Sammenlign metoder

	Svagt Overvåget Spørgsmål-Svar	Domæne-adaptiv Spørgsmålsbesvarelse
Fagområde	Dyb læring	Dyb læring
Familie	Machine learning	Machine learning
Oprindelsesår	2017–2019	2019–2020
Ophavsperson	Multiple authors (Clark, Gardner, Min et al.)	Multiple (e.g., Garg et al.; Yue et al.)
Type	Weakly supervised NLP model	Domain adaptation for extractive/generative QA
Oprindelig kilde	Clark, C., & Gardner, M. (2018). Simple and Effective Multi-Paragraph Reading Comprehension. In Proceedings of ACL 2018, pp. 845–855. Association for Computational Linguistics.	Garg, S., Vu, T., & Moschitti, A. (2020). TANDA: Transfer and Adapt Pre-Trained Transformer Models for Answer Sentence Selection. Proceedings of the AAAI Conference on Artificial Intelligence, 34(5), 7780–7788.
Aliasser	WS-QA, distantly supervised QA, noisy-label question answering, indirect supervision QA	DA-QA, domain-adapted QA, domain-specific question answering, cross-domain question answering
Relaterede	4	6
Resumé	Weakly supervised question answering (WS-QA) trains neural reading-comprehension models using indirect or automatically derived answer labels rather than expensive human-annotated span annotations. By exploiting distant supervision, heuristic labeling, or answer-presence signals, WS-QA makes QA feasible in domains and languages where full annotation is impractical.	Domain-adaptive Question Answering (DA-QA) adapts a pre-trained language model — typically BERT or RoBERTa — first trained on general QA benchmarks such as SQuAD to answer questions accurately in a new target domain (e.g., biomedical, legal, financial) where labelled data is scarce. Combining domain-adaptive pre-training with task fine-tuning yields substantially stronger performance than direct fine-tuning alone.