Tracked topic
Papers arxiv:2606.11075 Exploring the Design Space of Reward Backpropagation for Flow Matching Published on Jun 9 Submitted by Ruoyu on Jun 23 Tencent Authors: Ruoyu Wang , , , Yushi Huang , , Abstract FlowBP…
…Code-Driven Agentic Image Generation Published on May 28 Submitted by Zilong Huang on May 29 Tencent Hunyuan Authors: , , Zilong Huang , Dongzhi Jiang , , , Abstract GenClaw presents a code-driven agentic image generation…
…DiscoBench for Clarification-Aware Deep Search Published on Jun 26 Submitted by ZH on Jul 3 Tencent Hunyuan Authors: , , , , , Zhihao Zhu Abstract DiscoBench evaluates search agents' ability to handle ambiguous queries through…
…Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning Published on Aug 6 Submitted by 3 on Aug 7 Tencent Authors: , , , Yifu Guo , Zhengxi Lu , , , , , , , Abstract EnvACE trains large language model…
…Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification Published on May 7 Submitted by Qihang Fan on May 11 Tencent Authors: Qihang Fan , , , , Abstract UniPrefill is a prefill acceleration framework…
…An Empirical Study of Spatial Memory Staleness in VLM Agents Published on Aug 5 Submitted by Yushi SUN on Aug 6 Tencent Authors: Yushi Sun , Abstract Memory-augmented vision-language agents frequently…
…Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability Published on Jun 17 Submitted by haipengluo on Jun 18 Tencent Hunyuan Authors: Haipeng Luo , , , , , , Abstract GRPO algorithms face policy entropy collapse…
Papers arxiv:2608.05466 Recursive Synthesis for Long-Horizon Terminal Tasks Published on Aug 7 Submitted by LZX on Aug 6 #1 Paper of the day Tencent Hunyuan Authors: Zhongzhi Li , , , Ruhan…
…A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning Published on Jun 9 Submitted by Heming Zou on Jun 11 Tencent Authors: Heming Zou , , , , , , , , , , , Abstract TRACE is a rollout allocation…
…Group-based RLVR as Target-Projection on the LLM Response Simplex Published on May 7 Submitted by Yun Qu on May 11 Tencent Hunyuan Authors: Yun Qu , , , , , , , , , , , , , Abstract Group-based policy gradient…