paper-with-me

Papers

Long Context Question Answering via Supervised Contrastive Learning

2021-12-16 · NAACL 2022 7 · Avi Caciularu, Ido Dagan, Jacob Goldberger, Arman Cohan

Long-context question answering (QA) tasks require reasoning over a long document or multiple documents. Addressing these tasks often benefits from identifying a set of evidence spans (e.g., sentences), which provide supporting evidence for answering the question. In this work, we propose a novel method for equipping long-context QA models with an additional sequence-level objective for better identification of the supporting evidence. We achieve this via an additional contrastive supervision signal in finetuning, where the model is encouraged to explicitly discriminate supporting evidence sentences from negative ones by maximizing question-evidence similarity. The proposed additional loss exhibits consistent improvements on three different strong long-context transformer models, across two challenging question answering benchmarks -- HotpotQA and QAsper.

📄 PDF Abstract BibTeX arXiv:2112.08777

Code (2)

aviclu/long-context-qa-contrast 공식 구현 pytorch
aviclu/seq-contrast-qa 공식 구현 pytorch

Tasks

Contrastive LearningQuestion AnsweringVideo Generation

Similar Papers 제목 키워드 기반

elBERto: Self-supervised Commonsense Learning for Question Answering

2022-03-17 · Xunlin Zhan, Yuan Li, Xiao Dong, Xiaodan Liang 외

Commonsense question answering requires reasoning about everyday situations and causes and effects implicit in context. Typically, existing approaches first retrieve external evidence and then perform commonsense reasoni…

Question AnsweringRepresentation LearningRetrieval

MuDAF: Long-Context Multi-Document Attention Focusing through Contrastive Learning on Attention Heads

2025-02-19 · Weihao Liu, Ning Wu, Shiping Yang, Wenbiao Ding 외

Large Language Models (LLMs) frequently show distracted attention due to irrelevant information in the input, which severely impairs their long-context capabilities. Inspired by recent studies on the effectiveness of ret…

Contrastive LearningQuestion AnsweringRetrieval

QA Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation

2022-10-19 · Zhenrui Yue, Huimin Zeng, Bernhard Kratzwald, Stefan Feuerriegel 외

Question answering (QA) has recently shown impressive results for answering questions from customized domains. Yet, a common challenge is to adapt QA models to an unseen target domain. In this paper, we propose a novel s…

Contrastive LearningData AugmentationDomain AdaptationQuestion Answering

QAID: Question Answering Inspired Few-shot Intent Detection

2023-03-02 · Asaf Yehudai, Matan Vetzler, Yosi Mass, Koren Lazar 외

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 ques…

Intent DetectionQuestion AnsweringRetrieval

Momentum Contrastive Pre-training for Question Answering

2022-12-12 · Minda Hu, Muzhi Li, Yasheng Wang, Irwin King

Existing pre-training methods for extractive Question Answering (QA) generate cloze-like queries different from natural questions in syntax structure, which could overfit pre-trained models to simple keyword matching. In…

BenchmarkingContrastive LearningExtractive Question-AnsweringNatural Questions+1