Multi-agent paper authors: periodically flushing context and keeping a summary doc helps a lot
DimitrisPapail · x · 2026-09-22
In follow-up discussion on their multi-agent test-time communication paper, the authors share a complementary finding: in another project, they periodically flushed the agent's context and kept a running 'summary of what happened so far' doc — this helped a lot even at best-of-1, supporting the tunnel-vision hypothesis for long contexts.
More from Research
- 50+ NeurIPS 2025 papers found containing AI hallucinations that slipped past peer review — CSProfKGD · 2026-11-12
- Indie researcher pauses RNN training over environment bugs, shifts to FunSearch-style tasks — cephaloform · 2026-11-12
- Self-supervised visual similarity partially explains word generalization in humans and machines — Patricio M. Vera · 2026-11-12
- P2Echo: one text-prompted model segments multi-view echocardiography with high accuracy and low cost — UBC-V · 2026-11-12
- I Fixed Alamo Drafthouse's Search With Query Logs Alone: +18.4% MRR, No Model Training — CShorten30 · 2026-11-12
- NYU lab open-sources Outerloop: autonomous research agents that open PRs only when benchmarks improve — mengyer · 2026-11-12