New paper: test-time communication beats independent agents, Team@k > Best@k
DimitrisPapail · x · 2026-09-21
A new paper by Dimitris Papail and colleagues argues test-time communication is a next axis for scaling capabilities. N identical agents worked on the same task with no roles, sharing only a text log and told to "collaborate" — across three research-heavy tasks, communicating teams beat independent agents decisively (Team@k > Best@k). Commenters note that single-agent long-horizon gains saturate versus independent sampling, and ask whether communication could reshape the scaling curve to look more human-like, echoing the Hugging Face incident where agents exploited any channel they found.
More from coding & agent
- MechFaber: Claude Code designs a 99-part quadruped with firmware co-simulated in Renode and MuJoCo — SpeedyBrowser45 · 2026-09-22
- Exa MCP hits 5,000 GitHub stars as AI agents flock to its search integration — TheIshanGoswami · 2026-09-22
- 670,000 agent skills, no trust layer: bot scan finds 69% never reliably fire — markjeffrey · 2026-09-22
- Eight disruptive use cases for Jev: from millisecond evals to AI guardrails — nkmrao · 2026-09-22
- Training on production traces: single-trajectory RL may unlock continual learning — rhythmrg · 2026-09-22
- Anthropic's Swiss cheese model explains why passing evals isn't enough for agents — hugobowne · 2026-09-22