🤖 AI资讯日报

探讨传统务实企业如何超越AI炒作周期，强调可持续性而非短期热度。

讨论当前AI算法的实际应用痛点，反映技术落地中的挑战。

关注纽约地方法律144对AI行业的影响，涉及法规与伦理问题。

探讨NLP、AI等技术是短暂趋势还是长期变革，分析行业前景。

讨论AI进展速度是否呈指数级增长，涉及技术发展预测。

MIT Non-AI License

The AI Crackpot Index

The Next Bill Gates or Albert Einstein in AI “Chris Clark” – Yourobot

Common Lisp + Machine Learning Internship at Google (Mountain View, CA)

Ask HN: What would you read to learn about "artificial intelligence"?

Show HN: Startup Raising capital through Book Sales

Bioinformatician

提出DPPO算法，用策略散度约束替代PPO的启发式裁剪，解决大语言模型强化学习中的训练不稳定问题。

提出首个多尺度自回归蛋白质骨架生成框架PAR，通过粗到细的跨尺度预测实现高质量结构生成。

Recent progress has rapidly advanced our understanding of the mechanisms underlying in-context learning in modern attention-based neural networks. However, existing results focus exclusively on unimodal data; in contrast, the theoretical underpinnings of in-context learning for multi-modal data remain poorly understood. We introduce a mathematically tractable framework for studying multi-modal learning and explore when transformer-like architectures can recover Bayes-optimal performance in-context. To model multi-modal problems, we assume the observed data arises from a latent factor model. Our first result comprises a negative take on expressibility: we prove that single-layer, linear self-attention fails to recover the Bayes-optimal predictor uniformly over the task distribution. To address this limitation, we introduce a novel, linearized cross-attention mechanism, which we study in the regime where both the number of cross-attention layers and the context length are large. We show that this cross-attention mechanism is provably Bayes optimal when optimized using gradient flow. Our results underscore the benefits of depth for in-context learning and establish the provable utility of cross-attention for multi-modal distributions.

揭示数据中隐性效应的通用机制，提出LLS方法可从偏好数据集中选择子集以诱导模型特定行为。

提出群体进化智能体范式，通过组内经验共享实现开放式自我改进，在代码基准测试中超越现有方法。

介绍量子化学研究协作智能体Quntur，通过推理驱动决策支持计算化学实验的规划与执行。

提出键平滑性表征测试BSCT，作为机器学习原子间势能模型的评估与设计指导指标，提升物理合理性。

综述对比持续学习在物联网中的应用，连接算法设计与系统现实，提出物联网导向的参考架构。

推出持续机器人仿真套件CRoSS，为具有高任务多样性和真实物理模拟的可扩展强化学习提供基准测试。

介绍多模态自然语言驱动的分子编辑智能体Estructural，支持精确的三维分子几何生成与操作。

分析推理语言模型的内部机制，发现其在推理过程中动态优化表征，提出“流体推理表征”概念。

质疑AI能力呈指数增长的观点，提出更复杂模型表明拐点已近，强调现有预测的脆弱性。