Agent-Controlled Search Over Raw Chat Logs Beats Structured Memory Systems
_reachsumit · x · 2026-08-14
An arXiv paper titled When Your Agent Opens the Chat App introduces ReFind, a framework that challenges the current reliance on structured memory (like summaries, knowledge graphs, or trees) for AI agents.
ReFind builds no semantic structure at all. Instead, it leaves raw conversation histories unmodified and indexes them lexically at turn granularity. It uses an iterative keyword-search loop combined with four chat-native controls: session-aware rank fusion, local context expansion, temporal narrowing, and skipping already-inspected sessions.
Tested under the strict incremental multi-turn setting of MemoryAgentBench (2,800 questions covering multi-hop QA, event ordering, etc.), a ReFind system backed by GPT-4o-mini achieved the highest mean accuracy of 58.2%. It outperformed the strongest graph- and tree-based memory baselines, such as HippoRAG 2 (53.2%), proving that competent retrieval over raw logs is highly effective.
More from coding & agent
- Developer reflects: Superhuman AI coders enable effortless parallel problem-solving — repligate · 2026-08-14
- AI Models Prefer Hand-Coding Retry Logic Over Using Existing Libraries — rakyll · 2026-08-14
- Now It's Called 'Installing a Skill' — yuwen_lu_ · 2026-08-14
- ComfyUI Update: MiniMax H3 Now Supports Anchoring Image and Audio Guides at Any Frame — fruesome · 2026-08-14
- ComfyUI Merges PR: MiniMax H3 Now Supports Image and Audio Guides at Any Frame — fruesome · 2026-08-14
- Indie Dev Builds 7 AI Employees Using Hermes Agent — ayushtweetshere · 2026-08-14