acceptodds
Desk rejected at ICLR 2027

Residual Retrieval Memory for Forgetting Decisions in Hybrid RAG

Desk rejected at ICLR 2027, so left out of the paper lists and the map.

Abstract

Deciding whether a hybrid retrieval-augmented generation system has forgotten a fact requires testing the distributed state that can reconstruct it, not only the disappearance of one direct answer. On Natural Questions, eight repairs occupy a narrow 0.8–3.1% direct range while spanning 8.2–16.7% paraphrase and 10.4–21.3% multi-hop leakage, producing different stopping decisions from nearly indistinguishable direct scores. We formalize the missing decision object as residual retrieval memory and introduce the Residual Retrieval Access Benchmark (RRA-Bench), which profiles six reconstruction channels, separates query-only attacks from target-bearing reintroduction, and pairs fact compromise with retained utility and repair cost. A crossed study of 150 method–stage–seed cells finds that direct leakage has Spearman [0.15, 0.31] with indirect risk and only 0.61 [0.58, 0.65] AUROC for reintroduction-augmented AnyLeak@32; at a 1% direct threshold, 13 of 18 passing cells remain false-safe. PathSweep, a trace-conditioned allocator over dense, sparse, metadata, reranker, and cache surfaces, provides a constructive test under the same operations and budgets as fixed and validation-selected schedules; it lowers Hard Delete's reintroduction-augmented AnyLeak@32 from 38.6% [36.2,41.1] to 18.7% [16.4,21.2] while retaining 0.793 global MRR. Disjoint prompts, independently authored Qwen-2.5 attacks, 72-hour routine ingestion, and a component-replaced HotpotQA stack preserve the residual-risk separation hidden by direct scores. Residual retrieval memory thus preserves the distinctions needed to stop and rank hybrid-RAG repairs after direct answers disappear.

Primary area:
transfer learning, meta learning, and lifelong learning
Keywords:
memory editing and lifelong learning, Residual Retrieval Memory for Forgetting Decisions in Hybrid RAG

Related papers

Discussion (0)

Sign in to comment.