TreeOPSD is introduced, a critic-free, tree-aware posterior self-distillation framework for long-horizon agentic reinforcement learning that improves agentic capabilities across several competitive benchmarks and model scales, especially for long-horizon tasks.
Heming Zou, Qi Wang, Yi-Xiu Mao et al.· 2 citations
This paper revisits the standard state-only formulation of value estimation and proposes PPO, a self-privileged actor-critic framework that consistently improves value-estimation quality by a substantial margin and outperforms representative actor-critic and critic-free RLVR baselines on challenging mathematical reason...
Kun Liang, Chenming Tang, Clive Bai et al.· 0 citations
The Alpha-Stabler framework is proposed, a plug-and-play framework with a Predictor that monitors principal-subspace intrusion for early collapse warnings, and a Controller that removes the principal-subspace component of activation gradients during backpropagation while preserving the orthogonal complement.
Yu-Chen Cai, Ding Cao, Qi-Xiang Yin et al.· 1 citation
We present Hunyuan-A13B, an open-source large language model based on a Mixture-of-Experts architecture. It contains 80 billion total parameters but activates only 13 billion during inference, balancing model capability, computational efficiency, and deployment cost. The model is pretrained on a rigorously filtered 20T...
Tencent Hunyuan Team, Ao Liu, Bo Zhou et al.· 0 citations
Experiments on Qwen3-VL-4B and Qwen3-VL-8B show that incorporating pattern-specific penalties into reinforcement learning can mitigate cross-mode misalignment while incurring a marginal task performance trade-off.
Xin-Ming Wang, Wei-Nong Wang, Hong-Ming Yang et al.· 2 citations· ⚡1
This work compares three fusion paradigms by the artefacts they reuse and suggests that Merge when experts already exist and cheap fusion is paramount; Mix RL when training a unified model without experts, with domain proportions adjusted for cross-domain transfer; and MOPD when preserving domain-specific gains matters...
Sicheng Wu, Kai Yang, Yu-Chen Cai et al.· 5 citations· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.