FRESCO: Benchmarking and Optimizing Re-rankers for Evolving Semantic Conflict in Retrieval-Augmented Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: An, Sohyun, Lee, Hayeon, Yuan, Shuibenyang, Chen, Chun-cheng Jason, Hsieh, Cho-Jui, Mohan, Vijai, Min, Alexander
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915939321118720
author An, Sohyun
Lee, Hayeon
Yuan, Shuibenyang
Chen, Chun-cheng Jason
Hsieh, Cho-Jui
Mohan, Vijai
Min, Alexander
author_facet An, Sohyun
Lee, Hayeon
Yuan, Shuibenyang
Chen, Chun-cheng Jason
Hsieh, Cho-Jui
Mohan, Vijai
Min, Alexander
contents Retrieval-Augmented Generation (RAG) is a key approach to mitigating the temporal staleness of large language models (LLMs) by grounding responses in up-to-date evidence. Within the RAG pipeline, re-rankers play a pivotal role in selecting the most useful documents from retrieved candidates. However, existing benchmarks predominantly evaluate re-rankers in static settings and do not adequately assess performance under evolving information -- a critical gap, as real-world systems often must choose among temporally different pieces of evidence. To address this limitation, we introduce FRESCO (Factual Recency and Evolving Semantic COnflict), a benchmark for evaluating re-rankers in temporally dynamic contexts. By pairing recency-seeking queries with historical Wikipedia revisions, FRESCO tests whether re-rankers can prioritize factually recent evidence while maintaining semantic relevance. Our evaluation reveals a consistent failure mode across existing re-rankers: a strong bias toward older, semantically rich documents, even when they are factually obsolete. We further investigate an instruction optimization framework to mitigate this issue. By identifying Pareto-optimal instructions that balance Evolving and Non-Evolving Knowledge tasks, we obtain gains of up to 27% on Evolving Knowledge tasks while maintaining competitive performance on Non-Evolving Knowledge tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2604_14227
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FRESCO: Benchmarking and Optimizing Re-rankers for Evolving Semantic Conflict in Retrieval-Augmented Generation
An, Sohyun
Lee, Hayeon
Yuan, Shuibenyang
Chen, Chun-cheng Jason
Hsieh, Cho-Jui
Mohan, Vijai
Min, Alexander
Information Retrieval
Artificial Intelligence
Retrieval-Augmented Generation (RAG) is a key approach to mitigating the temporal staleness of large language models (LLMs) by grounding responses in up-to-date evidence. Within the RAG pipeline, re-rankers play a pivotal role in selecting the most useful documents from retrieved candidates. However, existing benchmarks predominantly evaluate re-rankers in static settings and do not adequately assess performance under evolving information -- a critical gap, as real-world systems often must choose among temporally different pieces of evidence. To address this limitation, we introduce FRESCO (Factual Recency and Evolving Semantic COnflict), a benchmark for evaluating re-rankers in temporally dynamic contexts. By pairing recency-seeking queries with historical Wikipedia revisions, FRESCO tests whether re-rankers can prioritize factually recent evidence while maintaining semantic relevance. Our evaluation reveals a consistent failure mode across existing re-rankers: a strong bias toward older, semantically rich documents, even when they are factually obsolete. We further investigate an instruction optimization framework to mitigate this issue. By identifying Pareto-optimal instructions that balance Evolving and Non-Evolving Knowledge tasks, we obtain gains of up to 27% on Evolving Knowledge tasks while maintaining competitive performance on Non-Evolving Knowledge tasks.
title FRESCO: Benchmarking and Optimizing Re-rankers for Evolving Semantic Conflict in Retrieval-Augmented Generation
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2604.14227