arXiv AI
Jun 11
ProcessThinker: Enhancing Multi-modal Large Language Models Reasoning via Rollout-based Process Reward
arXiv:2606. 11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning.
Video-HopChain: Multi-Hop Questions and Confidence-Gated Exploration for Video Reasoning Models
Video-HopChain introduces a new dataset of 22,550 multi‑hop video questions over 13,378 videos, each question consisting of three to six yes/no sub‑questions whose integer answers sum to a verifiable reward. Training a Qwen3‑VL‑8B model with GRPO on this dataset improves performance across eight video‑understanding benchmarks from 55.4 to 57.9, and the addition of Confidence‑Gated Exploration (CGE) raises the mean to 59.3. The authors release the dataset, checkpoint, and training code for further research.
Related stories
arXiv Machine Learning
Jul 7
Incentivizing Vision Language Models to Search for Long Video Question Answering
arXiv:2607. 02959v1 Announce Type: cross Abstract: We introduce VSeek, an agentic framework that transforms long-video question answering (LVQA) from a passive, single-pass perception task into a multi-turn retrieval process.
arXiv Computer Vision
2d ago
Token-Level Video Reinforcement Learning
arXiv:2610.01973v1 Announce Type: new Abstract: Reinforcement learning (RL) for video generation usually assigns one scalar reward to an entire sampled video. Yet a video is not uniformly flawed: som...
arXiv Computer Vision
Aug 24
Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs
The paper introduces OraRL, a reinforcement learning framework that leverages annotations as oracle rollouts to improve sample efficiency and scalability for video multimodal large language models (MLLMs). By decoupling advantage estimation and employing sign‑balanced pruning, OraRL achieves faster training and better performance across multiple video‑perception benchmarks compared to existing methods. The approach scales from 0.8B to 9B parameters and handles up to 100k prompts, delivering significant gains in temporal mIoU, tracking accuracy, segmentation, and spatial‑intelligence metrics.
arXiv AI
Jun 24
video-SALMONN-R$^3$: Learning to ReWatch, ReAsk, and ReAnswer for Efficient Video Understanding
arXiv:2606. 24477v1 Announce Type: cross Abstract: Video large language models (LLMs) are often constrained by computation and memory budgets, leading them to use reduced frame rates and spatial resolutions, which may cause them to miss critical information for question answering (QA).
arXiv Machine Learning
Jun 2
Towards Sparse Video Understanding and Reasoning
arXiv:2602. 13602v2 Announce Type: replace-cross Abstract: We present \revise (\underline{Re}asoning with \underline{Vi}deo \underline{S}parsity), a multi-round agent for video question answering (VQA).
arXiv AI
Jun 2
Video Reasoning without Training
arXiv:2510. 17045v2 Announce Type: replace-cross Abstract: Video reasoning using Large Multimodal Models (LMMs) relies on costly reinforcement learning (RL) and verbose chain-of-thought, resulting in substantial computational overhead during both training and inference.
arXiv AI
Jun 2
TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL
arXiv:2606. 01599v1 Announce Type: new Abstract: Reinforcement learning (RL) for visual reasoning needs scalable, verifiable, and controllable training signals.
arXiv Computer Vision
Sep 11
From Evaluation to Enhancement: Benchmarking and Improving Think-with-Video Reasoning for Video Generative Models
The paper introduces VWG-Bench, a benchmark covering nine reasoning dimensions and 38 tasks to evaluate video generative models on symbolic reasoning, physical laws, and goal pursuit. It also presents Vid-PRE, a prompt-rewriting framework that offloads reasoning to a VLM, improving logical performance without changing the generator architecture. Experiments show that current models excel at visual quality but struggle with logic-heavy tasks, while Vid-PRE significantly boosts reasoning across multiple generators.
arXiv AI
Jun 30
TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding
arXiv:2508. 07683v2 Announce Type: replace-cross Abstract: Video Temporal Grounding (VTG) aims to localize specific video segments corresponding to natural language queries.
arXiv AI
6d ago
Stepwise Intrinsic Rewards for Reasoning in Large Language Models
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
arXiv Computer Vision
4d ago
Learning via Self-Consistency for Diffusion-based Video Reasoning
arXiv:2609.36826v1 Announce Type: new Abstract: Video generation models have demonstrated emerging zero-shot capabilities for visual reasoning, perception, and other vision tasks. However, diffusion-...