135986, -0.057037354, 0.027450562, -0.0154800415, -0.020706177, -0.0009832382, 0.021957397, 0.021743774, -0.002664566, -0.011047363, -0.0054893494, -0.015129089, -0.05545044, 0.006225586, -0.023345947, ...
http://arxiv.org/abs/2307.14367v1
Prot2Text: Multimodal Protein's Function Generation with GNNs and Transformers
Hadi Abdine, Michail Chatzianastasis, Costas Bouyioukos, Michalis Vazirgiannis
q-bio.QM, cs.CL, cs.LG
The complex nature of big biological systems pushed some scientists to classify its understanding under the inconceivable missions. Different leveled challenges complicated this task, one of is the prediction of a protein's function. In recent years, significant progress has been made in this field through the developm...
http://arxiv.org/pdf/2307.14367v1
2023-07-25T09:35:43
2023-07-25T09:35:43
{ "title": [ -0.043823242, 0.010955811, -0.062561035, -0.07281494, 0.0066070557, 0.0026607513, -0.07507324, -0.0036411285, 0.0038223267, 0.0050354004, 0.031066895, -0.015411377, 0.006362915, -0.06298828, 0.00003707409, 0.022872925, 0.0058937073, ...
http://arxiv.org/abs/2307.13365v2
Empower Your Model with Longer and Better Context Comprehension
Yifei Gao, Lei Wang, Jun Fang, Longhua Hu, Jun Cheng
cs.CL, cs.AI, 68T07, 68T50
Recently, with the emergence of numerous Large Language Models (LLMs), the implementation of AI has entered a new era. Irrespective of these models' own capacity and structure, there is a growing demand for LLMs to possess enhanced comprehension of longer and more complex contexts with relatively smaller sizes. Models ...
http://arxiv.org/pdf/2307.13365v2
2023-07-27T10:17:18
2023-07-25T09:34:42
{ "title": [ 0.06021118, -0.0012788773, -0.019836426, -0.07867432, -0.014587402, -0.011482239, -0.047912598, -0.014556885, -0.008758545, 0.028778076, -0.042785645, -0.009635925, 0.0067749023, -0.034973145, 0.0413208, -0.01637268, 0.019729614, 0.0...
http://arxiv.org/abs/2307.13339v1
Analyzing Chain-of-Thought Prompting in Large Language Models via Gradient-based Feature Attributions
Skyler Wu, Eric Meng Shen, Charumathi Badrinath, Jiaqi Ma, Himabindu Lakkaraju
cs.CL, cs.AI
Chain-of-thought (CoT) prompting has been shown to empirically improve the accuracy of large language models (LLMs) on various question answering tasks. While understanding why CoT prompting is effective is crucial to ensuring that this phenomenon is a consequence of desired model behavior, little work has addressed th...
http://arxiv.org/pdf/2307.13339v1
2023-07-25T08:51:30
2023-07-25T08:51:30
{ "title": [ -0.0063934326, -0.01222229, -0.010917664, -0.042510986, -0.02684021, 0.004333496, -0.06298828, -0.022766113, -0.0044670105, 0.004047394, -0.027648926, 0.0012931824, 0.015838623, -0.027175903, -0.015464783, -0.008361816, -0.013877869, ...
http://arxiv.org/abs/2308.02506v1
Improving the Generalization Ability in Essay Coherence Evaluation through Monotonic Constraints
Chen Zheng, Huan Zhang, Yan Zhao, Yuxuan Lai
cs.CL
Coherence is a crucial aspect of evaluating text readability and can be assessed through two primary factors when evaluating an essay in a scoring scenario. The first factor is logical coherence, characterized by the appropriate use of discourse connectives and the establishment of logical relationships between sentenc...
http://arxiv.org/pdf/2308.02506v1
2023-07-25T08:26:46
2023-07-25T08:26:46
{ "title": [ 0.020401001, 0.033416748, -0.038909912, -0.047088623, 0.017807007, -0.019042969, -0.01398468, -0.043823242, -0.015991211, 0.05368042, -0.0072669983, 0.010185242, 0.005634308, 0.00065612793, -0.056427002, -0.019882202, -0.03338623, 0....
http://arxiv.org/abs/2307.13304v1
QuIP: 2-Bit Quantization of Large Language Models With Guarantees
Jerry Chee, Yaohui Cai, Volodymyr Kuleshov, Christopher De Sa
cs.LG, cs.CL
This work studies post-training parameter quantization in large language models (LLMs). We introduce quantization with incoherence processing (QuIP), a new method based on the insight that quantization benefits from incoherent weight and Hessian matrices, i.e., from the weights and the directions in which it is importa...
http://arxiv.org/pdf/2307.13304v1
2023-07-25T07:44:06
2023-07-25T07:44:06
{ "title": [ -0.06463623, -0.010238647, -0.056915283, -0.054901123, 0.00033807755, 0.008453369, -0.07171631, 0.010269165, -0.010787964, 0.05340576, -0.00002425909, -0.009857178, 0.018722534, -0.013450623, -0.050109863, 0.018432617, -0.01134491, 0...
http://arxiv.org/abs/2307.13298v1
An Intent Taxonomy of Legal Case Retrieval
Yunqiu Shao, Haitao Li, Yueyue Wu, Yiqun Liu, Qingyao Ai, Jiaxin Mao, Yixiao Ma, Shaoping Ma
cs.IR, cs.CL
Legal case retrieval is a special Information Retrieval~(IR) task focusing on legal case documents. Depending on the downstream tasks of the retrieved case documents, users' information needs in legal case retrieval could be significantly different from those in Web search and traditional ad-hoc retrieval tasks. While ...
http://arxiv.org/pdf/2307.13298v1
2023-07-25T07:27:32
2023-07-25T07:27:32
{ "title": [ 0.002922058, 0.005054474, -0.041870117, -0.054779053, 0.010894775, -0.017211914, -0.0016098022, 0.0099105835, 0.025527954, 0.032562256, -0.044921875, 0.00067043304, 0.021896362, -0.0060157776, -0.054992676, 0.010986328, -0.009094238, ...
http://arxiv.org/abs/2307.16663v1
Word Sense Disambiguation as a Game of Neurosymbolic Darts
Tiansi Dong, Rafet Sifa
cs.CL, cs.AI, cs.LG
Word Sense Disambiguation (WSD) is one of the hardest tasks in natural language understanding and knowledge engineering. The glass ceiling of 80% F1 score is recently achieved through supervised deep-learning, enriched by a variety of knowledge graphs. Here, we propose a novel neurosymbolic methodology that is able to ...
http://arxiv.org/pdf/2307.16663v1
2023-07-25T07:22:57
2023-07-25T07:22:57
{ "title": [ 0.018310547, -0.0030574799, 0.0018138885, -0.004901886, -0.015609741, -0.031066895, 0.020248413, -0.022064209, 0.01828003, 0.051574707, -0.020431519, -0.011146545, 0.05206299, -0.020828247, -0.047454834, 0.015052795, 0.011428833, 0.0...
http://arxiv.org/abs/2307.13269v1
LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Composition
Chengsong Huang, Qian Liu, Bill Yuchen Lin, Tianyu Pang, Chao Du, Min Lin
cs.CL, cs.AI
Low-rank adaptations (LoRA) are often employed to fine-tune large language models (LLMs) for new tasks. This paper investigates LoRA composability for cross-task generalization and introduces LoraHub, a strategic framework devised for the purposive assembly of LoRA modules trained on diverse given tasks, with the objec...
http://arxiv.org/pdf/2307.13269v1
2023-07-25T05:39:21
2023-07-25T05:39:21
{ "title": [ -0.044433594, -0.011657715, -0.0541687, -0.08093262, -0.00023007393, 0.007194519, -0.049438477, 0.00030755997, -0.0031147003, 0.050598145, -0.008300781, 0.030426025, -0.0013475418, -0.009246826, -0.0158844, 0.024307251, -0.020645142, ...
http://arxiv.org/abs/2307.13176v1
Schema-Driven Actionable Insight Generation and Smart Recommendation
Allmin Susaiyah, Aki Härmä, Milan Petković
cs.CL
In natural language generation (NLG), insight mining is seen as a data-to-text task, where data is mined for interesting patterns and verbalised into 'insight' statements. An 'over-generate and rank' paradigm is intuitively used to generate such insights. The multidimensionality and subjectivity of this process make it...
http://arxiv.org/pdf/2307.13176v1
2023-07-24T23:53:13
2023-07-24T23:53:13
{ "title": [ -0.0071907043, -0.009559631, -0.044433594, -0.0602417, 0.0021343231, -0.005554199, -0.022033691, 0.0036201477, 0.0025939941, 0.020507812, -0.018234253, -0.0016155243, -0.0146102905, -0.022720337, -0.028320312, 0.0022277832, 0.018112183, ...