TeamTR: Trust-Region Fine-Tuning for Multi-Agent LLM Coordination

• URL: http://arxiv.org/abs/2605.15207v1
• Date: Fri, 01 May 2026 23:42:57 GMT
• Title: TeamTR: Trust-Region Fine-Tuning for Multi-Agent LLM Coordination
• Abstract summary: TeamTR is a trust-region framework that resamples trajectories after each component update and enforces per-agent divergence control.<n>We show that TeamTR outperforms single-agent and sequential baselines with 7.1% on average, mitigates coordination regressions, and supports plug-and-play component replacement.
• Score: 11.65571332626047
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent LLM systems have shown promise for complex reasoning, yet recent evaluations reveal they often underperform single-model baselines. We identify a structural failure mode in sequential fine-tuning of shared-context teams: updating one agent shifts the team's context distribution, and when subsequent updates are evaluated on cached rollouts, this mismatch compounds. We formalize this as the compounding occupancy shift and prove that stale-occupancy evaluation incurs a penalty that scales quadratically with the number of agents. In contrast, intermediate-occupancy evaluation reduces this to linear scaling. We propose TeamTR, a trust-region framework that resamples trajectories after each component update and enforces per-agent divergence control, yielding rigorous per-update and per-stage improvement lower bounds. Experiments show that TeamTR outperforms single-agent and sequential baselines with 7.1% on average, mitigates coordination regressions, and supports plug-and-play component replacement. Code is available at https://github.com/Yydc/TeamTR.

Related papers

• Adversarial Review: Structured Disagreement for Grounded Agentic Code Review [0.0]
  We study whether a subagent paradigm can support a middle ground: minimal agentic cooperation without the overhead of large multi-agent teams.<n>We introduce Adversarial Review (AR), a minimal cooperative code-review protocol in which a main coding agent works with a reviewer and a critic agent.
  arXiv  Detail & Related papers  (2026-08-16T21:58:17Z)
• Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents [19.290397279235084]
  This paper aggregates the largest coordinated deep-dive of one MCP-based industrial-agent benchmark to date.<n>We argue that aggregate-score leaderboards systematically underspecify deployed-agent evaluation.
  arXiv  Detail & Related papers  (2026-06-18T02:02:46Z)
• Signal Reshaping for GRPO in Weak-Feedback Agentic Code Repair [36.31472731207028]
  We study signal reshaping for standard GRPO under such feedback.<n>Experiments show a clear end-to-end gain: full signal-reshaped GRPO improves strict compile-and-semantic accuracy.
  arXiv  Detail & Related papers  (2026-05-08T05:41:25Z)
• Improving the Efficiency of Language Agent Teams with Adaptive Task Graphs [52.26652574704317]
  Large language models (LLMs) are increasingly deployed in teams, yet existing coordination approaches often occupy two extremes.<n>We introduce Language Agent Teams for Task Evolution (LATTE), a framework for coordinating LLM teams inspired by distributed systems.
  arXiv  Detail & Related papers  (2026-05-07T14:19:17Z)
• CAPO: Counterfactual Credit Assignment in Sequential Cooperative Teams [6.8920524670882815]
  We introduce the Sequential Aristocrat Utility (SeqAU), the unique per-agent learning signal.<n>We derive CAPO, a critic-free policy-gradient algorithm.
  arXiv  Detail & Related papers  (2026-04-20T01:14:59Z)
• SAT: Sequential Agent Tuning for Coordinator Free Plug and Play Multi-LLM Training with Monotonic Improvement Guarantees [20.52379192411959]
  Large language models (LLMs) with a large number of parameters achieve strong performance but are often prohibitively expensive to deploy.<n>Recent work explores using teams of smaller, more efficient LLMs that collectively match or even outperform a single large model.<n>We address this by introducing Sequential Agent Tuning (SAT), a coordinator-free training paradigm.
  arXiv  Detail & Related papers  (2026-04-17T01:45:30Z)
• Scaling Test-Time Compute for Agentic Coding [126.72747643609274]
  We propose a test-time scaling framework for agentic coding based on compact representations of rollout trajectories.<n>Our framework converts each rollout into a structured summary that preserves its salient hypotheses, progress, and failure modes.<n>Our method consistently improves the performance of frontier coding agents across SWE-Bench Verified and Terminal-Bench v2.0.
  arXiv  Detail & Related papers  (2026-04-16T17:39:33Z)
• SQL-ASTRA: Alleviating Sparse Feedback in Agentic SQL via Column-Set Matching and Trajectory Aggregation [51.696027315573296]
  We propose a framework featuring a universal two-tiered reward mechanism designed to provide effective trajectory-level evaluation and dense step-level signals.<n>In traditional paradigms, rewards are determined solely by the final-turn feedback, which ignores the intermediate process and leads to ambiguous credit evaluation.<n>We show that our approach outperforms SOTA Arctic-Text2-R1-7B on BIRD 2.0 using identical models, propelling Text-to-reward convergence toward a robust multi-turn agent paradigm.
  arXiv  Detail & Related papers  (2026-03-17T06:26:36Z)
• AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation [39.61543921719145]
  AgentSelect is a benchmark that reframes agent selection as narrative query-to-agent recommendation.<n>It converts heterogeneous evaluation artifacts into unified, positive-only interaction data.<n>AgentSelect provides the first unified data and evaluation infrastructure for agent recommendation.
  arXiv  Detail & Related papers  (2026-03-04T06:17:51Z)
• DLLM Agent: See Farther, Run Faster [94.74432470237817]
  Diffusion large language models (DLLMs) have emerged as an alternative to autoregressive (AR) decoding with appealing efficiency and modeling properties.<n>We study this in a controlled setting by instantiatingDLLM and AR backbones within the same agent workflow.<n>We find thatDLLM Agents are on average over 30% faster end to end than AR agents, with some cases exceeding 8x speedup.
  arXiv  Detail & Related papers  (2026-02-07T09:01:18Z)
• MALT: Improving Reasoning with Multi-Agent LLM Training [67.76186488361685]
  MALT (Multi-Agent LLM Training) is a novel post-training strategy that divides the reasoning process into generation, verification, and refinement steps.<n>On MATH, GSM8K, and CSQA, MALT surpasses the same baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40% respectively.
  arXiv  Detail & Related papers  (2024-12-02T19:30:36Z)
• Order Matters: Agent-by-agent Policy Optimization [41.017093493743765]
  A sequential scheme that updates policies agent-by-agent provides another perspective and shows strong performance. We propose the textbfAgent-by-textbfagent textbfPolicy textbfOptimization (A2PO) algorithm to improve the sample efficiency.
  arXiv  Detail & Related papers  (2023-02-13T09:24:34Z)
• Group DETR: Fast DETR Training with Group-Wise One-to-Many Assignment [80.55064790937092]
  One-to-many assignment, assigning one ground-truth object to multiple predictions, succeeds in detection methods such as Faster R-CNN and FCOS. We introduce Group DETR, a simple yet efficient DETR training approach that introduces a group-wise way for one-to-many assignment. Experiments show that Group DETR significantly speeds up the training convergence and improves the performance of various DETR-based models.
  arXiv  Detail & Related papers  (2022-07-26T17:57:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.