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/2304.10495v1
A primer on getting neologisms from foreign languages to under-resourced languages
Luis Camacho
cs.CL
Mainly due to lack of support, most under-resourced languages have a reduced lexicon in most realms and domains of increasing importance, then their speakers need to significantly augment it. Although neologisms should arise from the languages themselves, external sources are widely accepted. However, we dispute the "c...
http://arxiv.org/pdf/2304.10495v1
2023-03-07T15:10:37
2023-03-07T15:10:37
{ "title": [ 0.011161804, -0.028930664, 0.008415222, -0.009300232, -0.05908203, -0.0030136108, -0.012245178, -0.010559082, -0.03274536, 0.03729248, -0.023834229, 0.020492554, 0.017974854, -0.031463623, 0.010322571, -0.023040771, -0.00074768066, 0...
http://arxiv.org/abs/2303.05391v2
Disambiguation of Company names via Deep Recurrent Networks
Alessandro Basile, Riccardo Crupi, Michele Grasso, Alessandro Mercanti, Daniele Regoli, Simone Scarsi, Shuyi Yang, Andrea Cosentini
cs.CL, cs.AI, cs.DB, cs.LG
Name Entity Disambiguation is the Natural Language Processing task of identifying textual records corresponding to the same Named Entity, i.e. real-world entities represented as a list of attributes (names, places, organisations, etc.). In this work, we face the task of disambiguating companies on the basis of their wr...
http://arxiv.org/pdf/2303.05391v2
2023-04-15T20:32:00
2023-03-07T15:07:57
{ "title": [ -0.026260376, 0.0074005127, -0.060668945, -0.05883789, 0.015403748, -0.0077934265, -0.050872803, 0.0055007935, 0.008712769, 0.042175293, -0.019882202, 0.011451721, 0.0060195923, 0.0070610046, -0.036987305, 0.0013093948, -0.017440796, ...
http://arxiv.org/abs/2303.03953v2
ChatGPT: Beginning of an End of Manual Linguistic Data Annotation? Use Case of Automatic Genre Identification
Taja Kuzman, Igor Mozetič, Nikola Ljubešić
cs.CL, cs.LG
ChatGPT has shown strong capabilities in natural language generation tasks, which naturally leads researchers to explore where its abilities end. In this paper, we examine whether ChatGPT can be used for zero-shot text classification, more specifically, automatic genre identification. We compare ChatGPT with a multilin...
http://arxiv.org/pdf/2303.03953v2
2023-03-08T09:35:09
2023-03-07T14:59:33
{ "title": [ -0.008766174, -0.0019779205, -0.04559326, -0.012817383, -0.019851685, -0.018997192, -0.02734375, -0.005428314, -0.03164673, 0.036468506, 0.020599365, 0.0027809143, 0.027053833, -0.049713135, 0.00957489, 0.0093688965, -0.017181396, 0....
http://arxiv.org/abs/2303.03948v1
A Meta-Evaluation of Faithfulness Metrics for Long-Form Hospital-Course Summarization
Griffin Adams, Jason Zucker, Noémie Elhadad
cs.CL
Long-form clinical summarization of hospital admissions has real-world significance because of its potential to help both clinicians and patients. The faithfulness of summaries is critical to their safe usage in clinical settings. To better understand the limitations of abstractive systems, as well as the suitability o...
http://arxiv.org/pdf/2303.03948v1
2023-03-07T14:57:06
2023-03-07T14:57:06
{ "title": [ -0.011833191, 0.01499939, -0.03277588, -0.060821533, 0.0060653687, 0.0041885376, -0.05038452, -0.039886475, -0.029312134, 0.056488037, -0.014167786, 0.031021118, -0.012802124, -0.0036411285, -0.028381348, -0.010131836, -0.0051612854, ...
http://arxiv.org/abs/2303.03926v1
Speak Foreign Languages with Your Own Voice: Cross-Lingual Neural Codec Language Modeling
Ziqiang Zhang, Long Zhou, Chengyi Wang, Sanyuan Chen, Yu Wu, Shujie Liu, Zhuo Chen, Yanqing Liu, Huaming Wang, Jinyu Li, Lei He, Sheng Zhao, Furu Wei
cs.CL, cs.AI, cs.SD, eess.AS
We propose a cross-lingual neural codec language model, VALL-E X, for cross-lingual speech synthesis. Specifically, we extend VALL-E and train a multi-lingual conditional codec language model to predict the acoustic token sequences of the target language speech by using both the source language speech and the target la...
http://arxiv.org/pdf/2303.03926v1
2023-03-07T14:31:55
2023-03-07T14:31:55
{ "title": [ -0.029464722, -0.0048675537, -0.027282715, -0.01687622, -0.022018433, -0.011650085, -0.035888672, -0.004234314, -0.010757446, 0.050598145, -0.003818512, -0.017196655, 0.034576416, -0.030059814, -0.031204224, 0.014839172, -0.017089844, ...
http://arxiv.org/abs/2303.03915v1
The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset
Hugo Laurençon, Lucile Saulnier, Thomas Wang, Christopher Akiki, Albert Villanova del Moral, Teven Le Scao, Leandro Von Werra, Chenghao Mou, Eduardo González Ponferrada, Huu Nguyen, Jörg Frohberg, Mario Šaško, Quentin Lhoest, Angelina McMillan-Major, Gerard Dupont, Stella Biderman, Anna Rogers, Loubna Ben allal, France...
cs.CL, cs.AI, I.2.7
As language models grow ever larger, the need for large-scale high-quality text datasets has never been more pressing, especially in multilingual settings. The BigScience workshop, a 1-year international and multidisciplinary initiative, was formed with the goal of researching and training large language models as a va...
http://arxiv.org/pdf/2303.03915v1
2023-03-07T14:25:44
2023-03-07T14:25:44
{ "title": [ -0.026168823, -0.009101868, -0.036590576, -0.060455322, 0.0060195923, 0.00818634, -0.062408447, -0.008888245, 0.020828247, 0.032806396, 0.007987976, -0.018234253, 0.006336212, -0.04196167, -0.031433105, 0.02104187, 0.011550903, 0.041...
http://arxiv.org/abs/2303.03912v1
Document-level Relation Extraction with Cross-sentence Reasoning Graph
Hongfei Liu, Zhao Kang, Lizong Zhang, Ling Tian, Fujun Hua
cs.CL, cs.AI, cs.LG, cs.SI
Relation extraction (RE) has recently moved from the sentence-level to document-level, which requires aggregating document information and using entities and mentions for reasoning. Existing works put entity nodes and mention nodes with similar representations in a document-level graph, whose complex edges may incur re...
http://arxiv.org/pdf/2303.03912v1
2023-03-07T14:14:12
2023-03-07T14:14:12
{ "title": [ -0.008171082, -0.016418457, -0.0335083, -0.08496094, -0.0043296814, 0.011947632, -0.05343628, 0.009803772, 0.039215088, 0.003982544, -0.023757935, 0.00818634, -0.0036582947, -0.010574341, -0.02973938, 0.003818512, -0.020843506, 0.054...
http://arxiv.org/abs/2303.03846v2
Larger language models do in-context learning differently
Jerry Wei, Jason Wei, Yi Tay, Dustin Tran, Albert Webson, Yifeng Lu, Xinyun Chen, Hanxiao Liu, Da Huang, Denny Zhou, Tengyu Ma
cs.CL
We study how in-context learning (ICL) in language models is affected by semantic priors versus input-label mappings. We investigate two setups-ICL with flipped labels and ICL with semantically-unrelated labels-across various model families (GPT-3, InstructGPT, Codex, PaLM, and Flan-PaLM). First, experiments on ICL wit...
http://arxiv.org/pdf/2303.03846v2
2023-03-08T07:37:43
2023-03-07T12:24:17
{ "title": [ 0.0003671646, -0.00020074844, -0.02154541, -0.008972168, -0.0501709, -0.027328491, -0.011917114, 0.0010910034, -0.04397583, 0.04159546, -0.058685303, -0.0049209595, -0.013938904, -0.015899658, 0.009140015, -0.037109375, 0.006198883, ...
http://arxiv.org/abs/2303.03840v2
A Challenging Benchmark for Low-Resource Learning
Yudong Wang, Chang Ma, Qingxiu Dong, Lingpeng Kong, Jingjing Xu
cs.CL, cs.AI
With promising yet saturated results in high-resource settings, low-resource datasets have gradually become popular benchmarks for evaluating the learning ability of advanced neural networks (e.g., BigBench, superGLUE). Some models even surpass humans according to benchmark test results. However, we find that there exi...
http://arxiv.org/pdf/2303.03840v2
2023-03-09T06:28:38
2023-03-07T12:10:47
{ "title": [ -0.010025024, 0.008056641, -0.012458801, -0.037139893, -0.01701355, -0.0141067505, -0.031066895, -0.019165039, 0.0011348724, 0.026779175, -0.023361206, 0.0055656433, -0.06317139, -0.027069092, -0.039764404, -0.009124756, 0.010734558, ...
http://arxiv.org/abs/2303.03836v2
Exploring the Feasibility of ChatGPT for Event Extraction
Jun Gao, Huan Zhao, Changlong Yu, Ruifeng Xu
cs.CL
Event extraction is a fundamental task in natural language processing that involves identifying and extracting information about events mentioned in text. However, it is a challenging task due to the lack of annotated data, which is expensive and time-consuming to obtain. The emergence of large language models (LLMs) s...
http://arxiv.org/pdf/2303.03836v2
2023-03-09T17:33:31
2023-03-07T12:03:58
{ "title": [ -0.0029239655, -0.0072288513, -0.030563354, -0.06970215, 0.018157959, 0.017044067, -0.043640137, -0.01158905, -0.008087158, 0.021774292, 0.0105896, 0.021530151, -0.010177612, -0.02848816, -0.027252197, 0.026138306, -0.026611328, 0.03...
http://arxiv.org/abs/2303.05388v1
German BERT Model for Legal Named Entity Recognition
Harshil Darji, Jelena Mitrović, Michael Granitzer
cs.CL, cs.LG
The use of BERT, one of the most popular language models, has led to improvements in many Natural Language Processing (NLP) tasks. One such task is Named Entity Recognition (NER) i.e. automatic identification of named entities such as location, person, organization, etc. from a given text. It is also an important base ...
http://arxiv.org/pdf/2303.05388v1
2023-03-07T11:54:39
2023-03-07T11:54:39
{ "title": [ -0.028808594, -0.01675415, -0.05532837, -0.043518066, -0.0206604, 0.0027637482, -0.049468994, 0.016937256, -0.00548172, 0.043548584, -0.031585693, 0.016433716, 0.06903076, -0.029907227, 0.0064926147, 0.031585693, -0.027053833, 0.0540...
http://arxiv.org/abs/2303.03750v2
Preparing the Vuk'uzenzele and ZA-gov-multilingual South African multilingual corpora
Richard Lastrucci, Isheanesu Dzingirai, Jenalea Rajab, Andani Madodonga, Matimba Shingange, Daniel Njini, Vukosi Marivate
cs.CL
This paper introduces two multilingual government themed corpora in various South African languages. The corpora were collected by gathering the South African Government newspaper (Vuk'uzenzele), as well as South African government speeches (ZA-gov-multilingual), that are translated into all 11 South African official l...
http://arxiv.org/pdf/2303.03750v2
2023-04-05T09:39:32
2023-03-07T09:20:09
{ "title": [ -0.019683838, -0.015808105, -0.037109375, -0.057769775, -0.03527832, 0.014511108, -0.07196045, 0.01524353, -0.047698975, 0.06518555, 0.009178162, 0.032806396, 0.06304932, -0.040802002, 0.038604736, 0.0071144104, -0.0031929016, 0.0459...
http://arxiv.org/abs/2303.03715v2
Universal resources for quantum computing
D. -S. Wang
quant-ph, cs.CL
Unravelling the source of quantum computing power has been a major goal in the field of quantum information science. In recent years, the quantum resource theory (QRT) has been established to characterize various quantum resources, yet their roles in quantum computing tasks still require investigation. The so-called un...
http://arxiv.org/pdf/2303.03715v2
2023-03-08T02:02:20
2023-03-07T07:56:38
{ "title": [ -0.020828247, 0.004611969, -0.007217407, -0.03878784, -0.012550354, -0.00440979, 0.003944397, -0.016677856, -0.00819397, 0.072387695, 0.017486572, 0.00806427, 0.020507812, -0.020202637, -0.032440186, 0.042175293, 0.012260437, 0.03105...
http://arxiv.org/abs/2303.03706v1
Classifying Text-Based Conspiracy Tweets related to COVID-19 using Contextualized Word Embeddings
Abdul Rehman, Rabeeh Ayaz Abbasi, Irfan ul Haq Qureshi, Akmal Saeed Khattak
cs.CL, cs.LG, 68T01, I.2.7
The FakeNews task in MediaEval 2022 investigates the challenge of finding accurate and high-performance models for the classification of conspiracy tweets related to COVID-19. In this paper, we used BERT, ELMO, and their combination for feature extraction and RandomForest as classifier. The results show that ELMO perfo...
http://arxiv.org/pdf/2303.03706v1
2023-03-07T07:42:26
2023-03-07T07:42:26
{ "title": [ 0.017227173, 0.010627747, -0.0362854, -0.05923462, 0.012229919, -0.0044059753, -0.024139404, -0.0068244934, -0.03378296, 0.015930176, -0.034484863, 0.03543091, 0.014053345, -0.017333984, 0.004146576, -0.017791748, -0.021453857, 0.054...
http://arxiv.org/abs/2303.03697v1
Stylometric Detection of AI-Generated Text in Twitter Timelines
Tharindu Kumarage, Joshua Garland, Amrita Bhattacharjee, Kirill Trapeznikov, Scott Ruston, Huan Liu
cs.CL, cs.LG
Recent advancements in pre-trained language models have enabled convenient methods for generating human-like text at a large scale. Though these generation capabilities hold great potential for breakthrough applications, it can also be a tool for an adversary to generate misinformation. In particular, social media plat...
http://arxiv.org/pdf/2303.03697v1
2023-03-07T07:26:09
2023-03-07T07:26:09
{ "title": [ -0.015899658, 0.010185242, -0.03829956, -0.036102295, 0.0022354126, -0.0008401871, -0.026062012, -0.014404297, -0.019897461, 0.03955078, -0.024505615, 0.047821045, 0.0069007874, -0.00083494186, -0.022949219, 0.0045166016, -0.017318726, ...
http://arxiv.org/abs/2303.03628v1
CoTEVer: Chain of Thought Prompting Annotation Toolkit for Explanation Verification
Seungone Kim, Se June Joo, Yul Jang, Hyungjoo Chae, Jinyoung Yeo
cs.CL, cs.LG
Chain-of-thought (CoT) prompting enables large language models (LLMs) to solve complex reasoning tasks by generating an explanation before the final prediction. Despite it's promising ability, a critical downside of CoT prompting is that the performance is greatly affected by the factuality of the generated explanation...
http://arxiv.org/pdf/2303.03628v1
2023-03-07T03:23:14
2023-03-07T03:23:14
{ "title": [ -0.011650085, -0.020950317, -0.002412796, -0.06732178, -0.02758789, 0.01739502, -0.058166504, -0.014595032, -0.0018463135, 0.00084257126, -0.013450623, 0.008071899, 0.0042152405, -0.030792236, -0.0074272156, 0.0068740845, -0.008407593, ...
http://arxiv.org/abs/2303.03608v1
Towards Interpretable and Efficient Automatic Reference-Based Summarization Evaluation
Yixin Liu, Alexander R. Fabbri, Yilun Zhao, Pengfei Liu, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, Dragomir Radev
cs.CL
Interpretability and efficiency are two important considerations for the adoption of neural automatic metrics. In this work, we develop strong-performing automatic metrics for reference-based summarization evaluation, based on a two-stage evaluation pipeline that first extracts basic information units from one text seq...
http://arxiv.org/pdf/2303.03608v1
2023-03-07T02:49:50
2023-03-07T02:49:50
{ "title": [ -0.018951416, 0.012680054, -0.032592773, -0.08679199, 0.010108948, 0.021865845, -0.070251465, -0.019805908, 0.009147644, 0.053771973, -0.015617371, 0.0063056946, 0.010009766, 0.012825012, -0.07354736, 0.011398315, -0.014190674, 0.040...
http://arxiv.org/abs/2303.03600v1
Adaptive Knowledge Distillation between Text and Speech Pre-trained Models
Jinjie Ni, Yukun Ma, Wen Wang, Qian Chen, Dianwen Ng, Han Lei, Trung Hieu Nguyen, Chong Zhang, Bin Ma, Erik Cambria
cs.CL, cs.LG, cs.SD, eess.AS
Learning on a massive amount of speech corpus leads to the recent success of many self-supervised speech models. With knowledge distillation, these models may also benefit from the knowledge encoded by language models that are pre-trained on rich sources of texts. The distillation process, however, is challenging due t...
http://arxiv.org/pdf/2303.03600v1
2023-03-07T02:31:57
2023-03-07T02:31:57
{ "title": [ -0.008522034, -0.0001436472, -0.062286377, -0.076049805, -0.007129669, -0.026000977, -0.07019043, 0.012924194, 0.011451721, 0.043151855, -0.022537231, -0.021560669, 0.022232056, -0.024536133, -0.03656006, -0.0059509277, 0.013114929, ...
http://arxiv.org/abs/2303.03593v1
ADELT: Transpilation Between Deep Learning Frameworks
Linyuan Gong, Jiayi Wang, Alvin Cheung
cs.CL, cs.LG
We propose Adversarial DEep Learning Transpiler (ADELT) for source-to-source transpilation between deep learning frameworks. Unlike prior approaches, we decouple the transpilation of code skeletons and the mapping of API keywords (an API function name or a parameter name). ADELT transpile code skeletons using few-shot ...
http://arxiv.org/pdf/2303.03593v1
2023-03-07T01:57:10
2023-03-07T01:57:10
{ "title": [ -0.009941101, 0.0060539246, -0.042816162, -0.070373535, 0.020889282, -0.010368347, -0.03161621, -0.02810669, -0.0040016174, 0.027908325, -0.006416321, 0.011009216, -0.009086609, -0.021881104, -0.023529053, 0.015838623, -0.0023479462, ...
http://arxiv.org/abs/2303.03548v1
Large Language Models as Zero-Shot Human Models for Human-Robot Interaction
Bowen Zhang, Harold Soh
cs.RO, cs.CL, cs.HC, cs.LG
Human models play a crucial role in human-robot interaction (HRI), enabling robots to consider the impact of their actions on people and plan their behavior accordingly. However, crafting good human models is challenging; capturing context-dependent human behavior requires significant prior knowledge and/or large amoun...
http://arxiv.org/pdf/2303.03548v1
2023-03-06T23:16:24
2023-03-06T23:16:24
{ "title": [ -0.016433716, -0.009895325, -0.031433105, -0.04824829, -0.009254456, -0.010383606, -0.04345703, 0.0036754608, -0.0013036728, 0.037628174, -0.011268616, -0.004070282, 0.0052604675, -0.0129776, -0.0440979, -0.0015029907, 0.0066184998, ...
http://arxiv.org/abs/2303.03542v1
Multi-resolution Interpretation and Diagnostics Tool for Natural Language Classifiers
Peyman Jalali, Nengfeng Zhou, Yufei Yu
cs.CL, cs.AI
Developing explainability methods for Natural Language Processing (NLP) models is a challenging task, for two main reasons. First, the high dimensionality of the data (large number of tokens) results in low coverage and in turn small contributions for the top tokens, compared to the overall model performance. Second, o...
http://arxiv.org/pdf/2303.03542v1
2023-03-06T22:59:02
2023-03-06T22:59:02
{ "title": [ -0.01260376, -0.0002142191, -0.025512695, -0.06982422, 0.006412506, -0.0029277802, -0.048950195, -0.0037155151, -0.0011281967, 0.017150879, -0.03213501, 0.0008678436, 0.006126404, -0.020019531, -0.029968262, -0.012039185, -0.015586853, ...
http://arxiv.org/abs/2303.06026v1
wav2vec and its current potential to Automatic Speech Recognition in German for the usage in Digital History: A comparative assessment of available ASR-technologies for the use in cultural heritage contexts
Michael Fleck, Wolfgang Göderle
eess.AS, cs.CL, cs.LG, cs.SD
In this case study we trained and published a state-of-the-art open-source model for Automatic Speech Recognition (ASR) for German to evaluate the current potential of this technology for the use in the larger context of Digital Humanities and cultural heritage indexation. Along with this paper we publish our wav2vec2 ...
http://arxiv.org/pdf/2303.06026v1
2023-03-06T22:24:31
2023-03-06T22:24:31
{ "title": [ 0.020050049, -0.009780884, -0.018829346, -0.002292633, -0.011039734, -0.027679443, 0.018951416, 0.00011104345, 0.028182983, 0.057250977, 0.0027828217, -0.0184021, 0.06451416, -0.046722412, -0.013679504, 0.022460938, 0.008987427, 0.04...
http://arxiv.org/abs/2303.03510v1
Guilt Detection in Text: A Step Towards Understanding Complex Emotions
Abdul Gafar Manuel Meque, Nisar Hussain, Grigori Sidorov, Alexander Gelbukh
cs.CL
We introduce a novel Natural Language Processing (NLP) task called Guilt detection, which focuses on detecting guilt in text. We identify guilt as a complex and vital emotion that has not been previously studied in NLP, and we aim to provide a more fine-grained analysis of it. To address the lack of publicly available ...
http://arxiv.org/pdf/2303.03510v1
2023-03-06T21:36:19
2023-03-06T21:36:19
{ "title": [ -0.0035209656, -0.0069389343, -0.026992798, -0.027893066, 0.00466156, -0.014266968, -0.0027713776, -0.005344391, 0.01576233, 0.0038223267, -0.021072388, -0.006839752, 0.004711151, -0.037475586, -0.037353516, -0.008544922, -0.017623901, ...
http://arxiv.org/abs/2303.03487v2
Two-stage Pipeline for Multilingual Dialect Detection
Ankit Vaidya, Aditya Kane
cs.CL, cs.AI
Dialect Identification is a crucial task for localizing various Large Language Models. This paper outlines our approach to the VarDial 2023 shared task. Here we have to identify three or two dialects from three languages each which results in a 9-way classification for Track-1 and 6-way classification for Track-2 respe...
http://arxiv.org/pdf/2303.03487v2
2023-03-28T17:55:52
2023-03-06T20:35:51
{ "title": [ -0.015434265, -0.0019330978, -0.029647827, -0.050750732, 0.0029563904, -0.0005302429, -0.044799805, 0.0015544891, -0.007221222, 0.043762207, -0.00630188, -0.011245728, 0.008422852, -0.028182983, -0.052459717, 0.0076446533, -0.0005059242,...
http://arxiv.org/abs/2303.03480v2
Can an Embodied Agent Find Your "Cat-shaped Mug"? LLM-Guided Exploration for Zero-Shot Object Navigation
Vishnu Sashank Dorbala, James F. Mullen Jr., Dinesh Manocha
cs.RO, cs.AI, cs.CL
We present LGX (Language-guided Exploration), a novel algorithm for Language-Driven Zero-Shot Object Goal Navigation (L-ZSON), where an embodied agent navigates to a uniquely described target object in a previously unseen environment. Our approach makes use of Large Language Models (LLMs) for this task by leveraging th...
http://arxiv.org/pdf/2303.03480v2
2023-11-05T20:49:11
2023-03-06T20:19:19
{ "title": [ 0.0024986267, -0.03237915, 0.014282227, 0.037750244, -0.06756592, -0.0309906, 0.005077362, -0.03225708, 0.009887695, 0.018936157, -0.023284912, 0.0034828186, 0.009002686, -0.036010742, -0.0061187744, -0.021835327, 0.013710022, 0.0126...
http://arxiv.org/abs/2303.05389v2
Depression Detection Using Digital Traces on Social Media: A Knowledge-aware Deep Learning Approach
Wenli Zhang, Jiaheng Xie, Zhu Zhang, Xiang Liu
cs.CL, cs.AI, cs.SI, stat.AP, H.4.m, K.5
Depression is a common disease worldwide. It is difficult to diagnose and continues to be underdiagnosed. Because depressed patients constantly share their symptoms, major life events, and treatments on social media, researchers are turning to user-generated digital traces on social media for depression detection. Such...
http://arxiv.org/pdf/2303.05389v2
2023-08-01T14:48:39
2023-03-06T20:08:07
{ "title": [ 0.027908325, 0.004360199, -0.0026416779, -0.055480957, 0.017562866, -0.0075263977, -0.022506714, -0.02243042, -0.0011100769, 0.02053833, -0.015296936, -0.0004274845, 0.014839172, -0.036712646, -0.03970337, -0.0049324036, 0.0028152466, ...
http://arxiv.org/abs/2303.03457v1
Spelling convention sensitivity in neural language models
Elizabeth Nielsen, Christo Kirov, Brian Roark
cs.CL
We examine whether large neural language models, trained on very large collections of varied English text, learn the potentially long-distance dependency of British versus American spelling conventions, i.e., whether spelling is consistently one or the other within model-generated strings. In contrast to long-distance ...
http://arxiv.org/pdf/2303.03457v1
2023-03-06T19:29:20
2023-03-06T19:29:20
{ "title": [ -0.033813477, 0.008102417, -0.028396606, -0.0070495605, -0.007587433, -0.010231018, -0.02571106, -0.020507812, 0.0026760101, 0.062561035, -0.013450623, -0.015289307, 0.03741455, -0.0028820038, -0.07116699, 0.020309448, -0.029678345, ...
http://arxiv.org/abs/2303.03363v2
Enhancing Activity Prediction Models in Drug Discovery with the Ability to Understand Human Language
Philipp Seidl, Andreu Vall, Sepp Hochreiter, Günter Klambauer
q-bio.BM, cs.CL, cs.LG, stat.ML
Activity and property prediction models are the central workhorses in drug discovery and materials sciences, but currently they have to be trained or fine-tuned for new tasks. Without training or fine-tuning, scientific language models could be used for such low-data tasks through their announced zero- and few-shot cap...
http://arxiv.org/pdf/2303.03363v2
2023-06-16T09:59:34
2023-03-06T18:49:09
{ "title": [ -0.022705078, -0.00881958, -0.05203247, -0.049102783, -0.0033340454, -0.02078247, -0.053100586, 0.010002136, 0.0016031265, 0.01461792, -0.011787415, -0.022201538, -0.013000488, -0.021621704, -0.04559326, -0.0061912537, -0.0066223145, ...
http://arxiv.org/abs/2303.03290v1
AmQA: Amharic Question Answering Dataset
Tilahun Abedissa, Ricardo Usbeck, Yaregal Assabie
cs.CL, cs.AI, cs.IR
Question Answering (QA) returns concise answers or answer lists from natural language text given a context document. Many resources go into curating QA datasets to advance robust models' development. There is a surge of QA datasets for languages like English, however, this is not true for Amharic. Amharic, the official...
http://arxiv.org/pdf/2303.03290v1
2023-03-06T17:06:50
2023-03-06T17:06:50
{ "title": [ -0.025390625, -0.020690918, -0.02420044, -0.06149292, -0.030151367, 0.0121154785, -0.061462402, 0.010215759, 0.0035934448, 0.03201294, -0.0077323914, -0.0017299652, 0.03277588, -0.035705566, -0.024978638, 0.02218628, -0.0043640137, 0...
http://arxiv.org/abs/2303.03283v2
The AI Ghostwriter Effect: When Users Do Not Perceive Ownership of AI-Generated Text But Self-Declare as Authors
Fiona Draxler, Anna Werner, Florian Lehmann, Matthias Hoppe, Albrecht Schmidt, Daniel Buschek, Robin Welsch
cs.HC, cs.CL
Human-AI interaction in text production increases complexity in authorship. In two empirical studies (n1 = 30 & n2 = 96), we investigate authorship and ownership in human-AI collaboration for personalized language generation. We show an AI Ghostwriter Effect: Users do not consider themselves the owners and authors of A...
http://arxiv.org/pdf/2303.03283v2
2023-11-07T13:40:12
2023-03-06T16:53:12
{ "title": [ 0.0005326271, 0.017868042, -0.028640747, 0.056243896, -0.019638062, -0.0423584, 0.04824829, -0.06549072, -0.031951904, 0.05053711, -0.024475098, 0.026397705, -0.010543823, -0.018936157, -0.03274536, -0.02168274, -0.036834717, 0.05749...
http://arxiv.org/abs/2303.03278v1
Faithfulness-Aware Decoding Strategies for Abstractive Summarization
David Wan, Mengwen Liu, Kathleen McKeown, Markus Dreyer, Mohit Bansal
cs.CL, cs.AI, cs.LG
Despite significant progress in understanding and improving faithfulness in abstractive summarization, the question of how decoding strategies affect faithfulness is less studied. We present a systematic study of the effect of generation techniques such as beam search and nucleus sampling on faithfulness in abstractive...
http://arxiv.org/pdf/2303.03278v1
2023-03-06T16:49:27
2023-03-06T16:49:27
{ "title": [ -0.011566162, 0.0028972626, -0.015289307, -0.09637451, -0.017547607, 0.022460938, -0.09741211, -0.014099121, 0.017715454, 0.07067871, 0.0058631897, 0.013214111, 0.016098022, -0.008056641, -0.06726074, 0.037841797, 0.005264282, 0.0354...
http://arxiv.org/abs/2303.05382v3
ChatGPT is on the Horizon: Could a Large Language Model be Suitable for Intelligent Traffic Safety Research and Applications?
Ou Zheng, Mohamed Abdel-Aty, Dongdong Wang, Zijin Wang, Shengxuan Ding
cs.CL, cs.AI, cs.LG
ChatGPT embarks on a new era of artificial intelligence and will revolutionize the way we approach intelligent traffic safety systems. This paper begins with a brief introduction about the development of large language models (LLMs). Next, we exemplify using ChatGPT to address key traffic safety issues. Furthermore, we...
http://arxiv.org/pdf/2303.05382v3
2023-09-05T18:13:24
2023-03-06T16:36:17
{ "title": [ -0.015396118, -0.008659363, -0.03829956, -0.032928467, -0.020095825, -0.0030937195, -0.012268066, 0.017608643, -0.030197144, 0.03930664, -0.03842163, 0.024383545, 0.02671814, -0.019165039, 0.01676941, -0.019866943, -0.0048828125, 0.0...
http://arxiv.org/abs/2303.03199v1
Choice Over Control: How Users Write with Large Language Models using Diegetic and Non-Diegetic Prompting
Hai Dang, Sven Goller, Florian Lehmann, Daniel Buschek
cs.HC, cs.CL, H.5.2; I.2.7
We propose a conceptual perspective on prompts for Large Language Models (LLMs) that distinguishes between (1) diegetic prompts (part of the narrative, e.g. "Once upon a time, I saw a fox..."), and (2) non-diegetic prompts (external, e.g. "Write about the adventures of the fox."). With this lens, we study how 129 crowd...
http://arxiv.org/pdf/2303.03199v1
2023-03-06T14:58:42
2023-03-06T14:58:42
{ "title": [ -0.0020599365, -0.0048980713, -0.0041923523, -0.024597168, -0.024856567, -0.0019741058, -0.03878784, -0.032226562, -0.030044556, 0.020187378, -0.042053223, 0.026016235, -0.002161026, -0.0038204193, -0.020751953, -0.018447876, -0.01306915...
http://arxiv.org/abs/2303.03171v1
Neighborhood Contrastive Transformer for Change Captioning
Yunbin Tu, Liang Li, Li Su, Ke Lu, Qingming Huang
cs.CV, cs.CL, cs.MM
Change captioning is to describe the semantic change between a pair of similar images in natural language. It is more challenging than general image captioning, because it requires capturing fine-grained change information while being immune to irrelevant viewpoint changes, and solving syntax ambiguity in change descri...
http://arxiv.org/pdf/2303.03171v1
2023-03-06T14:39:54
2023-03-06T14:39:54
{ "title": [ 0.043060303, 0.00033068657, -0.0042381287, -0.024169922, -0.028869629, -0.026657104, 0.0039405823, -0.037109375, -0.015792847, 0.06958008, -0.049957275, 0.006137848, 0.03765869, -0.011161804, -0.01259613, -0.02180481, 0.019622803, 0....
http://arxiv.org/abs/2303.03144v1
IPA-CLIP: Integrating Phonetic Priors into Vision and Language Pretraining
Chihaya Matsuhira, Marc A. Kastner, Takahiro Komamizu, Takatsugu Hirayama, Keisuke Doman, Yasutomo Kawanishi, Ichiro Ide
cs.MM, cs.CL, cs.SD, eess.AS
Recently, large-scale Vision and Language (V\&L) pretraining has become the standard backbone of many multimedia systems. While it has shown remarkable performance even in unseen situations, it often performs in ways not intuitive to humans. Particularly, they usually do not consider the pronunciation of the input, whi...
http://arxiv.org/pdf/2303.03144v1
2023-03-06T13:59:37
2023-03-06T13:59:37
{ "title": [ -0.00088739395, -0.008026123, -0.02015686, -0.052490234, -0.011383057, -0.001958847, -0.031982422, -0.003118515, -0.018508911, 0.014778137, -0.016159058, -0.000849247, -0.000664711, -0.041778564, -0.016983032, -0.01676941, 0.006385803, ...
http://arxiv.org/abs/2303.03124v2
IFAN: An Explainability-Focused Interaction Framework for Humans and NLP Models
Edoardo Mosca, Daryna Dementieva, Tohid Ebrahim Ajdari, Maximilian Kummeth, Kirill Gringauz, Yutong Zhou, Georg Groh
cs.CL, cs.AI
Interpretability and human oversight are fundamental pillars of deploying complex NLP models into real-world applications. However, applying explainability and human-in-the-loop methods requires technical proficiency. Despite existing toolkits for model understanding and analysis, options to integrate human feedback ar...
http://arxiv.org/pdf/2303.03124v2
2023-10-02T15:15:33
2023-03-06T13:37:59
{ "title": [ -0.0102005005, -0.0032749176, -0.014755249, -0.07501221, 0.0019454956, 0.019317627, -0.055358887, -0.0073432922, -0.00021481514, -0.015357971, -0.022857666, -0.00434494, -0.007896423, -0.01979065, -0.021194458, -0.014854431, -0.027236938...
http://arxiv.org/abs/2303.03103v1
Towards Zero-Shot Functional Compositionality of Language Models
Hangyeol Yu, Myeongho Jeong, Jamin Shin, Hyeongdon Moon, Juneyoung Park, Seungtaek Choi
cs.CL, cs.AI
Large Pre-trained Language Models (PLM) have become the most desirable starting point in the field of NLP, as they have become remarkably good at solving many individual tasks. Despite such success, in this paper, we argue that current paradigms of working with PLMs are neglecting a critical aspect of modeling human in...
http://arxiv.org/pdf/2303.03103v1
2023-03-06T13:15:25
2023-03-06T13:15:25
{ "title": [ -0.008857727, -0.013198853, -0.031829834, -0.10015869, -0.0019626617, 0.01576233, -0.095581055, 0.023910522, 0.012626648, 0.03918457, 0.001996994, -0.0017766953, 0.0029659271, -0.010528564, -0.067871094, 0.022140503, -0.022628784, 0....
http://arxiv.org/abs/2303.03053v1
Crowdsourcing on Sensitive Data with Privacy-Preserving Text Rewriting
Nina Mouhammad, Johannes Daxenberger, Benjamin Schiller, Ivan Habernal
cs.CL
Most tasks in NLP require labeled data. Data labeling is often done on crowdsourcing platforms due to scalability reasons. However, publishing data on public platforms can only be done if no privacy-relevant information is included. Textual data often contains sensitive information like person names or locations. In th...
http://arxiv.org/pdf/2303.03053v1
2023-03-06T11:54:58
2023-03-06T11:54:58
{ "title": [ -0.028442383, -0.0113220215, -0.042907715, -0.07727051, 0.009887695, 0.012756348, -0.046295166, 0.017425537, -0.038360596, 0.038360596, -0.0055618286, 0.025787354, -0.007583618, 0.0048294067, -0.039978027, -0.009407043, -0.012184143, ...
http://arxiv.org/abs/2303.03032v1
DeCap: Decoding CLIP Latents for Zero-Shot Captioning via Text-Only Training
Wei Li, Linchao Zhu, Longyin Wen, Yi Yang
cs.CV, cs.AI, cs.CL
Large-scale pre-trained multi-modal models (e.g., CLIP) demonstrate strong zero-shot transfer capability in many discriminative tasks. Their adaptation to zero-shot image-conditioned text generation tasks has drawn increasing interest. Prior arts approach to zero-shot captioning by either utilizing the existing large l...
http://arxiv.org/pdf/2303.03032v1
2023-03-06T11:02:47
2023-03-06T11:02:47
{ "title": [ -0.009147644, -0.0073394775, -0.023162842, -0.052856445, -0.018951416, -0.006626129, -0.018981934, -0.0009851456, 0.004798889, 0.04949951, -0.0140686035, 0.013435364, 0.005958557, -0.013282776, -0.03930664, 0.0036125183, 0.0020694733, ...
http://arxiv.org/abs/2303.03019v1
NxPlain: Web-based Tool for Discovery of Latent Concepts
Fahim Dalvi, Nadir Durrani, Hassan Sajjad, Tamim Jaban, Musab Husaini, Ummar Abbas
cs.CL
The proliferation of deep neural networks in various domains has seen an increased need for the interpretability of these models, especially in scenarios where fairness and trust are as important as model performance. A lot of independent work is being carried out to: i) analyze what linguistic and non-linguistic knowl...
http://arxiv.org/pdf/2303.03019v1
2023-03-06T10:45:24
2023-03-06T10:45:24
{ "title": [ -0.00869751, -0.007972717, -0.05819702, -0.03933716, 0.010726929, -0.02279663, 0.0030269623, -0.01889038, 0.0047950745, 0.014099121, -0.038635254, 0.015312195, -0.013038635, -0.018707275, -0.019699097, -0.0045814514, -0.0074424744, 0...
http://arxiv.org/abs/2303.03004v4
xCodeEval: A Large Scale Multilingual Multitask Benchmark for Code Understanding, Generation, Translation and Retrieval
Mohammad Abdullah Matin Khan, M Saiful Bari, Xuan Long Do, Weishi Wang, Md Rizwan Parvez, Shafiq Joty
cs.CL
Recently, pre-trained large language models (LLMs) have shown impressive abilities in generating codes from natural language descriptions, repairing buggy codes, translating codes between languages, and retrieving relevant code segments. However, the evaluation of these models has often been performed in a scattered wa...
http://arxiv.org/pdf/2303.03004v4
2023-11-06T07:16:58
2023-03-06T10:08:51
{ "title": [ -0.057800293, -0.022872925, -0.031463623, -0.045013428, -0.033966064, 0.0070381165, -0.09051514, -0.0030288696, 0.010848999, 0.045532227, 0.0017585754, 0.014602661, 0.03050232, -0.035308838, -0.01108551, 0.018035889, -0.008865356, 0....
http://arxiv.org/abs/2303.02995v1
HiCLIP: Contrastive Language-Image Pretraining with Hierarchy-aware Attention
Shijie Geng, Jianbo Yuan, Yu Tian, Yuxiao Chen, Yongfeng Zhang
cs.CV, cs.CL, cs.LG
The success of large-scale contrastive vision-language pretraining (CLIP) has benefited both visual recognition and multimodal content understanding. The concise design brings CLIP the advantage in inference efficiency against other vision-language models with heavier cross-attention fusion layers, making it a popular ...
http://arxiv.org/pdf/2303.02995v1
2023-03-06T09:44:01
2023-03-06T09:44:01
{ "title": [ 0.042907715, -0.0008239746, -0.0010137558, -0.06713867, 0.0016555786, -0.010368347, -0.011566162, -0.02432251, 0.020706177, 0.036590576, -0.00818634, 0.030319214, 0.0076141357, -0.01612854, -0.03353882, 0.016540527, -0.0002784729, 0....
http://arxiv.org/abs/2303.02915v1
GlobalNER: Incorporating Non-local Information into Named Entity Recognition
Chiao-Wei Hsu, Keh-Yih Su
cs.CL, cs.AI
Nowadays, many Natural Language Processing (NLP) tasks see the demand for incorporating knowledge external to the local information to further improve the performance. However, there is little related work on Named Entity Recognition (NER), which is one of the foundations of NLP. Specifically, no studies were conducted...
http://arxiv.org/pdf/2303.02915v1
2023-03-06T06:20:55
2023-03-06T06:20:55
{ "title": [ -0.015106201, -0.0010271072, -0.046173096, -0.08984375, 0.027633667, 0.0053710938, -0.040008545, 0.0010156631, -0.004562378, 0.015808105, -0.027923584, 0.025039673, 0.006576538, -0.0051879883, -0.011177063, -0.0019760132, -0.004852295, ...
http://arxiv.org/abs/2303.02913v1
OpenICL: An Open-Source Framework for In-context Learning
Zhenyu Wu, YaoXiang Wang, Jiacheng Ye, Jiangtao Feng, Jingjing Xu, Yu Qiao, Zhiyong Wu
cs.CL
In recent years, In-context Learning (ICL) has gained increasing attention and emerged as the new paradigm for large language model (LLM) evaluation. Unlike traditional fine-tuning methods, ICL instead adapts the pre-trained models to unseen tasks without any parameter updates. However, the implementation of ICL is sop...
http://arxiv.org/pdf/2303.02913v1
2023-03-06T06:20:25
2023-03-06T06:20:25
{ "title": [ -0.009437561, 0.0022773743, -0.05038452, -0.051849365, 0.022109985, -0.006011963, -0.03086853, -0.028900146, -0.018707275, 0.04345703, -0.021850586, 0.0022144318, -0.0047302246, -0.023452759, -0.028961182, -0.00920105, -0.0006418228, ...
http://arxiv.org/abs/2303.02909v2
Dynamic Prompting: A Unified Framework for Prompt Tuning
Xianjun Yang, Wei Cheng, Xujiang Zhao, Wenchao Yu, Linda Petzold, Haifeng Chen
cs.CL, cs.AI
It has been demonstrated that the art of prompt tuning is highly effective in efficiently extracting knowledge from pretrained foundation models, encompassing pretrained language models (PLMs), vision pretrained models, and vision-language (V-L) models. However, the efficacy of employing fixed soft prompts with a prede...
http://arxiv.org/pdf/2303.02909v2
2023-05-27T01:29:27
2023-03-06T06:04:46
{ "title": [ -0.017120361, -0.015960693, -0.025283813, -0.08312988, -0.0038547516, 0.010269165, -0.056488037, -0.006122589, -0.016738892, 0.014793396, -0.0124435425, -0.0025558472, -0.02041626, -0.035217285, -0.026351929, -0.002161026, 0.015090942, ...
http://arxiv.org/abs/2303.03919v1
Data Portraits: Recording Foundation Model Training Data
Marc Marone, Benjamin Van Durme
cs.LG, cs.CL
Foundation models are trained on increasingly immense and opaque datasets. Even while these models are now key in AI system building, it can be difficult to answer the straightforward question: has the model already encountered a given example during training? We therefore propose a widespread adoption of Data Portrait...
http://arxiv.org/pdf/2303.03919v1
2023-03-06T04:22:33
2023-03-06T04:22:33
{ "title": [ -0.0109939575, 0.0057411194, -0.06921387, -0.045196533, -0.017868042, -0.034851074, -0.044677734, -0.0060691833, -0.030380249, 0.016799927, -0.034606934, 0.0154418945, -0.025924683, -0.0357666, -0.002784729, -0.03137207, -0.0033874512, ...
http://arxiv.org/abs/2303.02861v1
Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning
Zhen Wang, Rameswar Panda, Leonid Karlinsky, Rogerio Feris, Huan Sun, Yoon Kim
cs.CL
Prompt tuning, in which a base pretrained model is adapted to each task via conditioning on learned prompt vectors, has emerged as a promising approach for efficiently adapting large language models to multiple downstream tasks. However, existing methods typically learn soft prompt vectors from scratch, and it has not ...
http://arxiv.org/pdf/2303.02861v1
2023-03-06T03:25:59
2023-03-06T03:25:59
{ "title": [ -0.003168106, -0.0020828247, -0.045928955, -0.06616211, -0.005088806, -0.021224976, -0.051452637, -0.005870819, -0.024124146, 0.012893677, 0.0075912476, 0.017410278, -0.026535034, -0.039794922, 0.00045990944, 0.0099105835, 0.00737381, ...
http://arxiv.org/abs/2303.02860v1
A Multi-Grained Self-Interpretable Symbolic-Neural Model For Single/Multi-Labeled Text Classification
Xiang Hu, Xinyu Kong, Kewei Tu
cs.CL, cs.LG
Deep neural networks based on layer-stacking architectures have historically suffered from poor inherent interpretability. Meanwhile, symbolic probabilistic models function with clear interpretability, but how to combine them with neural networks to enhance their performance remains to be explored. In this paper, we tr...
http://arxiv.org/pdf/2303.02860v1
2023-03-06T03:25:43
2023-03-06T03:25:43
{ "title": [ -0.01209259, 0.009269714, -0.017807007, -0.019042969, -0.012763977, 0.0046157837, -0.029907227, -0.028274536, -0.01675415, 0.03463745, -0.00086927414, 0.0033111572, -0.02658081, -0.010437012, -0.058288574, -0.0181427, -0.014526367, 0...
http://arxiv.org/abs/2303.02846v2
Reducing Spurious Correlations for Aspect-Based Sentiment Analysis with Variational Information Bottleneck and Contrastive Learning
Mingshan Chang, Min Yang, Qingshan Jiang, Ruifeng Xu
cs.CL, cs.AI
Deep learning techniques have dominated the literature on aspect-based sentiment analysis (ABSA), yielding state-of-the-art results. However, these deep models generally suffer from spurious correlation problems between input features and output labels, which creates significant barriers to robustness and generalizatio...
http://arxiv.org/pdf/2303.02846v2
2023-03-12T14:51:17
2023-03-06T02:52:37
{ "title": [ 0.0012636185, 0.005508423, -0.034851074, -0.04663086, 0.0107421875, -0.0044555664, -0.01083374, 0.000024199486, 0.04260254, 0.0647583, -0.008880615, -0.006679535, -0.0034828186, -0.00031971931, -0.10723877, 0.0018758774, -0.012290955, ...
http://arxiv.org/abs/2303.02841v2
Model-Agnostic Meta-Learning for Natural Language Understanding Tasks in Finance
Bixing Yan, Shaoling Chen, Yuxuan He, Zhihan Li
cs.CL, cs.AI, cs.LG
Natural language understanding(NLU) is challenging for finance due to the lack of annotated data and the specialized language in that domain. As a result, researchers have proposed to use pre-trained language model and multi-task learning to learn robust representations. However, aggressive fine-tuning often causes ove...
http://arxiv.org/pdf/2303.02841v2
2023-03-26T19:45:25
2023-03-06T02:24:48
{ "title": [ -0.027694702, -0.005279541, -0.0335083, -0.03387451, -0.01838684, -0.00207901, -0.052581787, -0.028839111, 0.008522034, 0.03955078, -0.03074646, -0.007575989, -0.008735657, -0.006210327, -0.07745361, -0.013748169, -0.012519836, 0.030...
http://arxiv.org/abs/2303.02758v1
WADER at SemEval-2023 Task 9: A Weak-labelling framework for Data augmentation in tExt Regression Tasks
Manan Suri, Aaryak Garg, Divya Chaudhary, Ian Gorton, Bijendra Kumar
cs.CL, cs.LG
Intimacy is an essential element of human relationships and language is a crucial means of conveying it. Textual intimacy analysis can reveal social norms in different contexts and serve as a benchmark for testing computational models' ability to understand social information. In this paper, we propose a novel weak-lab...
http://arxiv.org/pdf/2303.02758v1
2023-03-05T19:45:42
2023-03-05T19:45:42
{ "title": [ -0.014183044, -0.0059509277, -0.03692627, -0.03994751, 0.00018715858, -0.0017280579, -0.02734375, -0.011482239, -0.00548172, 0.019302368, -0.033477783, 0.04046631, -0.013160706, -0.011230469, -0.0135650635, 0.0060806274, -0.018127441, ...
http://arxiv.org/abs/2303.02707v1
FQP 2.0: Industry Trend Analysis via Hierarchical Financial Data
Hongyin Zhu
cs.CL, cs.CY
Analyzing trends across industries is critical to maintaining a healthy and stable economy. Previous research has mainly analyzed official statistics, which are more accurate but not necessarily real-time. In this paper, we propose a method for analyzing industry trends using stock market data. The difficulty of this t...
http://arxiv.org/pdf/2303.02707v1
2023-03-05T16:17:56
2023-03-05T16:17:56
{ "title": [ -0.022262573, -0.00806427, -0.076538086, -0.060943604, 0.022277832, -0.016647339, -0.02015686, -0.012069702, -0.013298035, 0.020370483, -0.026046753, 0.010231018, -0.025390625, -0.009819031, -0.04638672, -0.018295288, -0.011947632, 0...
http://arxiv.org/abs/2303.02677v1
Mining both Commonality and Specificity from Multiple Documents for Multi-Document Summarization
Bing Ma
cs.CL, cs.AI
The multi-document summarization task requires the designed summarizer to generate a short text that covers the important information of original documents and satisfies content diversity. This paper proposes a multi-document summarization approach based on hierarchical clustering of documents. It utilizes the construc...
http://arxiv.org/pdf/2303.02677v1
2023-03-05T14:25:05
2023-03-05T14:25:05
{ "title": [ -0.003932953, 0.0012073517, -0.03375244, -0.0869751, 0.0047950745, 0.01474762, -0.058380127, -0.0054779053, -0.003622055, 0.046936035, -0.017120361, 0.038482666, -0.0055618286, 0.0063323975, -0.04296875, 0.007835388, -0.02104187, 0.0...
http://arxiv.org/abs/2303.02601v1
Knowledge-Based Counterfactual Queries for Visual Question Answering
Theodoti Stoikou, Maria Lymperaiou, Giorgos Stamou
cs.CL
Visual Question Answering (VQA) has been a popular task that combines vision and language, with numerous relevant implementations in literature. Even though there are some attempts that approach explainability and robustness issues in VQA models, very few of them employ counterfactuals as a means of probing such challe...
http://arxiv.org/pdf/2303.02601v1
2023-03-05T08:00:30
2023-03-05T08:00:30
{ "title": [ -0.0042419434, -0.0012598038, -0.013206482, -0.022888184, -0.01058197, -0.007545471, -0.013130188, -0.018798828, 0.052520752, 0.026794434, -0.06201172, -0.010437012, -0.006340027, -0.00021362305, -0.078430176, -0.024810791, 0.022384644, ...
http://arxiv.org/abs/2303.05400v1
Prompt-Based Learning for Thread Structure Prediction in Cybersecurity Forums
Kazuaki Kashihara, Kuntal Kumar Pal, Chitta Baral, Robert P Trevino
cs.CL, cs.AI, cs.CR
With recent trends indicating cyber crimes increasing in both frequency and cost, it is imperative to develop new methods that leverage data-rich hacker forums to assist in combating ever evolving cyber threats. Defining interactions within these forums is critical as it facilitates identifying highly skilled users, wh...
http://arxiv.org/pdf/2303.05400v1
2023-03-05T04:26:17
2023-03-05T04:26:17
{ "title": [ 0.00012874603, 0.00062561035, -0.04534912, -0.03817749, -0.003255844, -0.011512756, -0.030975342, 0.0048446655, -0.026535034, 0.019119263, -0.0039749146, 0.012069702, -0.023956299, -0.043182373, -0.00623703, -0.015991211, 0.012214661, ...
http://arxiv.org/abs/2303.02577v2
Effectiveness of Data Augmentation for Parameter Efficient Tuning with Limited Data
Stephen Obadinma, Hongyu Guo, Xiaodan Zhu
cs.CL
Recent work has demonstrated that using parameter efficient tuning techniques such as prefix tuning (or P-tuning) on pretrained language models can yield performance that is comparable or superior to fine-tuning while dramatically reducing trainable parameters. Nevertheless, the effectiveness of such methods under the ...
http://arxiv.org/pdf/2303.02577v2
2023-06-29T06:13:01
2023-03-05T04:12:17
{ "title": [ 0.0046539307, -0.0051994324, -0.046539307, -0.06854248, 0.000813961, -0.012573242, -0.033447266, 0.0033226013, -0.0035114288, 0.008323669, -0.02519226, 0.026290894, 0.0016107559, -0.027252197, -0.009010315, 0.0129852295, 0.004436493, ...
http://arxiv.org/abs/2303.02563v4
FinXABSA: Explainable Finance through Aspect-Based Sentiment Analysis
Keane Ong, Wihan van der Heever, Ranjan Satapathy, Erik Cambria, Gianmarco Mengaldo
cs.CL
This paper presents a novel approach for explainability in financial analysis by deriving financially-explainable statistical relationships through aspect-based sentiment analysis, Pearson correlation, Granger causality & uncertainty coefficient. The proposed methodology involves constructing an aspect list from financ...
http://arxiv.org/pdf/2303.02563v4
2023-10-14T15:01:11
2023-03-05T03:18:56
{ "title": [ 0.0018978119, -0.011581421, -0.03387451, -0.061431885, 0.00894928, 0.00046253204, -0.019439697, -0.0019340515, 0.011268616, 0.041381836, -0.020690918, -0.013847351, 0.01776123, -0.0063209534, -0.064697266, -0.005836487, -0.021865845, ...
http://arxiv.org/abs/2303.02513v1
Model-Agnostic Meta-Learning for Multilingual Hate Speech Detection
Md Rabiul Awal, Roy Ka-Wei Lee, Eshaan Tanwar, Tanmay Garg, Tanmoy Chakraborty
cs.CL, cs.SI
Hate speech in social media is a growing phenomenon, and detecting such toxic content has recently gained significant traction in the research community. Existing studies have explored fine-tuning language models (LMs) to perform hate speech detection, and these solutions have yielded significant performance. However, ...
http://arxiv.org/pdf/2303.02513v1
2023-03-04T22:28:29
2023-03-04T22:28:29
{ "title": [ 0.0017995834, 0.008338928, -0.016571045, -0.037109375, -0.0025749207, -0.004261017, -0.04067993, -0.030181885, -0.004753113, 0.054138184, -0.0055160522, 0.0070762634, 0.0054779053, -0.007965088, -0.05670166, 0.005794525, -0.010108948, ...
http://arxiv.org/abs/2303.07103v2
Could a Large Language Model be Conscious?
David J. Chalmers
cs.AI, cs.CL, cs.LG
There has recently been widespread discussion of whether large language models might be sentient or conscious. Should we take this idea seriously? I will break down the strongest reasons for and against. Given mainstream assumptions in the science of consciousness, there are significant obstacles to consciousness in cu...
http://arxiv.org/pdf/2303.07103v2
2023-04-29T18:35:39
2023-03-04T19:14:20
{ "title": [ -0.018539429, 0.00051546097, -0.03213501, -0.012496948, -0.03970337, -0.01158905, -0.03817749, 0.007522583, 0.031463623, -0.00009781122, -0.051239014, -0.044952393, 0.0018920898, -0.03503418, -0.002922058, -0.032714844, 0.0039253235, ...
http://arxiv.org/abs/2303.02472v1
ESD: Expected Squared Difference as a Tuning-Free Trainable Calibration Measure
Hee Suk Yoon, Joshua Tian Jin Tee, Eunseop Yoon, Sunjae Yoon, Gwangsu Kim, Yingzhen Li, Chang D. Yoo
cs.LG, cs.AI, cs.CL, cs.CV
Studies have shown that modern neural networks tend to be poorly calibrated due to over-confident predictions. Traditionally, post-processing methods have been used to calibrate the model after training. In recent years, various trainable calibration measures have been proposed to incorporate them directly into the tra...
http://arxiv.org/pdf/2303.02472v1
2023-03-04T18:06:36
2023-03-04T18:06:36
{ "title": [ -0.03540039, -0.014297485, -0.047943115, -0.025314331, -0.0140686035, -0.009353638, -0.016113281, -0.03378296, 0.020904541, -0.0115737915, -0.013771057, 0.0008773804, -0.015991211, -0.054626465, -0.012199402, 0.015914917, 0.020568848, ...
http://arxiv.org/abs/2303.02469v1
Variational Quantum Classifiers for Natural-Language Text
Daniel T. Chang
cs.CL, cs.LG
As part of the recent research effort on quantum natural language processing (QNLP), variational quantum sentence classifiers (VQSCs) have been implemented and supported in lambeq / DisCoPy, based on the DisCoCat model of sentence meaning. We discuss in some detail VQSCs, including category theory, DisCoCat for modelin...
http://arxiv.org/pdf/2303.02469v1
2023-03-04T18:00:05
2023-03-04T18:00:05
{ "title": [ -0.024810791, 0.0059013367, -0.023544312, -0.004173279, 0.004764557, -0.01852417, -0.00793457, -0.031433105, 0.007865906, 0.0440979, 0.016326904, 0.0028762817, -0.009399414, -0.019958496, -0.076416016, 0.002155304, -0.027069092, 0.05...
http://arxiv.org/abs/2303.02468v3
Lon-ea at SemEval-2023 Task 11: A Comparison of Activation Functions for Soft and Hard Label Prediction
Peyman Hosseini, Mehran Hosseini, Sana Sabah Al-Azzawi, Marcus Liwicki, Ignacio Castro, Matthew Purver
cs.CL, cs.LG
We study the influence of different activation functions in the output layer of deep neural network models for soft and hard label prediction in the learning with disagreement task. In this task, the goal is to quantify the amount of disagreement via predicting soft labels. To predict the soft labels, we use BERT-based...
http://arxiv.org/pdf/2303.02468v3
2023-04-24T17:50:11
2023-03-04T17:59:43
{ "title": [ -0.00957489, -0.016326904, -0.026535034, -0.050689697, -0.009933472, -0.009963989, -0.023910522, 0.008766174, -0.0062942505, 0.029190063, -0.008674622, 0.02633667, -0.0023708344, -0.031280518, -0.003452301, 0.014907837, 0.008956909, ...
http://arxiv.org/abs/2303.02427v2
Self-tuning hyper-parameters for unsupervised cross-lingual tokenization
Anton Kolonin
cs.CL, cs.AI
We explore the possibility of meta-learning for the language-independent unsupervised tokenization problem for English, Russian, and Chinese. We implement the meta-learning approach for automatic determination of hyper-parameters of the unsupervised tokenization model proposed in earlier works, relying on various human...
http://arxiv.org/pdf/2303.02427v2
2023-04-04T15:34:50
2023-03-04T14:23:02
{ "title": [ -0.014602661, 0.0154953, -0.033203125, -0.078430176, 0.017578125, 0.018493652, -0.058502197, -0.008232117, 0.0029850006, 0.03479004, -0.014007568, 0.009262085, 0.011917114, -0.009086609, -0.037841797, 0.0071372986, -0.008399963, 0.05...
http://arxiv.org/abs/2303.02411v1
The Contribution of Knowledge in Visiolinguistic Learning: A Survey on Tasks and Challenges
Maria Lymperaiou, Giorgos Stamou
cs.CL, cs.AI, cs.CV
Recent advancements in visiolinguistic (VL) learning have allowed the development of multiple models and techniques that offer several impressive implementations, able to currently resolve a variety of tasks that require the collaboration of vision and language. Current datasets used for VL pre-training only contain a ...
http://arxiv.org/pdf/2303.02411v1
2023-03-04T13:12:18
2023-03-04T13:12:18
{ "title": [ 0.0030612946, -0.028060913, -0.024795532, -0.037994385, -0.010643005, -0.03451538, 0.004470825, 0.011192322, -0.020095825, 0.015914917, -0.03253174, 0.024795532, -0.03677368, -0.01524353, -0.041259766, -0.023544312, 0.006198883, 0.03...
http://arxiv.org/abs/2303.02399v1
RweetMiner: Automatic identification and categorization of help requests on twitter during disasters
Irfan Ullah, Sharifullah Khan, Muhammad Imran, Young-Koo Lee
cs.CL, cs.AI
Catastrophic events create uncertain situations for humanitarian organizations locating and providing aid to affected people. Many people turn to social media during disasters for requesting help and/or providing relief to others. However, the majority of social media posts seeking help could not properly be detected a...
http://arxiv.org/pdf/2303.02399v1
2023-03-04T12:21:45
2023-03-04T12:21:45
{ "title": [ 0.016586304, 0.020721436, -0.029022217, -0.07312012, 0.030212402, -0.0066070557, -0.016952515, -0.0082473755, -0.03173828, 0.05239868, -0.035980225, 0.039855957, 0.02218628, -0.009811401, -0.020767212, 0.00040841103, -0.010932922, 0....
http://arxiv.org/abs/2303.02357v1
DiTTO: A Feature Representation Imitation Approach for Improving Cross-Lingual Transfer
Shanu Kumar, Abbaraju Soujanya, Sandipan Dandapat, Sunayana Sitaram, Monojit Choudhury
cs.CL, cs.AI
Zero-shot cross-lingual transfer is promising, however has been shown to be sub-optimal, with inferior transfer performance across low-resource languages. In this work, we envision languages as domains for improving zero-shot transfer by jointly reducing the feature incongruity between the source and the target languag...
http://arxiv.org/pdf/2303.02357v1
2023-03-04T08:42:50
2023-03-04T08:42:50
{ "title": [ -0.0017061234, -0.012329102, -0.0022201538, -0.04046631, 0.0040130615, 0.006198883, -0.028152466, -0.025131226, 0.022491455, 0.032165527, -0.01399231, 0.0042533875, 0.03894043, -0.02609253, -0.029693604, 0.013641357, -0.009437561, 0....
http://arxiv.org/abs/2303.05398v1
MathPrompter: Mathematical Reasoning using Large Language Models
Shima Imani, Liang Du, Harsh Shrivastava
cs.CL, cs.AI
Large Language Models (LLMs) have limited performance when solving arithmetic reasoning tasks and often provide incorrect answers. Unlike natural language understanding, math problems typically have a single correct answer, making the task of generating accurate solutions more challenging for LLMs. To the best of our k...
http://arxiv.org/pdf/2303.05398v1
2023-03-04T04:43:49
2023-03-04T04:43:49
{ "title": [ -0.031051636, -0.016647339, -0.044403076, -0.061401367, 0.011360168, -0.0035591125, -0.037109375, -0.0013036728, 0.01625061, 0.01309967, -0.013923645, -0.013122559, 0.013404846, -0.028442383, -0.011550903, 0.0072250366, -0.0021247864, ...
http://arxiv.org/abs/2303.02260v2
Learning to reason over visual objects
Shanka Subhra Mondal, Taylor Webb, Jonathan D. Cohen
cs.CV, cs.CL
A core component of human intelligence is the ability to identify abstract patterns inherent in complex, high-dimensional perceptual data, as exemplified by visual reasoning tasks such as Raven's Progressive Matrices (RPM). Motivated by the goal of designing AI systems with this capacity, recent work has focused on eva...
http://arxiv.org/pdf/2303.02260v2
2023-10-26T21:24:47
2023-03-03T23:19:42
{ "title": [ -0.0010843277, -0.011543274, -0.023727417, -0.046844482, -0.0046844482, -0.023254395, 0.0040245056, -0.00793457, 0.0033779144, 0.0006828308, -0.047424316, -0.009513855, -0.036315918, -0.01222229, -0.037109375, -0.027267456, 0.0102005005,...
http://arxiv.org/abs/2303.02242v2
TrojText: Test-time Invisible Textual Trojan Insertion
Qian Lou, Yepeng Liu, Bo Feng
cs.CL
In Natural Language Processing (NLP), intelligent neuron models can be susceptible to textual Trojan attacks. Such attacks occur when Trojan models behave normally for standard inputs but generate malicious output for inputs that contain a specific trigger. Syntactic-structure triggers, which are invisible, are becomin...
http://arxiv.org/pdf/2303.02242v2
2023-08-22T02:34:19
2023-03-03T22:19:22
{ "title": [ -0.004508972, -0.008232117, -0.027267456, -0.050567627, -0.0029735565, -0.002872467, -0.009651184, -0.008918762, -0.05239868, 0.049804688, -0.010276794, 0.024673462, 0.014053345, -0.015472412, -0.011924744, -0.021392822, -0.02571106, ...
http://arxiv.org/abs/2303.02206v2
Domain Specific Question Answering Over Knowledge Graphs Using Logical Programming and Large Language Models
Navid Madani, Rohini K. Srihari, Kenneth Joseph
cs.LG, cs.CL
Answering questions over domain-specific graphs requires a tailored approach due to the limited number of relations and the specific nature of the domain. Our approach integrates classic logical programming languages into large language models (LLMs), enabling the utilization of logical reasoning capabilities to tackle...
http://arxiv.org/pdf/2303.02206v2
2023-08-23T14:23:48
2023-03-03T20:35:38
{ "title": [ -0.037872314, -0.017196655, -0.060821533, -0.065979004, -0.017318726, 0.0075569153, -0.07543945, 0.021102905, 0.019012451, 0.026443481, -0.02468872, -0.01184845, 0.0064697266, -0.010276794, -0.052886963, -0.0049743652, -0.017333984, ...
http://arxiv.org/abs/2303.02198v1
Exploring Data Augmentation Methods on Social Media Corpora
Isabel Garcia Pietri, Kineret Stanley
cs.CL
Data augmentation has proven widely effective in computer vision. In Natural Language Processing (NLP) data augmentation remains an area of active research. There is no widely accepted augmentation technique that works well across tasks and model architectures. In this paper we explore data augmentation techniques in t...
http://arxiv.org/pdf/2303.02198v1
2023-03-03T20:15:35
2023-03-03T20:15:35
{ "title": [ 0.017547607, 0.015823364, -0.0524292, -0.055664062, 0.031402588, -0.023117065, -0.0040283203, 0.004966736, -0.008872986, 0.023605347, -0.050598145, 0.035949707, -0.018600464, 0.00617218, -0.02508545, -0.011146545, -0.016586304, 0.072...
http://arxiv.org/abs/2303.02151v1
Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners
Renrui Zhang, Xiangfei Hu, Bohao Li, Siyuan Huang, Hanqiu Deng, Hongsheng Li, Yu Qiao, Peng Gao
cs.CV, cs.CL
Visual recognition in low-data regimes requires deep neural networks to learn generalized representations from limited training samples. Recently, CLIP-based methods have shown promising few-shot performance benefited from the contrastive language-image pre-training. We then question, if the more diverse pre-training k...
http://arxiv.org/pdf/2303.02151v1
2023-03-03T18:58:16
2023-03-03T18:58:16
{ "title": [ 0.021057129, -0.0074653625, -0.04425049, -0.06173706, -0.013267517, -0.02746582, -0.050750732, -0.0079956055, -0.010536194, -0.009635925, -0.045654297, 0.025024414, -0.057861328, -0.03186035, 0.0050735474, -0.04574585, 0.012275696, 0...
http://arxiv.org/abs/2303.09397v1
Cryptocurrency Price Prediction using Twitter Sentiment Analysis
Haritha GB, Sahana N. B
q-fin.ST, cs.AI, cs.CL, cs.LG
The cryptocurrency ecosystem has been the centre of discussion on many social media platforms, following its noted volatility and varied opinions. Twitter is rapidly being utilised as a news source and a medium for bitcoin discussion. Our algorithm seeks to use historical prices and sentiment of tweets to forecast the ...
http://arxiv.org/pdf/2303.09397v1
2023-03-03T18:42:01
2023-03-03T18:42:01
{ "title": [ -0.012290955, 0.0065193176, -0.03741455, -0.028549194, 0.0053215027, 0.0012283325, -0.0025978088, -0.023468018, -0.0064353943, 0.06756592, -0.028198242, -0.0054130554, 0.0135269165, 0.0014467239, -0.072387695, -0.0056495667, -0.015960693...
http://arxiv.org/abs/2303.03177v1
Pre-trained Model Representations and their Robustness against Noise for Speech Emotion Analysis
Vikramjit Mitra, Vasudha Kowtha, Hsiang-Yun Sherry Chien, Erdrin Azemi, Carlos Avendano
eess.AS, cs.CL, cs.LG, cs.SD
Pre-trained model representations have demonstrated state-of-the-art performance in speech recognition, natural language processing, and other applications. Speech models, such as Bidirectional Encoder Representations from Transformers (BERT) and Hidden units BERT (HuBERT), have enabled generating lexical and acoustic ...
http://arxiv.org/pdf/2303.03177v1
2023-03-03T18:22:32
2023-03-03T18:22:32
{ "title": [ 0.029785156, 0.009346008, -0.028686523, -0.08074951, 0.019592285, -0.01763916, -0.040100098, -0.0016450882, 0.026657104, 0.047821045, -0.008735657, -0.0077819824, 0.036987305, -0.02619934, -0.06100464, 0.01637268, 0.0053977966, 0.014...
http://arxiv.org/abs/2303.02078v1
Who could be behind QAnon? Authorship attribution with supervised machine-learning
Florian Cafiero, Jean-Baptiste Camps
cs.CL
A series of social media posts signed under the pseudonym "Q", started a movement known as QAnon, which led some of its most radical supporters to violent and illegal actions. To identify the person(s) behind Q, we evaluate the coincidence between the linguistic properties of the texts written by Q and to those written...
http://arxiv.org/pdf/2303.02078v1
2023-03-03T16:50:12
2023-03-03T16:50:12
{ "title": [ -0.031463623, 0.00907135, -0.060638428, -0.012611389, 0.0030155182, -0.022109985, -0.007949829, -0.024902344, -0.06311035, 0.015945435, 0.015808105, 0.008460999, 0.0012283325, -0.04147339, 0.018127441, -0.021820068, -0.021408081, 0.0...
http://arxiv.org/abs/2303.03186v1
Will Affective Computing Emerge from Foundation Models and General AI? A First Evaluation on ChatGPT
Mostafa M. Amin, Erik Cambria, Björn W. Schuller
cs.CL, cs.AI
ChatGPT has shown the potential of emerging general artificial intelligence capabilities, as it has demonstrated competent performance across many natural language processing tasks. In this work, we evaluate the capabilities of ChatGPT to perform text classification on three affective computing problems, namely, big-fi...
http://arxiv.org/pdf/2303.03186v1
2023-03-03T16:11:37
2023-03-03T16:11:37
{ "title": [ -0.017074585, -0.013191223, -0.012413025, 0.0016403198, -0.012969971, -0.015281677, 0.025238037, -0.01902771, 0.028381348, 0.027023315, -0.039031982, 0.010070801, 0.0013828278, -0.013923645, -0.04815674, 0.004989624, -0.017105103, 0....
http://arxiv.org/abs/2303.01912v2
Ancient Chinese Word Segmentation and Part-of-Speech Tagging Using Distant Supervision
Shuo Feng, Piji Li
cs.CL
Ancient Chinese word segmentation (WSG) and part-of-speech tagging (POS) are important to study ancient Chinese, but the amount of ancient Chinese WSG and POS tagging data is still rare. In this paper, we propose a novel augmentation method of ancient Chinese WSG and POS tagging data using distant supervision over para...
http://arxiv.org/pdf/2303.01912v2
2023-03-06T09:29:40
2023-03-03T13:24:17
{ "title": [ -0.01902771, 0.019897461, -0.022445679, -0.032226562, 0.018341064, -0.0046043396, -0.013038635, -0.014328003, -0.0036792755, 0.026855469, -0.048034668, 0.01638794, 0.014015198, -0.016479492, -0.03439331, -0.004501343, -0.007949829, 0...
http://arxiv.org/abs/2303.01911v2
Investigating the Translation Performance of a Large Multilingual Language Model: the Case of BLOOM
Rachel Bawden, François Yvon
cs.CL
The NLP community recently saw the release of a new large open-access multilingual language model, BLOOM (BigScience et al., 2022) covering 46 languages. We focus on BLOOM's multilingual ability by evaluating its machine translation performance across several datasets (WMT, Flores-101 and DiaBLa) and language pairs (hi...
http://arxiv.org/pdf/2303.01911v2
2023-05-09T12:21:58
2023-03-03T13:23:42
{ "title": [ -0.039215088, -0.011543274, -0.0368042, -0.05319214, -0.00617218, 0.00365448, -0.05267334, 0.0063591003, -0.0046691895, 0.064819336, -0.0031776428, 0.008125305, 0.052459717, -0.01574707, -0.05819702, 0.023712158, -0.008216858, 0.0203...
http://arxiv.org/abs/2303.01903v2
Prompting Large Language Models with Answer Heuristics for Knowledge-based Visual Question Answering
Zhenwei Shao, Zhou Yu, Meng Wang, Jun Yu
cs.CV, cs.CL, cs.LG
Knowledge-based visual question answering (VQA) requires external knowledge beyond the image to answer the question. Early studies retrieve required knowledge from explicit knowledge bases (KBs), which often introduces irrelevant information to the question, hence restricting the performance of their models. Recent wor...
http://arxiv.org/pdf/2303.01903v2
2023-03-16T01:49:29
2023-03-03T13:05:15
{ "title": [ -0.023162842, -0.009178162, -0.018432617, -0.052093506, -0.0234375, 0.008522034, -0.05114746, 0.0025463104, 0.03744507, 0.008743286, -0.04144287, 0.0028839111, -0.008460999, -0.0059509277, -0.05001831, -0.0048294067, 0.03186035, -0.0...
http://arxiv.org/abs/2303.03165v1
Multi label classification of Artificial Intelligence related patents using Modified D2SBERT and Sentence Attention mechanism
Yongmin Yoo, Tak-Sung Heo, Dongjin Lim, Deaho Seo
cs.CL, cs.AI
Patent classification is an essential task in patent information management and patent knowledge mining. It is very important to classify patents related to artificial intelligence, which is the biggest topic these days. However, artificial intelligence-related patents are very difficult to classify because it is a mix...
http://arxiv.org/pdf/2303.03165v1
2023-03-03T12:27:24
2023-03-03T12:27:24
{ "title": [ -0.0059127808, 0.0033836365, -0.020584106, -0.010803223, -0.010223389, -0.0234375, 0.021575928, -0.019729614, -0.008216858, 0.04135132, -0.030395508, -0.0011882782, 0.010574341, -0.027740479, -0.021087646, -0.02394104, -0.021133423, ...
http://arxiv.org/abs/2303.05401v1
Early Warning Signals of Social Instabilities in Twitter Data
Vahid Shamsaddini, Henry Kirveslahti, Raphael Reinauer, Wallyson Lemes de Oliveira, Matteo Caorsi, Etienne Voutaz
cs.CL, cs.LG, cs.SI, 68
The goal of this project is to create and study novel techniques to identify early warning signals for socially disruptive events, like riots, wars, or revolutions using only publicly available data on social media. Such techniques need to be robust enough to work on real-time data: to achieve this goal we propose a to...
http://arxiv.org/pdf/2303.05401v1
2023-03-03T11:18:02
2023-03-03T11:18:02
{ "title": [ 0.02798462, 0.017471313, -0.019821167, -0.050872803, 0.028366089, -0.020690918, 0.0048332214, -0.02180481, -0.013656616, 0.033935547, -0.039367676, 0.025375366, -0.007896423, -0.0045318604, -0.032440186, -0.007331848, 0.01626587, 0.0...
http://arxiv.org/abs/2303.01847v1
Mapping Wordnets on the Fly with Permanent Sense Keys
Eric Kafe
cs.CL
Most of the major databases on the semantic web have links to Princeton WordNet (PWN) synonym set (synset) identifiers, which differ for each PWN release, and are thus incompatible between versions. On the other hand, both PWN and the more recent Open English Wordnet (OEWN) provide permanent word sense identifiers (the...
http://arxiv.org/pdf/2303.01847v1
2023-03-03T11:01:10
2023-03-03T11:01:10
{ "title": [ -0.004245758, -0.019470215, 0.018844604, -0.017227173, -0.03930664, 0.016723633, -0.019302368, -0.03338623, 0.020019531, 0.056365967, -0.04095459, -0.01322937, 0.0435791, -0.01361084, -0.047668457, 0.0071868896, 0.011817932, 0.020050...
http://arxiv.org/abs/2303.01795v1
PAGE: A Position-Aware Graph-Based Model for Emotion Cause Entailment in Conversation
Xiaojie Gu, Renze Lou, Lin Sun, Shangxin Li
cs.CL
Conversational Causal Emotion Entailment (C2E2) is a task that aims at recognizing the causes corresponding to a target emotion in a conversation. The order of utterances in the conversation affects the causal inference. However, most current position encoding strategies ignore the order relation among utterances and s...
http://arxiv.org/pdf/2303.01795v1
2023-03-03T09:13:23
2023-03-03T09:13:23
{ "title": [ 0.0095825195, -0.0041770935, -0.034179688, -0.07940674, -0.00049591064, -0.0026950836, -0.05307007, 0.015556335, 0.030685425, 0.014503479, 0.0059280396, -0.011444092, 0.031280518, -0.03439331, -0.03692627, 0.012466431, -0.020477295, ...
http://arxiv.org/abs/2303.01794v2
Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News
Yuta Koreeda, Ken-ichi Yokote, Hiroaki Ozaki, Atsuki Yamaguchi, Masaya Tsunokake, Yasuhiro Sogawa
cs.CL, cs.AI
This paper explains the participation of team Hitachi to SemEval-2023 Task 3 "Detecting the genre, the framing, and the persuasion techniques in online news in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the task and the low-resource setting, we investigated different cross-lingual and mult...
http://arxiv.org/pdf/2303.01794v2
2023-04-25T05:44:49
2023-03-03T09:12:55
{ "title": [ -0.0035037994, -0.022888184, -0.060791016, -0.012031555, -0.008033752, -0.039611816, -0.00504303, -0.000039041042, -0.022918701, 0.042907715, 0.0035171509, 0.04248047, 0.025802612, -0.017578125, 0.0001912117, 0.012275696, -0.027496338, ...
http://arxiv.org/abs/2303.01793v1
Exploiting Language Relatedness in Machine Translation Through Domain Adaptation Techniques
Amit Kumar, Rupjyoti Baruah, Ajay Pratap, Mayank Swarnkar, Anil Kumar Singh
cs.CL, cs.AI
One of the significant challenges of Machine Translation (MT) is the scarcity of large amounts of data, mainly parallel sentence aligned corpora. If the evaluation is as rigorous as resource-rich languages, both Neural Machine Translation (NMT) and Statistical Machine Translation (SMT) can produce good results with suc...
http://arxiv.org/pdf/2303.01793v1
2023-03-03T09:07:30
2023-03-03T09:07:30
{ "title": [ -0.027908325, -0.0049819946, -0.02078247, -0.061676025, -0.0005259514, 0.003768921, -0.044677734, 0.0018281937, -0.013267517, 0.05203247, 0.019439697, 0.0032730103, 0.037231445, -0.01637268, -0.03994751, 0.017852783, -0.012496948, 0....
http://arxiv.org/abs/2303.01781v1
Meme Sentiment Analysis Enhanced with Multimodal Spatial Encoding and Facial Embedding
Muzhaffar Hazman, Susan McKeever, Josephine Griffith
cs.CV, cs.CL
Internet memes are characterised by the interspersing of text amongst visual elements. State-of-the-art multimodal meme classifiers do not account for the relative positions of these elements across the two modalities, despite the latent meaning associated with where text and visual elements are placed. Against two mem...
http://arxiv.org/pdf/2303.01781v1
2023-03-03T08:44:20
2023-03-03T08:44:20
{ "title": [ 0.02748108, 0.004234314, -0.029663086, -0.018829346, 0.0008764267, -0.033691406, 0.015197754, -0.023757935, 0.04058838, 0.045562744, -0.010223389, 0.0017127991, 0.01802063, -0.01474762, -0.057006836, 0.025970459, -0.012321472, 0.0387...
http://arxiv.org/abs/2303.01742v1
NCL: Textual Backdoor Defense Using Noise-augmented Contrastive Learning
Shengfang Zhai, Qingni Shen, Xiaoyi Chen, Weilong Wang, Cong Li, Yuejian Fang, Zhonghai Wu
cs.CR, cs.CL
At present, backdoor attacks attract attention as they do great harm to deep learning models. The adversary poisons the training data making the model being injected with a backdoor after being trained unconsciously by victims using the poisoned dataset. In the field of text, however, existing works do not provide suff...
http://arxiv.org/pdf/2303.01742v1
2023-03-03T07:07:04
2023-03-03T07:07:04
{ "title": [ -0.02178955, 0.0013046265, -0.0082092285, -0.040985107, -0.005519867, 0.0030822754, -0.02017212, -0.009750366, -0.0037727356, 0.021713257, -0.0026741028, 0.02003479, -0.009979248, -0.026748657, -0.035705566, -0.012786865, -0.020004272, ...
http://arxiv.org/abs/2304.11076v1
Can ChatGPT-like Generative Models Guarantee Factual Accuracy? On the Mistakes of New Generation Search Engines
Ruochen Zhao, Xingxuan Li, Yew Ken Chia, Bosheng Ding, Lidong Bing
cs.CL, cs.AI
Although large conversational AI models such as OpenAI's ChatGPT have demonstrated great potential, we question whether such models can guarantee factual accuracy. Recently, technology companies such as Microsoft and Google have announced new services which aim to combine search engines with conversational AI. However,...
http://arxiv.org/pdf/2304.11076v1
2023-03-03T04:27:44
2023-03-03T04:27:44
{ "title": [ 0.009674072, -0.014190674, -0.016906738, -0.016723633, -0.016448975, -0.024780273, 0.004501343, -0.02368164, -0.011291504, 0.029907227, -0.02331543, 0.020111084, 0.004573822, -0.018432617, -0.023971558, -0.006816864, -0.016433716, 0....
http://arxiv.org/abs/2303.01694v1
DWFormer: Dynamic Window transFormer for Speech Emotion Recognition
Shuaiqi Chen, Xiaofen Xing, Weibin Zhang, Weidong Chen, Xiangmin Xu
cs.SD, cs.CL, eess.AS
Speech emotion recognition is crucial to human-computer interaction. The temporal regions that represent different emotions scatter in different parts of the speech locally. Moreover, the temporal scales of important information may vary over a large range within and across speech segments. Although transformer-based m...
http://arxiv.org/pdf/2303.01694v1
2023-03-03T03:26:53
2023-03-03T03:26:53
{ "title": [ 0.0069732666, -0.00724411, -0.05709839, -0.07476807, 0.0119018555, -0.016555786, -0.022064209, 0.01159668, 0.008255005, 0.04574585, -0.02368164, -0.012916565, 0.04360962, -0.02407837, -0.049591064, 0.0047073364, 0.0062408447, 0.02093...
http://arxiv.org/abs/2303.03922v1
Structure Pretraining and Prompt Tuning for Knowledge Graph Transfer
Wen Zhang, Yushan Zhu, Mingyang Chen, Yuxia Geng, Yufeng Huang, Yajing Xu, Wenting Song, Huajun Chen
cs.LG, cs.AI, cs.CL
Knowledge graphs (KG) are essential background knowledge providers in many tasks. When designing models for KG-related tasks, one of the key tasks is to devise the Knowledge Representation and Fusion (KRF) module that learns the representation of elements from KGs and fuses them with task representations. While due to ...
http://arxiv.org/pdf/2303.03922v1
2023-03-03T02:58:17
2023-03-03T02:58:17
{ "title": [ -0.021392822, -0.012557983, -0.052703857, -0.058746338, -0.01928711, -0.008300781, -0.052093506, -0.00041770935, -0.034118652, 0.0021381378, -0.000107228756, 0.02545166, -0.038726807, -0.03366089, 0.005958557, -0.008552551, 0.011688232, ...
http://arxiv.org/abs/2303.03329v1
End-to-End Speech Recognition: A Survey
Rohit Prabhavalkar, Takaaki Hori, Tara N. Sainath, Ralf Schlüter, Shinji Watanabe
eess.AS, cs.CL, cs.SD
In the last decade of automatic speech recognition (ASR) research, the introduction of deep learning brought considerable reductions in word error rate of more than 50% relative, compared to modeling without deep learning. In the wake of this transition, a number of all-neural ASR architectures were introduced. These s...
http://arxiv.org/pdf/2303.03329v1
2023-03-03T01:46:41
2023-03-03T01:46:41
{ "title": [ -0.02746582, 0.0073127747, -0.03717041, -0.085754395, 0.03817749, 0.016586304, -0.039978027, 0.022827148, -0.0062828064, 0.040740967, -0.008728027, -0.0008177757, 0.00042057037, -0.020828247, -0.045135498, 0.013916016, -0.013671875, ...
http://arxiv.org/abs/2303.01645v1
APIContext2Com: Code Comment Generation by Incorporating Pre-Defined API Documentation
Ramin Shahbazi, Fatemeh Fard
cs.SE, cs.CL, cs.LG
Code comments are significantly helpful in comprehending software programs and also aid developers to save a great deal of time in software maintenance. Code comment generation aims to automatically predict comments in natural language given a code snippet. Several works investigate the effect of integrating external k...
http://arxiv.org/pdf/2303.01645v1
2023-03-03T00:38:01
2023-03-03T00:38:01
{ "title": [ -0.025909424, -0.046966553, -0.06036377, -0.09069824, -0.022735596, -0.008422852, -0.07019043, 0.030258179, -0.012023926, 0.0440979, -0.041656494, 0.014694214, 0.0104522705, -0.009307861, -0.02406311, -0.021194458, 0.008522034, 0.036...
http://arxiv.org/abs/2303.01593v2
QAID: Question Answering Inspired Few-shot Intent Detection
Asaf Yehudai, Matan Vetzler, Yosi Mass, Koren Lazar, Doron Cohen, Boaz Carmeli
cs.CL, cs.AI, cs.IR, cs.LG
Intent detection with semantically similar fine-grained intents is a challenging task. To address it, we reformulate intent detection as a question-answering retrieval task by treating utterances and intent names as questions and answers. To that end, we utilize a question-answering retrieval architecture and adopt a t...
http://arxiv.org/pdf/2303.01593v2
2023-03-21T14:22:00
2023-03-02T21:35:15
{ "title": [ 0.014595032, 0.0005092621, -0.025604248, -0.018188477, 0.012573242, -0.030975342, 0.016799927, -0.012702942, 0.0259552, 0.009864807, -0.026611328, 0.00021862984, -0.013748169, -0.018753052, -0.04458618, -0.007785797, 0.00007247925, 0...
http://arxiv.org/abs/2303.01590v4
Technical report: Graph Neural Networks go Grammatical
Jason Piquenot, Aldo Moscatelli, Maxime Bérar, Pierre Héroux, Romain raveaux, Jean-Yves Ramel, Sébastien Adam
cs.LG, cs.CL
This paper introduces a framework for formally establishing a connection between a portion of an algebraic language and a Graph Neural Network (GNN). The framework leverages Context-Free Grammars (CFG) to organize algebraic operations into generative rules that can be translated into a GNN layer model. As CFGs derived ...
http://arxiv.org/pdf/2303.01590v4
2023-10-04T08:47:30
2023-03-02T21:27:54
{ "title": [ -0.0146484375, -0.01361084, -0.03289795, -0.046203613, -0.012382507, -0.012413025, -0.0357666, -0.0012607574, -0.018341064, 0.02583313, -0.0015535355, -0.0010566711, -0.008560181, -0.038513184, -0.015571594, 0.005531311, -0.0021762848, ...
http://arxiv.org/abs/2303.01580v2
Mixture of Soft Prompts for Controllable Data Generation
Derek Chen, Celine Lee, Yunan Lu, Domenic Rosati, Zhou Yu
cs.CL
Large language models (LLMs) effectively generate fluent text when the target output follows natural language patterns. However, structured prediction tasks confine the output format to a limited ontology, causing even very large models to struggle since they were never trained with such restrictions in mind. The diffi...
http://arxiv.org/pdf/2303.01580v2
2023-10-18T03:31:02
2023-03-02T21:13:56
{ "title": [ -0.025436401, -0.019165039, -0.06665039, -0.06726074, -0.024795532, -0.0034008026, -0.05166626, 0.023239136, -0.012306213, 0.018081665, -0.046936035, 0.003921509, -0.001616478, -0.034362793, 0.013183594, -0.012321472, 0.024658203, 0....
http://arxiv.org/abs/2303.01502v1
Computational Language Acquisition with Theory of Mind
Andy Liu, Hao Zhu, Emmy Liu, Yonatan Bisk, Graham Neubig
cs.CL, cs.AI
Unlike current state-of-the-art language models, young children actively acquire language through interactions with their surrounding environment and caretakers. One mechanism that has been argued to be critical to language learning is the ability to infer the mental states of other agents in social environments, coine...
http://arxiv.org/pdf/2303.01502v1
2023-03-02T18:59:46
2023-03-02T18:59:46
{ "title": [ -0.028518677, -0.030639648, 0.0036849976, -0.028259277, -0.02079773, 0.00957489, -0.002506256, 0.005268097, -0.021881104, 0.014640808, -0.013153076, 0.010917664, -0.0033054352, -0.029937744, -0.023406982, 0.0008420944, -0.024536133, ...
http://arxiv.org/abs/2303.01490v1
Language Variety Identification with True Labels
Marcos Zampieri, Kai North, Tommi Jauhiainen, Mariano Felice, Neha Kumari, Nishant Nair, Yash Bangera
cs.CL
Language identification is an important first step in many IR and NLP applications. Most publicly available language identification datasets, however, are compiled under the assumption that the gold label of each instance is determined by where texts are retrieved from. Research has shown that this is a problematic ass...
http://arxiv.org/pdf/2303.01490v1
2023-03-02T18:51:58
2023-03-02T18:51:58
{ "title": [ -0.006477356, 0.009399414, -0.03994751, -0.03152466, -0.0055274963, -0.023468018, -0.021850586, -0.007820129, -0.022232056, 0.033477783, 0.01739502, -0.012832642, -0.012123108, -0.053894043, -0.03378296, 0.004459381, -0.010192871, 0....
http://arxiv.org/abs/2303.01432v2
WiCE: Real-World Entailment for Claims in Wikipedia
Ryo Kamoi, Tanya Goyal, Juan Diego Rodriguez, Greg Durrett
cs.CL
Textual entailment models are increasingly applied in settings like fact-checking, presupposition verification in question answering, or summary evaluation. However, these represent a significant domain shift from existing entailment datasets, and models underperform as a result. We propose WiCE, a new fine-grained tex...
http://arxiv.org/pdf/2303.01432v2
2023-10-22T18:11:08
2023-03-02T17:45:32
{ "title": [ 0.003528595, -0.012451172, -0.05178833, -0.054107666, 0.009178162, -0.026412964, 0.008514404, 0.003583908, -0.025375366, 0.02331543, -0.013061523, 0.017669678, 0.04598999, -0.03616333, 0.017822266, 0.017837524, -0.030700684, 0.068969...
http://arxiv.org/abs/2303.03387v3
CoSyn: Detecting Implicit Hate Speech in Online Conversations Using a Context Synergized Hyperbolic Network
Sreyan Ghosh, Manan Suri, Purva Chiniya, Utkarsh Tyagi, Sonal Kumar, Dinesh Manocha
cs.LG, cs.AI, cs.CL, cs.SI
The tremendous growth of social media users interacting in online conversations has led to significant growth in hate speech, affecting people from various demographics. Most of the prior works focus on detecting explicit hate speech, which is overt and leverages hateful phrases, with very little work focusing on detec...
http://arxiv.org/pdf/2303.03387v3
2023-10-24T12:57:08
2023-03-02T17:30:43
{ "title": [ 0.033355713, 0.022521973, -0.043273926, -0.056152344, 0.035705566, -0.030883789, 0.002714157, -0.006175995, 0.0009994507, 0.03668213, -0.028549194, 0.00083494186, 0.029846191, -0.014007568, -0.036254883, -0.012306213, -0.019363403, 0...
http://arxiv.org/abs/2303.01421v1
Semiparametric Language Models Are Scalable Continual Learners
Guangyue Peng, Tao Ge, Si-Qing Chen, Furu Wei, Houfeng Wang
cs.CL, cs.LG
Semiparametric language models (LMs) have shown promise in continuously learning from new text data by combining a parameterized neural LM with a growable non-parametric memory for memorizing new content. However, conventional semiparametric LMs will finally become prohibitive for computing and storing if they are appl...
http://arxiv.org/pdf/2303.01421v1
2023-03-02T17:15:02
2023-03-02T17:15:02
{ "title": [ -0.0079956055, 0.005657196, -0.04031372, -0.06903076, -0.013442993, -0.019119263, -0.055755615, 0.009399414, -0.01725769, 0.016220093, -0.035888672, -0.005657196, -0.027496338, -0.025054932, -0.04309082, -0.014625549, -0.0070381165, ...