1
小米雷军总结 YU7 八大环节测试直播：实际过程非常顺利，目前整个测试团队有 800 多人
IT之家 6 月 13 日消息，小米董事长兼 CEO 雷军今天在盐城试验场完成了一场 YU7 测试直播，随后他发布长文，对本次直播进行了总结。 雷军表示，今天的测试共有 8 大环节、26 个项目、11 项挑战， 原计划 7 个小时，实际测试过程非常顺利，只用了 5 个半小时就全部完成 。测试车型是 …
2
PaLMR: Towards Faithful Visual Reasoning via Multimodal Process Alignment
PaLMR通过多模态过程对齐实现可信视觉推理，提升大模型对图像的理解与逻辑一致性。
arXiv:2603.06652v2 Announce Type: replace-cross Abstract: Reinforcement learning has recently improved the reasoning ability of Large Language Models …
3
4
Show HN: Realistic procedural fire effect via naive algorithm
仅用朴素算法实现的逼真火焰特效演示，视觉效果惊艳且代码开源可学习。
Source code: https://github.com/Leftium/fx/blob/main/src/routes/fire-plas... I made this naive fire effect as realistic as possible; arguably more rea…
5
Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
无需额外训练，用现成大模型就能给数学推理过程打分，性能媲美专用过程奖励模型。
arXiv:2606.01682v1 Announce Type: cross Abstract: Selecting the best response from multiple small-model samples using a stronger scorer is a simple in…
6
DeepTool: Scaling Interleaved Deliberation in Tool-Integrated Reasoning via Process-Supervised Reinforcement Learning
DeepTool通过过程监督强化学习实现工具集成推理中的交错深思，提升LLM在策略规划与自纠正上的表现。
arXiv:2605.29568v1 Announce Type: new Abstract: Tool-Integrated Reasoning (TIR) extends LLM capabilities by leveraging external environments. However,…
7
Verifiable Process Rewards for Agentic Reasoning
提出可验证过程奖励机制，让智能体推理更可信可解释，强化学习新思路。
arXiv:2605.10325v2 Announce Type: replace Abstract: Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of larg…
8
Understanding and Mitigating Premature Confidence for Better LLM Reasoning
最新研究揭示LLM长思维链中“过早自信”导致的逻辑缺口，并提出基于过程奖励模型的缓解策略，提升推理质量。
arXiv:2605.24396v1 Announce Type: new Abstract: Long chains of thought (CoT) from current language models frequently contain logical gaps and unjustif…
9
A Contractive Feedback Semantics for Reinforcement Learning
从封闭马尔可夫决策过程跳脱，用组合视角打造强化学习的新型收缩反馈语义。
arXiv:2605.24759v1 Announce Type: new Abstract: Discounted reinforcement learning is usually presented through Bellman equations on closed Markov deci…
10
HawkesLLM: Semantic Uncertainty Propagation in Agentic Text Simulation
Hawkes过程结合LLM，让智能体文本模拟中的语义不确定性传播更精准可控。
arXiv:2605.23043v1 Announce Type: new Abstract: Agentic text-simulation systems write in sequence, with each item becoming possible context for later …
11
VRPRM: Process Reward Modeling via Visual Reasoning
通过视觉推理提升过程奖励建模精度，为复杂任务训练提供新思路。
arXiv:2508.03556v3 Announce Type: replace Abstract: Process Reward Model (PRM) is widely used in the post-training of Large Language Model (LLM) becau…
12
rePIRL: Learn PRM with Inverse RL for LLM Reasoning
用逆强化学习从推理轨迹中自动学习过程奖励模型，有效提升大语言模型的复杂推理能力。
arXiv:2602.07832v2 Announce Type: replace Abstract: Process rewards have been widely used in deep reinforcement learning to improve training efficienc…
13
Investigating unauthorized access to GitHub-owned repositories
GitHub首席安全官亲述内部仓库遭非法入侵的完整调查过程
If any impact is discovered, customers will be notified via established incident response and notification channels. The post Investigating unauthoriz…
14
Does AI Know How Many Tokens It Is Burning
AI按token计费看似精确，但背后推理过程复杂且隐藏，揭露了费用计算的模糊性。
The strange thing about the modern AI bill is that it looks precise while the work behind it feels mysterious. A user types a short request, a model t…
15
Mat\'ern Gaussian Processes on Graphs
图上的Matérn高斯过程：理论推导与图结构结合的创新方法，为图数据建模提供新视角。
arXiv:2010.15538v4 Announce Type: replace-cross Abstract: Gaussian processes are a versatile framework for learning unknown functions in a manner that…
16
Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training
超越正确性：通过强化学习调和过程与结果奖励，为模型训练提供新视角
arXiv:2509.03403v2 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) improves final-answer accuracy on reasoning …
17
Designing Against the Gallery: A Two-Year Journey to a Layered Portfolio Experience
两年打磨个人作品集，如何用分层体验打破传统画廊思维，设计创意者的灵感实录。
A two-year journey to create a layered, engaging portfolio beyond the traditional gallery.
18
Variational Optimality of F\"ollmer Processes in Generative Diffusions
从变分角度揭示Föllmer过程在生成扩散中的最优性，理论数学与AI生成的深度交叉
arXiv:2602.10989v2 Announce Type: replace-cross Abstract: We construct and analyze generative diffusions that transport a point mass to a prescribed t…
19
Addressing Terminal Constraints in Data-Driven Demand Response Scheduling
数据驱动方法解决化工流程需求响应调度中的终端约束难题，助力柔性操作与电网平衡。
arXiv:2605.14741v1 Announce Type: cross Abstract: Electrified chemical processes are incentivized by exposure to time-varying electricity markets to o…
20
Best-of-Both-Worlds for Heavy-Tailed Markov Decision Processes
提出首个在重尾MDP上同时实现随机与对抗环境最优遗憾的BoBW算法，突破保守局限。
arXiv:2602.01295v3 Announce Type: replace Abstract: We investigate episodic Markov Decision Processes with heavy-tailed losses (HTMDPs). Existing appr…