AdaStep: Adaptive Step Credit Weighting for Agentic Reinforcement Learning
cs.LG, cs.CL
Submitted: 2026-10-02
Updated: 2026-10-11
Comments: 21 pages, 3 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
- Progress- and Reliability-Oriented Group Policy Optimization for Agentic Reinforcement Learning
- GLM-5: from Vibe Coding to Agentic Engineering
- When Denser Credit Is Not Enough: Evidence-Calibrated Policy Optimization for Long-Horizon LLM Agent Training
- SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning
- Length-Unbiased Sequence Policy Optimization: Revealing and Controlling Response Length Variation in RLVR
- Qwen2.5 Technical Report
- High-Dimensional Continuous Control Using Generalized Advantage Estimation
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning
- CAPO: Towards Enhancing LLM Reasoning through Generative Credit Assignment
- Qwen3 Technical Report
- Train Less, Learn More: Adaptive Efficient Rollout Optimization for Group-Based Reinforcement Learning
- PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks