{"slug": "unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics", "title": "UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics", "summary": "Researchers introduced UniCAR-RL, an annotation-free reinforcement learning framework that decouples perception and reasoning optimization in Multimodal Large Language Models (MLLMs) to improve complex mathematical visual reasoning, according to arXiv paper 2609.13849v1. UniCAR-RL uses three branches — Caption-RL for verifier-guided perception, Reasoning-RL for logic based on a gold image description, and QA-RL for end-to-end alignment — and substantially improves MLLM mathematical and visual reasoning using only raw short-answer data, with strong generalization across architectures and scales.", "body_md": "arXiv:2609.13849v1 Announce Type: new \nAbstract: Multimodal Large Language Models (MLLMs) often struggle with complex mathematical visual reasoning primarily due to a lack of fine-grained perception, causing initial visual hallucinations to directly trigger cascading reasoning failures. In traditional end-to-end reinforcement learning (RL), sparse rewards fail to decouple perceptual hallucinations from logical missteps, hindering targeted perception optimization. Alternatively, fine-tuning with perception-enhanced CoT data incurs high costs and hallucinations. In this paper, we address these challenges by proposing UniCAR-RL, an annotation-free RL framework. By explicitly decoupling the optimization of perception and reasoning during the training process, it achieves isolation and optimization of both capabilities. Specifically, UniCAR-RL consists of three synergistic branches: 1) a Caption-RL branch that optimizes perception capabilities through verifier-guided reasoning validation; 2) a Reasoning-RL branch that performs logical reasoning based on a gold image description to halt cascading errors; 3) a QA-RL branch that retains native end-to-end alignment to ensure robust question-answering performance. Experiments show that UniCAR-RL substantially improves MLLMs' mathematical and visual reasoning using only raw short-answer data. Furthermore, it demonstrates strong generalization across diverse architectures and scales.", "url": "https://wpnews.pro/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics", "canonical_source": "https://www.machinebrief.com/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mat-bim9", "published_at": "2026-09-15 04:00:00+00:00", "updated_at": "2026-09-15 05:31:40.080142+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research", "computer-vision"], "entities": ["UniCAR-RL", "Multimodal Large Language Models", "arXiv", "Caption-RL", "Reasoning-RL", "QA-RL"], "alternates": {"html": "https://wpnews.pro/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics", "markdown": "https://wpnews.pro/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics.md", "text": "https://wpnews.pro/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics.txt", "jsonld": "https://wpnews.pro/news/unicar-rl-seeing-better-before-thinking-deeper-in-visual-mathematics.jsonld"}}