AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

From Volume to Value: Preference-Aligned Memory Construction for On-Device RAG

Changmin Lee, Jaemin Kim, Taesik Gong

ICML 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning · 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
add semantic-based coarse filtering alone (ablation)Contriever RAG, Llama-3.1-8B · PrefWikipreference-following accuracy (%): 40.88 -> 37.26Hurt
add semantic-based coarse filtering alone (ablation)Contriever RAG, Llama-3.1-8B · PrefRQpreference-following accuracy (%): 68.11 -> 69.00No clear effect
add coarse filtering alone: memory reduction (ablation)Contriever RAG, Llama-3.1-8B · PrefWikimemory usage (MB): 142.16 -> 1.83Improved
add preference-aligned fine verification to coarse filtering (ablation)EPIC, Llama-3.1-8B · PrefWikipreference-following accuracy (%): 37.26 -> 52.53Improved
add preference-aligned fine verification to coarse filtering (ablation)EPIC, Llama-3.1-8B · PrefEvalpreference-following accuracy (%): 27.89 -> 61.58Improved
add fine verification: memory (ablation)EPIC, Llama-3.1-8B · PrefWikimemory usage (MB): 1.83 -> 0.27Improved
add preference-guided query steering (ablation)EPIC, Llama-3.1-8B · PrefWikipreference-following accuracy (%): 52.53 -> 54.07Improved
add preference-guided query steering (ablation)EPIC, Llama-3.1-8B · PrefELI5preference-following accuracy (%): 85.48 -> 87.95Improved
add preference-guided query steering (ablation)EPIC, Llama-3.1-8B · PrefEvalpreference-following accuracy (%): 61.58 -> 65.61Improved
add fine verification (F) to HippoRAG 2 (ablation)HippoRAG 2, Llama-3.1-8B · PrefWikipreference-following accuracy (%): +10.57%pImproved
add coarse filtering to F in HippoRAG 2 (ablation)HippoRAG 2 + F · PrefWikimemory and indexing latency: 6.08x memory reduction, 8.31x indexing latency reductionImproved
add query steering to HippoRAG 2 + C + F (ablation)HippoRAG 2 + C + F · PrefWikipreference-following accuracy (%): +6.49%pImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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