RSIAgent improves agents without weight updates, beats GPT-6 Astra on OSWorld 2.0
Roger_M_Taylor · x · 2026-09-16
Researchers introduce RSIAgent, a framework for recursive self-improvement via autonomous exploration: the agent decides what to explore, executes tasks, verifies outcomes, and consolidates stable action-condition-outcome relationships into memory for reuse — dubbed Scaling Experience.
With base model weights frozen (Kimi-K3 and GLM-5.3), it still keeps improving through acquired experience:
- OSWorld 2.0 (0808 offline): 78.98% Partial Score vs 72.60% for GPT-6 Astra
- Agents' Last Exam (Near-term): 84.82% vs 82.26% for GPT-6 Astra
The takeaway: agents can keep getting better by acquiring, verifying, and reusing their own experience without any weight updates.
Related event: RSIAgent Achieves Recursive Self-Improvement Without Training(3 posts)→
More from coding & agent
- Microsoft's MemSteer turns unlabeled user histories into agent memory supervision — MicrosoftResearch · 2026-11-02
- She fired herself from her own Instagram: AI agent runs the account, views up 82.3% — heyshrutimishra · 2026-11-02
- Sourcegraph: AI writes code faster than teams can review it, driving more outages — AI Engineer · 2026-11-02
- Tsinghua & Microsoft's HarnessMM cuts agent token use by up to 87.93% while raising success rate — rohanpaul_ai · 2026-11-02
- GitHub launches Agent HQ: Anthropic, OpenAI, Google coding agents in Copilot — emmanuelvivier · 2026-11-02
- DeepSeek open-sources Super-Agent, a coding agent that reads team chats to minimize human input — emmanuelvivier · 2026-11-02