Ant Group & RUC Introduce SearchOS: Shared Memory Prevents Search Agent Loops
jiqizhixin · x · 2026-08-08
To address the issue of AI search agents getting stuck in loops instead of finding answers, Ant Group and Renmin University of China introduced SearchOS, a multi-agent framework.
The core mechanism of SearchOS is providing a shared memory for all search agents. Instead of working blindly, agents track progress through a common state that logs completed tasks, gathered evidence, coverage gaps, and past failures, ensuring the team never repeats a dead-end.
Additionally, a smart scheduler fills idle slots by assigning the most promising unfinished work, while a middleware layer monitors stalled searches and steps in before time and budget run out. On the WideSearch and GISA benchmarks, SearchOS outperformed all single-agent and multi-agent baselines in quality, completeness, and efficiency for open-domain information seeking.
More from coding & agent
- Constraints you force on models today will become tech debt in months, warns dev — zack_overflow · 2026-10-16
- Liquid AI tutorial: fine-tune your d1-3B decision model in 30 minutes on a free GPU — helloiamleonie · 2026-10-16
- Alibaba open-sources OpenGCUA, a full-stack generalist computer-use agent; +19.6 points on OSWorld-v2 across 8 models — Tongyi-MAI · 2026-10-16
- BotBeat: a social network where all users are AI agents, exposed via 32 MCP tools — Dismal_Piccolo4973 · 2026-10-16
- DKGI paper: agents write their own tools to search knowledge graphs, beating baselines by 6.7 Hit@1 — yuz9yuz · 2026-10-16
- Dev open-sources lgtmaybe: running PR code reviews on a local 27B model via hunk splitting — mattjcoles · 2026-10-16