Research

Paper

TESTING February 26, 2026

AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning

Authors

Yutong Wang, Siyuan Xiong, Xuebo Liu, Wenkang Zhou, Liang Ding, Miao Zhang, Min Zhang

Abstract

While Multi-Agent Systems (MAS) excel in complex reasoning, they suffer from the cascading impact of erroneous information generated by individual participants. Current solutions often resort to rigid structural engineering or expensive fine-tuning, limiting their deployability and adaptability. We propose AgentDropoutV2, a test-time rectify-or-reject pruning framework designed to dynamically optimize MAS information flow without retraining. Our approach acts as an active firewall, intercepting agent outputs and employing a retrieval-augmented rectifier to iteratively correct errors based on a failure-driven indicator pool. This mechanism allows for the precise identification of potential errors using distilled failure patterns as prior knowledge. Irreparable outputs are subsequently pruned to prevent error propagation, while a fallback strategy preserves system integrity. Empirical results on extensive math benchmarks show that AgentDropoutV2 significantly boosts the MAS's task performance, achieving an average accuracy gain of 6.3 percentage points on math benchmarks. Furthermore, the system exhibits robust generalization and adaptivity, dynamically modulating rectification efforts based on task difficulty while leveraging context-aware indicators to resolve a wide spectrum of error patterns. Our code and dataset are released at https://github.com/TonySY2/AgentDropoutV2.

Metadata

arXiv ID: 2602.23258
Provider: ARXIV
Primary Category: cs.AI
Published: 2026-02-26
Fetched: 2026-02-27 04:35

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23258v1</id>\n    <title>AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning</title>\n    <updated>2026-02-26T17:31:43Z</updated>\n    <link href='https://arxiv.org/abs/2602.23258v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23258v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>While Multi-Agent Systems (MAS) excel in complex reasoning, they suffer from the cascading impact of erroneous information generated by individual participants. Current solutions often resort to rigid structural engineering or expensive fine-tuning, limiting their deployability and adaptability. We propose AgentDropoutV2, a test-time rectify-or-reject pruning framework designed to dynamically optimize MAS information flow without retraining. Our approach acts as an active firewall, intercepting agent outputs and employing a retrieval-augmented rectifier to iteratively correct errors based on a failure-driven indicator pool. This mechanism allows for the precise identification of potential errors using distilled failure patterns as prior knowledge. Irreparable outputs are subsequently pruned to prevent error propagation, while a fallback strategy preserves system integrity. Empirical results on extensive math benchmarks show that AgentDropoutV2 significantly boosts the MAS's task performance, achieving an average accuracy gain of 6.3 percentage points on math benchmarks. Furthermore, the system exhibits robust generalization and adaptivity, dynamically modulating rectification efforts based on task difficulty while leveraging context-aware indicators to resolve a wide spectrum of error patterns. Our code and dataset are released at https://github.com/TonySY2/AgentDropoutV2.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <published>2026-02-26T17:31:43Z</published>\n    <arxiv:primary_category term='cs.AI'/>\n    <author>\n      <name>Yutong Wang</name>\n    </author>\n    <author>\n      <name>Siyuan Xiong</name>\n    </author>\n    <author>\n      <name>Xuebo Liu</name>\n    </author>\n    <author>\n      <name>Wenkang Zhou</name>\n    </author>\n    <author>\n      <name>Liang Ding</name>\n    </author>\n    <author>\n      <name>Miao Zhang</name>\n    </author>\n    <author>\n      <name>Min Zhang</name>\n    </author>\n  </entry>"
}