AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Ranking Free RAG: Replacing Re-ranking with Selection in RAG for Sensitive Domains

Yash Saxena, Ankur Padia, Mandar Chaudhary, Kalpa Gunaratna, srinivasan parthasarathy, Manas Gaur

ICML 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 0 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
replace top-k re-ranking with rationale-driven adaptive selectionMETEORA · Context Prioritization, 6 datasetsRecall (average): 0.82 -> 0.93Improved
add DPO tuning of the rationale generator (ablation)METEORA · MAUDRecall: 0.65 -> 0.72Improved
add DPO tuning, average over datasets (ablation)METEORA · Context Prioritization, 6 datasetsRecall (average): 0.88 -> 0.93Improved
add context expansion (ablation)METEORA · Context Prioritization, 6 datasetsRecall / Precision (average): Recall 0.89 -> 0.93; Precision 0.23 -> 0.19Mixed
add context expansion on MAUD (ablation)METEORA · MAUDRecall: 0.66 -> 0.72Improved
add context expansion on ContractNLI (ablation)METEORA · ContractNLIRecall: 0.97 -> 1.00No clear effect
add Verifier LLM (ablation)METEORA · Context Prioritization, 6 datasetsPrecision (average): 0.17 -> 0.19Improved
add Verifier LLM (ablation)METEORA · Context Prioritization, 6 datasetsRecall (average): 0.94 -> 0.93Hurt
add Verifier LLM for corpus poisoningMETEORA · Corpus poisoning defense, QASPERF1: 0.00 -> 0.44Improved
add verifier, latency cost (ablation)METEORA · LatencyTotal time (s): 1.77 -> 2.91Hurt
adaptive selection with fewer evidence chunks vs SBERT and RankRAGMETEORA · LatencyTotal time (s): 4.04 (SBERT) -> 2.91Improved
rationale-driven selection on FinQA (short documents)METEORA · FinQARecall: 0.98 -> 0.95Hurt

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