1
发现频道：10款大家发现的好评软件[2026年第34期]
精选近期社区好评的10款软件，从白板工具到复古扫雷，涵盖实用与趣味，值得把玩。
最近10日，来自小众软件论坛的发现频道的热门排行榜，由系统自动生成，直接列出来： 序号 主题 1️⃣ 【开发者自荐】史前动物博物馆，和孩子一起转着看 18 个史前动物 2️⃣ 【自荐】全网热点APP 3️⃣ 免费桌面护眼软件，蓝光过滤 + 可爱宠物久坐提醒，不锁屏不打扰 4️⃣ 基于
2
Evolutionary Algorithm-Guided LLMs for Physics-Informed Neural Network Design
进化算法加持LLM，让物理信息神经网络设计告别试错，智能更高效。
arXiv:2607.15560v1 Announce Type: cross Abstract: Physics-informed neural networks (PINNs) are unusually sensitive to interacting choices of architect…
3
Modularity-Free Conflict-Averse Training for Generalized PINNs
被ICASSP 2026接收的PINN训练新方法，提出无模块化冲突避免策略，提升泛化能力。
arXiv:2606.20156v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have become a powerful framework for solving PDEs by embeddin…
4
Show HN: Kickbacks.ai – pays you for Claude Code spinner impressions
让Claude Code的旋转展示为你赚钱，轻松赚取额外收入
Article URL: https://kickbacks.ai Comments URL: https://news.ycombinator.com/item?id=48500004 Points: 2 # Comments: 0
5
Loss-Conditional PINNs for Parametric PDE Families
引入损失条件机制优化PINNs训练，高效求解参数化偏微分方程族
arXiv:2606.04420v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) approximate solutions of ODEs and PDEs by minimising a weight…
6
PINNs Failure Modes are Overfitting
揭示物理信息神经网络（PINNs）失败的根本原因——过拟合，为改进PINN训练提供新视角
arXiv:2605.30910v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) are a common class of machine learning-based partial differen…
7
Spinning Up in Deep RL: Workshop review
OpenAI首次深度强化学习教育工作坊回顾，亮点在于Spinning Up项目的实践教学与社区反响。
On February 2, we held our first Spinning Up Workshop as part of our new education initiative at OpenAI.