This work studies how supervised fine-tuning data quality and reinforcement learning (RL) optimization design affect performance on Countdown arithmetic reasoning under a reinforcement learning with verifiable rewards (RLVR) setup, and investigates whether replay-based reinforcement learning can improve sample efficiency during RL fine-tuning.
VarIance Guided Online Rollout allocation (VIGOR) is proposed which instead of allocating a fixed rollout budget per example, begins with a small number of rollouts for all examples in a batch and iteratively allocates additional rollouts to those with the highest group reward variance until a fixed total rollout budge...
Heyang Jiang, Henry Liu, Baharan Mirzasoleiman· arXiv.org· 1 citation
DataFlex-RL, an evaluation platform for comparing choices under a common GRPO recipe, is introduced, finding that changing the data policy measurably changes the training process but does not produce a reproducible improvement over uniform training.
Hao Liang, Ming-Rui Chen, Hengyi Feng et al.· 2 citations
It is shown that on-policy reinforcement learning with verifiable rewards (RLVR) can improve the current objective while making successful behaviors for later objectives too rare to sample and reinforce, and effective rewardable support is defined as successful trajectories reachable within a fixed rollout budget.
Shaohang Wei, Z.Y. Su, Feifan Song et al.· 0 citations
Empirical findings suggest that CudaPerf significantly outperforms strong baselines, including Qwen-3-32B and CUDA Agent by achieving up to 5X and 3.32X improvements in speedup, and 17%&7% improvements in correctness, respectively.
Q. I. Mahmud, Nesreen K. Ahmed, Ali Jannesari· arXiv.org· 0 citations
Results support a qualified internalized-search reading: under the recipe the authors test, much of the measured RL gain corresponds to a change in sampling efficiency toward operating points the base model can already reach under search.
Wen-He Sun, Cun-Xiang Wang, Zi-Jun Yao et al.· 0 citations
DATPO (Difficulty-Adaptive Sentence-entropy-guided Tree-structured Policy Optimization) integrates difficulty-adaptive tree search with a sibling-diversity advantage term, explicitly promoting semantic diversity to expand reasoning coverage during training.
Young Kyu Yu, Sanghwan Jang, Hwanjo Yu· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
