Selective Successor Eligibility Traces for RLVR

arXiv:2605.05965 · cs.LG, cs.AI · Submitted 2026-05-07 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-05-07

Updated: 2026-10-13

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Trace — 10× in this paper · explained in 2 episodes
• Credit assignment — 3× in this paper · explained in 3 episodes
• GRPO — 2× in this paper · explained in 6 episodes
• Group Relative Policy Optimization — 1× in this paper
• Group Relative Policy Optimization (GRPO) — 1× in this paper · explained in 6 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning with Verifiable Rewards — 1× in this paper
• Verifiable Rewards — 1× in this paper · explained in 2 episodes

Sources

• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Proximal Policy Optimization Algorithms
• GRPO-$\lambda$: Credit Assignment improves LLM Reasoning
• Selective Credit Assignment
• Group Sequence Policy Optimization
• REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization
• On-Policy RL with Optimal Reward Baseline
• R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning
• MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention
• GEPO: Group Expectation Policy Optimization for Stable Heterogeneous Reinforcement Learning
• ExGRPO: Learning to Reason from Experience
• RL-PLUS: Countering Capability Boundary Collapse of LLMs in Reinforcement Learning with Hybrid-policy Optimization
• Reasoning with Exploration: An Entropy Perspective
• Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Qwen3 Technical Report
• Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning
• A Minimalist Approach to LLM Reasoning: from Rejection Sampling to Reinforce

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