V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement Learning
cs.CV, cs.AI
Submitted: 2026-08-26
Updated: 2026-08-26
Project page: https://shulin16.github.io/v-rubrics
Terminology
Sources
- OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles
- Insight-V++: Towards Advanced Long-Chain Visual Reasoning with Multimodal Large Language Models
- LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training
- Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition
- Qwen3-VL Technical Report
- MM-PRM: Enhancing Multimodal Mathematical Reasoning with Scalable Step-Level Supervision
- Qwen2.5-VL Technical Report
- VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Models
- SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models
- OneThinker: All-in-one Reasoning Model for Image and Video
- UniT: Unified Multimodal Chain-of-Thought Test-time Scaling
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- MLLM-Bench: Evaluating Multimodal LLMs with Per-sample Criteria
- A Diagram Is Worth A Dozen Images
- Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
- ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
- ChartQA-X: Generating Explanations for Visual Chart Reasoning
- Tulu 3: Pushing Frontiers in Open Language Model Post-Training
- Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models
- MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models