More Than the Final Answer: Improving Visual Extraction and Logical Consistency in Vision-Language Models
Year: 2025
Type: preprint
Abstract: Reinforcement learning from verifiable rewards (RLVR) has recently been extended from text-only LLMs to vision-language models (VLMs) to elicit long-chain multimodal reasoning. However, RLVR-trained VLMs still exhibit two persistent failure modes: inaccurate visual extraction (missing or hallucinating details) and logically inconsistent chains-of-thought, largely because verifiable signals supervise only the final answer. We propose PeRL-VL (Perception and Reasoning Learning for Vision-Language Models), a decoupled framework that separately improves visual perception and textual reasoning on top of RLVR. For perception, PeRL-VL introduces a VLM-based description reward that scores the model's self-generated image descriptions for faithfulness and sufficiency. For reasoning, PeRL-VL adds a text-only Reasoning SFT stage on logic-rich chain-of-thought data, enhancing coherence and logical consistency independently of vision. Across diverse multimodal benchmarks, PeRL-VL improves average Pass@1 accuracy from 63.3% (base Qwen2.5-VL-7B) to 68.8%, outperforming standard RLVR, text-only reasoning SFT, and naive multimodal distillation from GPT-4o.
Source: arXiv (Cornell University)
Authors: Just, Hoang Anh, Fan, Yifei, Zhao, Handong, Gu, Jiuxiang, Zhang, Ruiyi, Jenni, Simon, Kafle, Kushal, Jia, Ruoxi, Shi, Jing
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4417448911
Cites:
Cited by:
Related to: