AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Mnemosyne: Accelerating Multi-Hop Question Answering via Cache Hit Order Fitting

Haizhou Du, Jiujiu Li, Dongyang Li, Luobin Huang, Lisheng Wang

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Explanatory and multiple-choice QA benchmarks

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 cache-aware order fitting (graph reordering of queries) (ablation)Mnemosyne · HotpotQA (sparse)cache hit rate: 22.9 -> 35.3Improved
add cache-aware order fitting (ablation)Mnemosyne · HotpotQA (sparse)latency (s): 0.28 -> 0.18Improved
add L1 exact-match cache (ablation)Mnemosyne · HotpotQA (sparse)F1: 54.2 -> 54.6Improved
add L2 entity-document cache (ablation)Mnemosyne · HotpotQA (sparse)cache hit rate: 4.1 -> 35.3Improved
add idempotency rule (ablation)Mnemosyne · HotpotQA (sparse)F1: 54.0 -> 54.6Improved
raise cache tolerance to 0.4 (ablation)Mnemosyne · HotpotQA (sparse)F1: 54.6 -> 42.4Hurt
lower cache tolerance to 0.1 (ablation)Mnemosyne · HotpotQA (sparse)F1: 54.6 -> 48.9Hurt
increase cache capacity from 20 to 40 (ablation)Mnemosyne · HotpotQA (sparse)F1: 54.6 -> 55.7Improved
increase cache capacity from 40 to 80 (ablation)Mnemosyne · HotpotQA (sparse)F1: 55.7 -> 55.9No clear effect
use approximate semantic (Proximity) caching instead of entity-level cachingRAG caching baseline · HotpotQA (exact dense)delta answer F1: -13.6Hurt
exact-match query caching (CAG-R)naive RAG · four MHQA datasetsretrieval speedup: 0.87-0.98xHurt

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