Value Forcing: Aligning Return Distributions at All Noise Levels
cs.LG, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Code: https://github.com/chongyi-zheng/value-forcing
Project page: https://chongyi-zheng.github.io/value-forcing
Terminology
Sources
- Pessimistic Bootstrapping for Uncertainty-Driven Offline Reinforcement Learning
- Neural Temporal-Difference and Q-Learning Provably Converge to Global Optima
- Flow Actor-Critic for Offline Reinforcement Learning
- Unleashing Flow Policies with Distributional Critics
- Value Flows
- Expressive Value Learning for Scalable Offline Reinforcement Learning
- Quantile-Coupled Flow Matching for Distributional Reinforcement Learning
- IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies
- Gaussian Error Linear Units (GELUs)
- Classifier-Free Diffusion Guidance
- Adam: A Method for Stochastic Optimization
- Dual-Flow Reinforcement Learning with State-Aware Exploration
- Reinforcement Learning with Action Chunking
- Flow Matching for Generative Modeling
- Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
- Behavioral Cloning Mystery
- OGBench: Benchmarking Offline Goal-Conditioned RL
- Horizon Reduction Makes RL Scalable
- Flow Q-Learning
- Proximal Policy Optimization Algorithms
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