DISCOVER: Online Variance-Guided Data Discovery for Budgeted Multimodal GRPO
cs.LG, cs.AI
Submitted: 2026-11-08
Updated: 2026-11-08
Project page: https://peymoon.github.io/discover/Abstract
Terminology
Sources
- Qwen3-VL Technical Report
- Prompt replay: speeding up grpo with on-policy reuse of high-signal prompts
- SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models
- VADE: Variance-Aware Dynamic Sampling via Online Sample-Level Difficulty Estimation for Multimodal RL
- MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources
- Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking
- LIMR: Less is More for RL Scaling
- Less is More: High-value Data Selection for Visual Instruction Tuning
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- A Relative-Budget Theory for Reinforcement Learning with Verifiable Rewards in Large Language Model Reasoning
- Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis