BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards (2026.findings-acl)
Copied to clipboard
| Challenge: | Critic-free reinforcement learning with verifiable rewards (RLVR) is a practical paradigm for aligning Large Language Models. |
| Approach: | They propose a framework that stabilizes advantage estimation by combining prompt-local on-policy statistics with semantic-cluster-conditioned historical moments. |
| Outcome: | Experiments show that RLVR improves training stability and performance compared to critic-based methods . compared with other approaches, RL VR improves in cold-start regimes with binary verifiers . |
Similar Papers
PVPO: Pre-Estimated Value-Based Policy Optimization for Agentic Reasoning (2026.findings-acl)
Copied to clipboard
| Challenge: | extending grouping-based methods to agentic reasoning presents unique challenges . frequent environment interactions and tool invocations render intra-group advantage estimation unstable . |
| Approach: | They propose a grouping-based method that uses a single round of rollouts to stabilize advantage estimation. |
| Outcome: | a new RL framework outperforms grouping-based methods in retrieval tasks and advanced mathematical reasoning benchmarks. |
Think Outside the Policy: In-Context Steered Policy Optimization (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing Reinforcement Learning from Verifiable Rewards (RLVR) methods exhibit limited exploration due to reliance on on-policy rollouts which are limited to the current policy’s distribution, resulting in narrow trajectory diversity. |
| Approach: | They propose a framework that leverages the in-context learning capability of Large Reasoning Models to provide expert guidance using existing datasets. |
| Outcome: | The proposed framework improves RLVR performance and training stability on mathematical reasoning benchmarks. |
MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning (2026.acl-long)
Copied to clipboard
Xiaoliang Fu, Jiaye Lin, Yangyi Fang, Binbin Zheng, Chaowen Hu, Zekai Shao, Cong Qin, Lu Pan, Ke Zeng, Xunliang Cai
| Challenge: | Existing RLVR algorithms rely on rigid, uniform, and symmetric trust region mechanisms . current algorithms lack robustness, asymmetric signal reliability and inefficient gradient utilization . |
| Approach: | They propose a framework to harmonize three dimensions of RLVR algorithms, a paper argues . a binary cutoff is used to discard valuable reinforcement signals, they argue . |
| Outcome: | The proposed framework outperforms baselines in evaluating a robust RLVR solution. |
Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing RLVR algorithms focus on different granularities and have complementary strengths and limitations. |
| Approach: | They propose a framework for reinforcement learning with verifiable rewards that bridges RLVR and GSPO . group-level importance ratios are used to update a policy, which preserves fine-grained credit assignment . |
| Outcome: | The proposed framework outperforms existing methods on seven reasoning benchmarks. |
DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning (2026.findings-acl)
Copied to clipboard
Xiwen Chen, Wenhui Zhu, Peijie Qiu, Xuanzhao Dong, Hao Wang, Haiyu Wu, Huayu Li, Aris Sotiras, Yalin Wang, Abolfazl Razi
| Challenge: | Existing methods for group-relative policy optimization rely on scalar correctness rewards that are often non-injective with respect to semantic content. |
| Approach: | They propose a framework that calibrates the reward signal using the semantic density of sampled groups. |
| Outcome: | The proposed framework outperforms strong baselines on five math benchmarks with 7,000 samples and 55 cost. |
G2RPO-A: Guided Group Relative Policy Optimization with Adaptive Guidance (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in reasoning-centric large language models (LLMs) have significantly expanded the performance boundaries of LLMs, showcasing the immense potential of reasoning-enhanced models. |
| Approach: | They propose an adaptive algorithm that injects ground-truth reasoning steps into roll-out trajectories to compensate for SLMs’ inherent weaknesses. |
| Outcome: | Experiments on mathematical reasoning and code-generation benchmarks confirm that G2RPO-A substantially outperforms vanilla GRPO. |
Outcome-Grounded Advantage Reshaping for Fine-Grained Credit Assignment in Mathematical Reasoning (2026.acl-long)
Copied to clipboard
Ziheng Li, Liu Kang, Feng Xiao, Luxi Xing, Qingyi Si, Zhuoran Li, Weikang Gong, Deqing Yang, Yanghua Xiao, Hongcheng Guo
| Challenge: | Group Relative Policy Optimization (GRPO) uses a coarse-grained credit assignment mechanism that propagates group-level rewards uniformly to to every token in a sequence, neglecting the varying contribution of individual reasoning steps. |
| Approach: | They introduce Outcome-grounded Advantage Reshaping (OAR) which redistributes advantages based on how much each token influences the model’s final answer. |
| Outcome: | Empirical results show that OAR-G outperforms GRPO on a high-fidelity attribution signal and suppresses low-impact tokens while preserving the advantage mass. |
MDP-GRPO: Stabilized Group Relative Policy Optimization for Multi-Constraint Instruction Following (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) can follow many natural-language instructions, yet they remain brittle when a request bundles multiple explicit constraints, such as asking the LLM to respond in a particular structure with an exact ending phrase. |
| Approach: | They propose a method which stabilizes learning through multi-temperature sampling to increase reward dispersion, dual-anchor advantages to restore gradients in homogeneous groups, prospect-theoretic shaping to bound updates and penalize violations based on Kahneman Tversky’s theory and asymmetric KL regularization. |
| Outcome: | The proposed method outperforms standard GRPO on FollowBench, IFEval, and a curated multi-constraint dataset, improving strict constraint satisfaction by up to 5.0% on Llama-3.2-3B. |
Unlocking Exploration in RLVR: Uncertainty-aware Advantage Shaping for Deeper Reasoning (2026.findings-acl)
Copied to clipboard
| Challenge: | Reinforcement Learning with Verifiable Rewards (RLVR) has shown significant promise for enhancing the reasoning capabilities of large language models (LLMs). |
| Approach: | They propose a model-free method that refines credit assignment by leveraging the model's internal uncertainty signals. |
| Outcome: | Extensive experiments on five mathematical reasoning benchmarks show that the proposed method outperforms strong RLVR baselines on multiple model scales, including 1.5B and 7B. |
Low-probability Tokens Sustain Exploration in Reinforcement Learning with Verifiable Reward (2026.findings-acl)
Copied to clipboard
Guanhua Huang, Tingqiang Xu, Mingze Wang, Qi Yi, Xue Gong, Siheng Li, Ruibin Xiong, Kejiao Li, Yuhao Jiang, Bo Zhou
| Challenge: | Recent studies show that RLVR training is slow and results plateau as policy entropy collapses . low-probability regularization (Lp-Reg) reduces the number of low-quality exploratory tokens induced by RL training . |
| Approach: | They propose a method to reduce RLVR over-penalization by eliminating low-probability exploratory tokens . they propose 'Low-provability Regularization' to reduce the gradual elimination of low-quality exploratory entropy tokens. |
| Outcome: | The proposed method eliminates low-probability exploratory tokens and prevents suppression of potentially valuable low-property candidates. |