| title | Visual Para-Thinker++: A Single-Policy Multi-Agent Framework for Visual Reasoning |
|---|---|
| authors | Haoran Xu, Hongyu Wang, Yifei Gao, Jiaze Li, Zizhao Tong, Xiaofeng Zhang, Xiaosong Yuan |
| date | 2026-06-08 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2606.09290 |
| upvotes | 2 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Visual reasoning requires integrating evidence distributed across regions, attributes, and relations, making single-chain reasoning prone to early perceptual commitment and hallucination. We propose Visual Para-Thinker++, a single-policy multi-agent framework in which one shared MLLM policy is instantiated as role-conditioned Main, Worker, and Summary Agents. The Main Agent decomposes the task with fixed allocation patterns; Worker Agents reason in parallel under context isolation; and the Summary Agent reconciles full Worker reasoning traces rather than majority-voting on final labels. The shared policy is trained by Multi-Agent Capability Injection and Role-Decoupled Multi-Agent Optimization, which assign role-specific rewards and advantages to corresponding token segments to reduce gradient conflict among collaborative roles. A native inference engine enables efficient multi-agent rollout through shared visual prefix and KV cache reuse. Across V*, CountBench, the RefCOCO family, and HallusionBench, Visual Para-Thinker++ consistently outperforms single-trajectory and inference-time parallel baselines, with especially strong gains on hallucination-sensitive visual reasoning. |
cite
@misc{xu2026visual,
title={Visual Para-Thinker++: A Single-Policy Multi-Agent Framework for Visual Reasoning},
author={Haoran Xu and Hongyu Wang and Yifei Gao and Jiaze Li and Zizhao Tong and Xiaofeng Zhang and Xiaosong Yuan},
year={2026},
eprint={2606.09290},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2606.09290}
}