paper-with-me

Papers

When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset

2021-04-18 · Lucia Zheng, Neel Guha, Brandon R. Anderson, Peter Henderson, Daniel E. Ho

While self-supervised learning has made rapid advances in natural language processing, it remains unclear when researchers should engage in resource-intensive domain-specific pretraining (domain pretraining). The law, puzzlingly, has yielded few documented instances of substantial gains to domain pretraining in spite of the fact that legal language is widely seen to be unique. We hypothesize that these existing results stem from the fact that existing legal NLP tasks are too easy and fail to meet conditions for when domain pretraining can help. To address this, we first present CaseHOLD (Case Holdings On Legal Decisions), a new dataset comprised of over 53,000+ multiple choice questions to identify the relevant holding of a cited case. This dataset presents a fundamental task to lawyers and is both legally meaningful and difficult from an NLP perspective (F1 of 0.4 with a BiLSTM baseline). Second, we assess performance gains on CaseHOLD and existing legal NLP datasets. While a Transformer architecture (BERT) pretrained on a general corpus (Google Books and Wikipedia) improves performance, domain pretraining (using corpus of approximately 3.5M decisions across all courts in the U.S. that is larger than BERT's) with a custom legal vocabulary exhibits the most substantial performance gains with CaseHOLD (gain of 7.2% on F1, representing a 12% improvement on BERT) and consistent performance gains across two other legal tasks. Third, we show that domain pretraining may be warranted when the task exhibits sufficient similarity to the pretraining corpus: the level of performance increase in three legal tasks was directly tied to the domain specificity of the task. Our findings inform when researchers should engage resource-intensive pretraining and show that Transformer-based architectures, too, learn embeddings suggestive of distinct legal language.

📄 PDF Abstract BibTeX arXiv:2104.08671

Code (2)

reglab/casehold 공식 구현 pytorch
trusthlt/privacy-legal-nlp-lm jax

Tasks

Multiple-choiceQuestion AnsweringSelf-Supervised LearningSpecificityText Classification

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Tanh Activation 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Sigmoid Activation 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

Does GNN Pretraining Help Molecular Representation?

2022-07-13 · Ruoxi Sun, Hanjun Dai, Adams Wei Yu

Extracting informative representations of molecules using Graph neural networks (GNNs) is crucial in AI-driven drug discovery. Recently, the graph research community has been trying to replicate the success of self-super…

Drug Discoverymolecular representation

Resource and data efficient self supervised learning

2021-09-03 · Ozan Ciga, Tony Xu, Anne L. Martel

We investigate the utility of pretraining by contrastive self supervised learning on both natural-scene and medical imaging datasets when the unlabeled dataset size is small, or when the diversity within the unlabeled se…

DiversitySelf-Supervised Learning

When Does Contrastive Visual Representation Learning Work?

2021-05-12 · CVPR 2022 1 · Elijah Cole, Xuan Yang, Kimberly Wilber, Oisin Mac Aodha 외

Recent self-supervised representation learning techniques have largely closed the gap between supervised and unsupervised learning on ImageNet classification. While the particulars of pretraining on ImageNet are now rela…

Contrastive LearningFine-Grained Image ClassificationRepresentation LearningSelf-Supervised Learning

Evaluating Protein Transfer Learning with TAPE

2019-06-19 · NeurIPS 2019 12 · Roshan Rao, Nicholas Bhattacharya, Neil Thomas, Yan Duan 외

Protein modeling is an increasingly popular area of machine learning research. Semi-supervised learning has emerged as an important paradigm in protein modeling due to the high cost of acquiring supervised protein labels…

BIG-bench Machine LearningRepresentation LearningTransfer Learning

United We Pretrain, Divided We Fail! Representation Learning for Time Series by Pretraining on 75 Datasets at Once

2024-02-23 · Maurice Kraus, Felix Divo, David Steinmann, Devendra Singh Dhami 외

In natural language processing and vision, pretraining is utilized to learn effective representations. Unfortunately, the success of pretraining does not easily carry over to time series due to potential mismatch between…

Representation LearningTime Series