AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

BiCA: Effective Biomedical Dense Retrieval with Citation-Aware Hard Negatives

Aarush Sinha, Pavan Kumar S, Roshan Balaji, Nirav Pravinbhai Bhatt

AAAI 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Cross-domain text and graph transfer · 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
fine-tune GTE-small on citation-aware hard negativesGTE-small · NQnDCG@10 (x100): 32.0 -> 50.2Improved
fine-tune GTE-base on citation-aware hard negativesGTE-base · ArguAnanDCG@10 (x100): 41.0 -> 57.1Improved
fine-tune GTE-small on citation-aware hard negativesGTE-small · SciFactnDCG@10 (x100): 72.7 -> 72.7No clear effect
fine-tune GTE-small on citation-aware hard negativesGTE-small · NFCorpusnDCG@10 (x100): 34.9 -> 34.7No clear effect
fine-tune on citation-aware negatives (average over 14 BEIR tasks)GTE-base · BEIR 14 tasksnDCG@10 gain: +6.8Improved
use 3 traversal paths instead of 2 (ablation)BiCA mining, bert-base · 5 BEIR tasks avgnDCG@10: 0.2068 -> 0.2739Improved
use 3 traversal paths instead of 4 (ablation)BiCA mining, bert-base · 5 BEIR tasks avgnDCG@10: 0.2184 -> 0.2739Improved
use 3 traversal paths instead of 5 (ablation)BiCA mining, bert-base · 5 BEIR tasks avgnDCG@10: 0.2734 -> 0.2739No clear effect
use path length 3 instead of 2 (ablation)BiCA mining, bert-base · 5 BEIR tasks avgnDCG@10: 0.2041 -> 0.2739Improved
use path length 3 instead of 5 (ablation)BiCA mining, bert-base · 5 BEIR tasks avgnDCG@10: 0.2580 -> 0.2739Improved
train on 20k triplets instead of 1k (ablation)bert-base-uncased · ArguAnanDCG@10: 0.384 -> 0.444Improved
train on 20k triplets instead of 1k (ablation)bert-base-uncased · SciFactnDCG@10: 0.262 -> 0.493Improved

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