Chinese content将估计生成形式化为推荐过程，提出正则化感知推荐者遗憾
这篇论文把 RL 里的估计问题换了个推荐视角，还证明了均匀采样就能达到遗憾下界，做 MCTS 或 bandit 的可以看看。
#MCTS
12 stories · 2 in 7 days · 5 in 30 days
Stories, products and papers tagged "MCTS", newest first.
Oct 26
Oct 24
CodeEvolve: Salesforce Research's LLM Evolutionary System for Code Enhancement
Verified fitness gates, not raw model skill, are what turn LLM-generated code edits into production-safe optimisations — the correctness pipeline is the real contribution here. Engineering teams maintaining legacy codebases, especially Salesforce Apex and Java shops, can apply the runtime-validated search pattern to automate hot-path optimisation without shipping regressions.
Oct 16
IneqAgent Pairs LLM Agents with Symbolic Backends for Lean4 Inequality Proving
The interesting part is the division of labour: the LLM handles high-level lemma strategy while exact symbolic backends supply certificates, keeping every step Lean-verified and cheap at low inference budgets. Formal-methods researchers and theorem-proving engineers can reuse the MCTS-plus-backend pattern for other proof domains. It also gives DeepSeek-V4-Flash a concrete reasoning benchmark data point against other backbones tested in the paper.
Sep 28
Chinese contentPoLar 复现研究：把 Transformer 层当作可调用的函数库
有人在 5 个模型上复现了 PoLar 这篇论文，结论一半成立一半翻车，还发现纠错程序脆弱得离谱，做层路由研究的可以看看。
Sakana AI Introduces SAIL for Test-Time Robot Trajectory Scaling
Test-time scaling, proven in language reasoning, may transfer to robot control without touching model weights — SAIL is an early sign of that shift. Robotics researchers and embodied-AI builders can apply the simulator-plus-VLM-feedback loop to their own stacks, trading compute for reliability. The work competes conceptually with end-to-end approaches like GPT-6 Astra's direct robot operation rather than replacing model training.
Sep 9
Aug 18
Aug 7
Papers09:40
Chinese contentMicroEvo：知识引导的LLM采样用于微架构设计空间探索MicroEvo用LLM引导芯片微架构设计搜索，比NSGA-II质量高36.2%，效率提升10倍多，代码已开源。
Jul 30
Models10:32
Chinese contentBelief-Guided架构：减少围棋AI对MCTS依赖的新方法这篇论文提出一个新架构，让围棋AI不用大量搜索也能在普通电脑上达到专业水平，比AlphaZero更省资源。
Chinese contentKernel Forge: 用MCTS优化CUDA内核的开源Agent Harness
想又快又稳地优化CUDA内核？Kernel Forge用MCTS自动探索，比手动写靠谱多了，实测14个内核超基线。
Jul 15
Papers11:02
Chinese content动态资源分配改进Ensemble Determinization MCTS这篇论文给做游戏AI的同行提供了一个很实用的优化思路：不用固定分配计算资源，而是根据搜索情况动态调整，实测对三个桌游都有明显提升。
Jul 7
Papers11:19
Chinese contentGraph Sparse Sampling：打破连续MDP规划中的视界诅咒这篇论文提出一种新算法GSS，通过共享未来样本来避免树搜索的指数爆炸，在长视界连续控制任务中表现更好。