AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

CSMeD: Bridging the Dataset Gap in Automated Citation Screening for Systematic Literature Reviews

Wojciech Kusa, Oscar E. Mendoza, Matthias Samwald, Petr Knoth, Allan Hanbury

NeurIPS 2023 · Large language models › Retrieval and knowledge · approach: Domain-specific QA datasets · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use the SLR abstract rather than title or eligibility criteria as the query representation (ablation)BM25 and MiniLM-L6-v2 · CSMeD-dev-cochraneNDCG@10: BM25: 0.438 / 0.286 -> 0.503; MiniLM: 0.470 / 0.299 -> 0.517Improved
Replace BM25 with MiniLM-L6-v2 using abstractsCSMeD zero-shot retrieval · CSMeD-dev-cochraneNDCG@10: 0.503 -> 0.517Improved
Replace BM25 with MiniLM-L6-v2 using abstractsCSMeD zero-shot retrieval · CSMeD-dev-cochraneLast Rel: 63.6 -> 65.5Hurt
Use eligibility criteria rather than abstract as the query representation (ablation)BM25 and MiniLM-L6-v2 · CSMeD-dev-cochraneNDCG@10: BM25: 0.503 -> 0.286; MiniLM: 0.517 -> 0.299Hurt
Use BigBird rather than Longformer architecture (ablation)Open-domain full-text screening models · CSMeD-ft-testF1-score: Longformer: 0.398; BigBird-roberta-base: 0.575Improved
Use BigBird rather than Longformer architecture (ablation)Clinical full-text screening models · CSMeD-ft-testF1-score: Clinical-Longformer: 0.435; Clinical-BigBird: 0.609Improved
Use clinical-domain pretraining rather than open-domain pretraining (ablation)Longformer and BigBird full-text screening models · CSMeD-ft-testF1-score: Longformer: 0.398 -> Clinical-Longformer: 0.435; BigBird-roberta-base: 0.575 -> Clinical-BigBird: 0.609Mixed
Use GPT-4-8k rather than the other evaluated modelsFull-text screening · CSMeD-ft-test-smallF1-score: GPT-4-8k-0314: 0.660; the difference is not statistically significantNo clear effect
Use GPT-3.5-turbo-16k rather than GPT-3.5-turbo-0301Zero-shot full-text screening · CSMeD-ft-test-smallPrecision: 0.585 -> 0.712Improved
Use Clinical-BigBird rather than GPT-3.5-turbo-16kFull-text screening · CSMeD-ft-testF1-score: 0.475 -> 0.609Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy