Learning When to Think: Dual-Reference Offline Optimization for Adaptive VLM Reasoning
cs.AI
Submitted: 2026-11-06
Updated: 2026-11-18
Code: https://github.com/HongbinLin3589/DR-BOLT
Terminology
Sources
- Qwen Technical Report
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning
- OverThink: Slowdown Attacks on Reasoning LLMs
- Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning
- Learning to Think Fast and Slow for Visual Language Models
- DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization
- Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Reference-Sampled Boltzmann Projection for KL-Regularized RLVR: Target-Matched Weighted SFT, Finite One-Shot Gaps, and Policy Mirror Descent
- VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning
- Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Thinking with Nothinking Calibration: A New In-Context Learning Paradigm in Reasoning Large Language Models
- Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation
- Qwen3 Technical Report
- R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning
- Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Search
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents