More Than the Final Answer: Improving Visual Extraction and Logical Consistency in Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Just, Hoang Anh, Fan, Yifei, Zhao, Handong, Gu, Jiuxiang, Zhang, Ruiyi, Jenni, Simon, Kafle, Kushal, Jia, Ruoxi, Shi, Jing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914200644747264
author Just, Hoang Anh
Fan, Yifei
Zhao, Handong
Gu, Jiuxiang
Zhang, Ruiyi
Jenni, Simon
Kafle, Kushal
Jia, Ruoxi
Shi, Jing
author_facet Just, Hoang Anh
Fan, Yifei
Zhao, Handong
Gu, Jiuxiang
Zhang, Ruiyi
Jenni, Simon
Kafle, Kushal
Jia, Ruoxi
Shi, Jing
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_12487
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle More Than the Final Answer: Improving Visual Extraction and Logical Consistency in Vision-Language Models
Just, Hoang Anh
Fan, Yifei
Zhao, Handong
Gu, Jiuxiang
Zhang, Ruiyi
Jenni, Simon
Kafle, Kushal
Jia, Ruoxi
Shi, Jing
Computer Vision and Pattern Recognition
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.
title More Than the Final Answer: Improving Visual Extraction and Logical Consistency in Vision-Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.12487