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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace top-k re-ranking with rationale-driven adaptive selection | METEORA · Context Prioritization, 6 datasets | Recall (average): 0.82 -> 0.93 | Improved |
| add DPO tuning of the rationale generator (ablation) | METEORA · MAUD | Recall: 0.65 -> 0.72 | Improved |
| add DPO tuning, average over datasets (ablation) | METEORA · Context Prioritization, 6 datasets | Recall (average): 0.88 -> 0.93 | Improved |
| add context expansion (ablation) | METEORA · Context Prioritization, 6 datasets | Recall / Precision (average): Recall 0.89 -> 0.93; Precision 0.23 -> 0.19 | Mixed |
| add context expansion on MAUD (ablation) | METEORA · MAUD | Recall: 0.66 -> 0.72 | Improved |
| add context expansion on ContractNLI (ablation) | METEORA · ContractNLI | Recall: 0.97 -> 1.00 | No clear effect |
| add Verifier LLM (ablation) | METEORA · Context Prioritization, 6 datasets | Precision (average): 0.17 -> 0.19 | Improved |
| add Verifier LLM (ablation) | METEORA · Context Prioritization, 6 datasets | Recall (average): 0.94 -> 0.93 | Hurt |
| add Verifier LLM for corpus poisoning | METEORA · Corpus poisoning defense, QASPER | F1: 0.00 -> 0.44 | Improved |
| add verifier, latency cost (ablation) | METEORA · Latency | Total time (s): 1.77 -> 2.91 | Hurt |
| adaptive selection with fewer evidence chunks vs SBERT and RankRAG | METEORA · Latency | Total time (s): 4.04 (SBERT) -> 2.91 | Improved |
| rationale-driven selection on FinQA (short documents) | METEORA · FinQA | Recall: 0.98 -> 0.95 | Hurt |
Closest papers by meaning
- Worse than Zero-shot? A Fact-Checking Dataset for Evaluating the Robustness of RAG Against Misleading Retrievals (NeurIPS 2025 · similarity 0.91)
- On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains (ICML 2025 · similarity 0.89)
- Retrieval is Not Enough: Enhancing RAG through Test-Time Critique and Optimization (NeurIPS 2025 · similarity 0.88)
- RAGGED: Towards Informed Design of Scalable and Stable RAG Systems (ICML 2025 · similarity 0.87)
- Follow My Instruction and Spill the Beans: Scalable Data Extraction from Retrieval-Augmented Generation Systems (ICLR 2025 · similarity 0.87)
- ReliabilityRAG: Effective and Provably Robust Defense for RAG-based Web-Search (NeurIPS 2025 · similarity 0.87)
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.
