arXiv:2606. 01599v1 Announce Type: new Abstract: Reinforcement learning (RL) for visual reasoning needs scalable, verifiable, and controllable training signals.
arXiv:2608. 08326v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as an effective approach for improving multimodal reasoning.
arXiv:2606. 29984v1 Announce Type: new Abstract: Reinforcement Learning (RL) is an important paradigm for improving the reasoning capabilities of Vision-Language Models (VLMs).
arXiv:2604. 04917v3 Announce Type: replace-cross Abstract: What does it take to build a visual reasoner that works across charts, science, spatial understanding, and open-ended tasks?
arXiv:2608. 19669v1 Announce Type: cross Abstract: Latent reasoning has advanced multimodal reasoning through a two-stage training paradigm: (1) a helper image is encoded into latent tokens to teach visual chain-of-thought during a supervised fine-tuning (SFT) stage, and (2) these latent tokens are further refined with reward feedback during a reinforcement learning (RL) stage.
arXiv:2606. 11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning.
VBVR-Pro is a closed‑loop testbed that enables native visual reasoning through generation, offering 300 procedurally generated tasks that scale training and allow strong transfer to external benchmarks. It supplies verifiable reward scorers based on deterministic, task‑specific rules, outperforming VLM‑as‑a‑judge approaches and providing reliable signals for reinforcement learning. The suite also facilitates controlled modality studies, revealing that video generation excels at persistent spatiotemporal tracking while interleaved generation offers a compute‑efficient alternative, and highlights the importance of vision‑native trajectories for reasoning.
Latent reasoning has advanced multimodal reasoning through a two-stage training paradigm: (1) a helper image is encoded into latent tokens to teach visual chain-of-thought during a supervised fine-tuning (SFT) stage, and (2) these latent tokens are further refined with reward feedback during a reinforcement learning (RL) stage. In this paper, we identify two key limitations of this framework, one in each stage.
Reinforcement learning with verifiable rewards (RLVR) has emerged as an effective approach for improving multimodal reasoning. However, most existing methods evaluate an entire response using a binary reward based only on final-answer correctness, thereby discarding the supervision available in intermediate reasoning steps.
arXiv:2512. 03438v3 Announce Type: replace Abstract: Agentic reasoning models trained with multimodal reinforcement learning (MMRL) have become increasingly capable, yet they are almost universally optimized using sparse, outcome-based rewards computed based on the final answers.
arXiv:2607. 16727v1 Announce Type: new Abstract: Autoregressive multimodal large language models (MLLMs) suffer from error snowballing: a single incorrect inference early in a chainof-thought (CoT) trace corrupts all downstream reasoning.
The paper introduces V‑Rubrics, a reinforcement‑learning framework that evaluates vision‑language model responses by breaking them into atomic propositions and scoring them on Visual Faithfulness, Reasoning Consistency, and Instruction Following. Using a fine‑tuned Qwen3‑VL‑8B‑Instruct model and a newly created 50K‑example V‑Rubrics dataset, the authors demonstrate that rubric‑based GRPO outperforms both a shared SFT baseline and an answer‑only GRPO, especially on knowledge‑oriented and visually grounded reasoning tasks.
UniCAR‑RL is an annotation‑free reinforcement learning framework designed to improve multimodal large language models’ visual mathematics reasoning. It decouples perception and reasoning by using three branches: Caption‑RL for perception optimization, Reasoning‑RL for logical reasoning with a gold image description, and QA‑RL for end‑to‑end question answering. Experiments show significant gains in mathematical and visual reasoning across various model architectures and scales using only raw short‑answer data.