id stringlengths 33 34 | doc_url stringlengths 33 34 | title stringlengths 10 211 | publication_date timestamp[s] | update_date timestamp[s] | authors stringlengths 5 20.1k | category_primary stringclasses 78
values | category_all stringlengths 5 138 | abstract stringlengths 123 3.09k | embeddings listlengths 768 768 |
|---|---|---|---|---|---|---|---|---|---|
http://arxiv.org/abs/2212.03813v4 | http://arxiv.org/pdf/2212.03813v4 | Robustness of Learning from Task Instructions | 2022-12-07T17:54:59 | 2023-05-23T13:39:39 | Jiasheng Gu, Hongyu Zhao, Hanzi Xu, Liangyu Nie, Hongyuan Mei, Wenpeng Yin | cs.CL | cs.CL | Traditional supervised learning mostly works on individual tasks and requires
training on a large set of task-specific examples. This paradigm seriously
hinders the development of task generalization since preparing a task-specific
example set is costly. To build a system that can quickly and easily generalize
to new t... | [
0.0653905347,
-1.0149949789,
0.4767991006,
0.3362721503,
-0.3244452775,
-0.2532216012,
-0.472494334,
-0.1662518084,
-0.2943106592,
0.4506784678,
-0.5923339128,
-0.5929360986,
-0.8654900789,
0.1911830902,
-0.3633796275,
1.2917960882,
-0.2176746577,
-0.1706789136,
-0.3153348863,
... |
http://arxiv.org/abs/2212.03869v1 | http://arxiv.org/pdf/2212.03869v1 | Pre-Training With Scientific Text Improves Educational Question Generation | 2022-12-07T17:17:58 | 2022-12-07T17:17:58 | Hamze Muse, Sahan Bulathwela, Emine Yilmaz | cs.CL | cs.CL, cs.AI, cs.CY, cs.IR, cs.LG, stat.ML, H.3.3; J.1; I.2.0 | With the boom of digital educational materials and scalable e-learning
systems, the potential for realising AI-assisted personalised learning has
skyrocketed. In this landscape, the automatic generation of educational
questions will play a key role, enabling scalable self-assessment when a global
population is manoeuvr... | [
-0.5750786066,
-1.1101056337,
0.3676655889,
-0.2552350461,
-0.0910321027,
0.0610474572,
0.0957903415,
-0.3486667871,
-0.0352382064,
0.5176400542,
-0.597871244,
-0.1877368391,
-0.2861850262,
0.4249358773,
-0.2804042399,
0.9652938843,
0.226755023,
0.14841187,
-0.8019682765,
0.206... |
http://arxiv.org/abs/2212.03760v5 | http://arxiv.org/pdf/2212.03760v5 | Pivotal Role of Language Modeling in Recommender Systems: Enriching Task-specific and Task-agnostic Representation Learning | 2022-12-07T16:31:14 | 2023-05-13T08:16:21 | Kyuyong Shin, Hanock Kwak, Wonjae Kim, Jisu Jeong, Seungjae Jung, Kyung-Min Kim, Jung-Woo Ha, Sang-Woo Lee | cs.IR | cs.IR, cs.CL | Recent studies have proposed unified user modeling frameworks that leverage
user behavior data from various applications. Many of them benefit from
utilizing users' behavior sequences as plain texts, representing rich
information in any domain or system without losing generality. Hence, a
question arises: Can language ... | [
-0.0499419235,
-0.4765660465,
0.6684715748,
-0.2198244184,
0.1764863133,
-0.0388594121,
-0.1187742874,
-0.484431833,
-0.0088425241,
0.7748830318,
-0.4039211273,
-0.2286399603,
-0.3915661871,
-0.3742326796,
-0.7151010633,
1.0765900612,
0.0214484297,
0.0325794071,
-0.318467021,
-... |
http://arxiv.org/abs/2212.03749v2 | http://arxiv.org/pdf/2212.03749v2 | Memorization of Named Entities in Fine-tuned BERT Models | 2022-12-07T16:20:50 | 2023-10-10T14:32:43 | Andor Diera, Nicolas Lell, Aygul Garifullina, Ansgar Scherp | cs.CL | cs.CL | Privacy preserving deep learning is an emerging field in machine learning
that aims to mitigate the privacy risks in the use of deep neural networks. One
such risk is training data extraction from language models that have been
trained on datasets, which contain personal and privacy sensitive information.
In our study,... | [
-0.4355195761,
-0.8772716522,
0.2738810182,
0.0138776144,
-0.2231796086,
0.066740185,
-0.5071439147,
-0.5971159935,
-0.1447179615,
0.6789722443,
-0.6113520861,
-0.5887595415,
-0.6730189919,
-0.0602249578,
-0.6209687591,
1.3942658901,
0.1958648562,
-0.1384606659,
0.1183899194,
0... |
http://arxiv.org/abs/2212.03747v1 | http://arxiv.org/pdf/2212.03747v1 | Harnessing Knowledge and Reasoning for Human-Like Natural Language Generation: A Brief Review | 2022-12-07T16:18:19 | 2022-12-07T16:18:19 | Jiangjie Chen, Yanghua Xiao | cs.CL | cs.CL | The rapid development and application of natural language generation (NLG)
techniques has revolutionized the field of automatic text production. However,
these techniques are still limited in their ability to produce human-like text
that is truly reasonable and informative. In this paper, we explore the
importance of N... | [
-0.5109455585,
-0.7584338784,
0.661379993,
0.0006341389,
-0.1137793511,
0.0802954435,
-0.2401145697,
-0.591445148,
0.0187001005,
0.8596716523,
-0.7227902412,
-0.430552721,
-0.3289038241,
0.486759454,
-0.4638659954,
0.9062114358,
0.3860182166,
-0.3110252321,
-0.1501520127,
0.262... |
http://arxiv.org/abs/2212.03657v1 | http://arxiv.org/pdf/2212.03657v1 | M3ST: Mix at Three Levels for Speech Translation | 2022-12-07T14:22:00 | 2022-12-07T14:22:00 | Xuxin Cheng, Qianqian Dong, Fengpeng Yue, Tom Ko, Mingxuan Wang, Yuexian Zou | cs.CL | cs.CL, cs.SD, eess.AS | How to solve the data scarcity problem for end-to-end speech-to-text
translation (ST)? It's well known that data augmentation is an efficient method
to improve performance for many tasks by enlarging the dataset. In this paper,
we propose Mix at three levels for Speech Translation (M^3ST) method to
increase the diversi... | [
-0.7654612064,
-0.7210476398,
0.0655653924,
0.2226041555,
-0.691612184,
0.1769881397,
-0.2209240496,
-0.5798761249,
0.0327027738,
0.550186336,
-0.6288769245,
-0.4861771166,
-0.7325940132,
0.3086903393,
-0.3697560132,
0.9411324859,
-0.1298658103,
0.2466372699,
0.2955064774,
-0.2... |
http://arxiv.org/abs/2212.03613v3 | http://arxiv.org/pdf/2212.03613v3 | G-MAP: General Memory-Augmented Pre-trained Language Model for Domain Tasks | 2022-12-07T13:07:24 | 2024-02-17T19:31:47 | Zhongwei Wan, Yichun Yin, Wei Zhang, Jiaxin Shi, Lifeng Shang, Guangyong Chen, Xin Jiang, Qun Liu | cs.CL | cs.CL | Recently, domain-specific PLMs have been proposed to boost the task
performance of specific domains (e.g., biomedical and computer science) by
continuing to pre-train general PLMs with domain-specific corpora. However,
this Domain-Adaptive Pre-Training (DAPT; Gururangan et al. (2020)) tends to
forget the previous gener... | [
-0.4916910827,
-0.9110229015,
0.5589989424,
-0.0670298114,
-0.3274866343,
0.168591246,
-0.2810854018,
-0.3798279762,
0.3721026778,
0.5506874919,
-0.3144362271,
-0.5722145438,
-0.7289314866,
-0.3753888905,
-0.175597474,
1.0512530804,
-0.207994014,
-0.0748092756,
-0.6269605756,
-... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.