Skip to content

CollabFlow: Recursive Self-Improvement of Agent Collaboration

Sep 2026 · arXiv.org · Vol abs/2609.38662 · 0 citations · 49 references
Computer Science

TL;DR

This work proposes CollabFlow, an RSI system of Learned Agent Collaboration: a trainable Collab-Director constructs teams of complete Agents, a frozen executor runs them, and each round's outcomes retrain the director, and each round's outcomes retrain the director.

Abstract

Recursive self-improvement (RSI) lets a system improve from its own outcomes; in LLM-based multi-agent systems, Agents refine one another within a task, and outcomes improve how they collaborate across tasks. However, existing multi-agent collaboration leaves this loop open: collaboration is pre-defined at the operator level, topology-only learning keeps verbatim exchange that propagates errors, and reward maximization on a system's own outcomes concentrates on a few teams. To address these challenges, we propose CollabFlow, an RSI system of Learned Agent Collaboration: a trainable Collab-Director constructs teams of complete Agents, a frozen executor runs them, and each round's outcomes retrain the director. Within each round, the edges of a collaboration graph carry protocols of Evidence-Conditioned Communication: a receiver adopts a differing answer only when the sender's evidence is stronger by a margin, so the director learns who communicates and how. Across rounds, we further propose Collaborative Trajectory Balance (CTB), a flow-based objective that credits each team once across its construction orders and targets a reward-proportional distribution over teams, so several good teams stay in play. We also bound how far this self-generated target moves between rounds, which shrinks as records accumulate. On twelve datasets, CollabFlow outperforms all baselines and keeps improving across rounds. Code is available at https://anonymous.4open.science/r/CollabFlow-631E.

View source

Similar papers

#natural language process... Preprint Oct 2026

From Agents to Edges: Paired Optimization for Multi-Agent Collaboration

Strong individual agents do not guarantee effective collaboration. Communication must align what one agent conveys with how another interprets and uses it. This dependency makes the communication edge a natural unit of optimization. We introduce Paired Automated Communication Optimization (PACO), which jointly optimize...

Feng Lin, Xiao-Tian Yu, Yi-Fan Yang et al. · 0 citations
Sep 2026

AgentWorld: Benchmarking Long-Horizon Collaboration of Multi-agent LLMs

To quantify collaboration effectiveness in addition to conventional binary task success, Causal Collaboration Effectiveness (CCE) is proposed, a graph-based metric that traces causal dependencies between agent actions and measures what fraction of a team's effort actually contributed to the outcome.

Raphael Shu, Yu-Sen Zhang, Y. Cho et al. · 2 citations

Rethinking Multi-Agent Collaboration: When More Is Less

SAIGE, a lightweight multi-agent collaboration mechanism based on Semantic-Aware Incremental Graph Evolution, is proposed, suggesting that multi-agent superiority is bounded by task structure rather than universal, and that more agents do not necessarily make a system more intelligent.

Yizhen Yuan, Yi-Bo Wu, Yi-Han Zhang et al. · 0 citations
#artificial intelligence Preprint Oct 2026

Training LLMs for Multi-Agent Collaboration with Verifiable Information Rewards

Effective collaboration among large language model (LLM) agents in multi-agent systems requires sharing relevant evidence, requesting missing information, and incorporating teammates'contributions into decisions. Training these behaviors is challenging because final task outcomes provide limited feedback on the value o...

Ke-Ru Chen, Sen-Fon Lin, Ying-Bin Liang et al. · 0 citations
#machine learning Preprint Oct 2026

RSIGym: A Flexible Environment for Recursive Self-Improvement

Recursive self-improvement requires carrying accepted changes into later improvement cycles, while studying agent-proposed changes also requires substantial research infrastructure. Existing settings often leave agents to rebuild routine infrastructure or restrict exploration to individual components. We introduce RSIG...

Fan-Qing Meng, Ling-Xiao Du, Hao-Cheng Lu et al. · 4 citations
#artificial intelligence Review Oct 2026

RSI-Forge: From Research Papers to Environments for Recursive Self-Improvement

RSI-Forge is introduced, a multi-agent pipeline that turns published papers into executable environments for self-improvement and provides a scalable approach to constructing research environments for training and evaluating self-improving agents.

Ren-Xiong Wang, Darvin Yi, Abril Herrlein et al. · 2 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.