一作质问持续学习何时真正需要「学习」，揭示任务无关场景下模型无需更新即可泛化，引发对学习本质的再思考。
arXiv:2607.07847v1 Announce Type: new Abstract: As large language models (LLMs) become increasingly capable, the next question is how can we enable mo…
生成式AI模型是否会概率性地“复制”训练数据？这篇论文从理论上剖析了模型记忆与版权边界。
Article URL: https://download.ssrn.com/2026/7/6/7067878.pdf?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjELH%2F%2F%2F%…
颠覆知识蒸馏常规认知：预训练表征只有等价类意义，匹配坐标是伪命题
arXiv:2607.03572v1 Announce Type: cross Abstract: Knowledge distillation is usually framed as a choice of what to match in the teacher - its logits, h…
大模型先验如何提升程序搜索中的经验风险最小化？理论+方法前沿新作
arXiv:2510.14331v3 Announce Type: replace Abstract: We study program-learning methods that are efficient in both samples and computation. Classical le…
凸学习与非仿射聚合看似美妙结合，实则暗藏“危险关系”，这篇论文揭示了潜藏的理论风险。
arXiv:2606.28123v1 Announce Type: new Abstract: Last-iterate convergence and generalization guarantees in first-order convex learning hinge on the mon…
探索测度近似的新结构化方法，为机器学习和概率论提供理论工具
arXiv:2310.09149v3 Announce Type: replace-cross Abstract: We study the approximation of probability measures in the Wasserstein-$p$ distance by struct…
提出“无窥视调优”方法，为大模型后训练提供可证明的泛化界限与鲁棒性保障。
arXiv:2507.01752v4 Announce Type: replace-cross Abstract: Gradient-based optimization is the workhorse of deep learning, offering efficient and scalab…
揭示Grokking中权重范数如何通过交叉熵下的logit尺度中介作用控制延迟泛化，为理解神经网络泛化机制提供新视角。
arXiv:2606.18465v1 Announce Type: new Abstract: Grokking, the delayed jump from memorization to generalization, is usually tied to the weight norm: a …
利用对称性隐私保护新范式，正交等变Transformer让大模型推理更安全
arXiv:2606.16461v1 Announce Type: new Abstract: Running large language models locally is often impractical, pushing inference on sensitive text to thi…
揭秘Transformer缩放定律背后的学习动力学与泛化机制，87页长文深度统一理论框架。
arXiv:2512.22088v3 Announce Type: replace-cross Abstract: The scaling law, a cornerstone of Large Language Model (LLM) development, predicts improveme…
将AI学习比作冰融化，用参数效应重新理解模型复杂度，颠覆传统视角。
Article URL: https://tcz.hu/blog/2026/02/26/singular-learning-theory/ Comments URL: https://news.ycombinator.com/item?id=48476343 Points: 1 # Comments…
探究核赌博机问题的算法与极小极大复杂度，理论机器学习新进展
arXiv:2606.11171v1 Announce Type: new Abstract: Gaussian-process upper confidence bound (GP-UCB) and decision-estimation-coefficient (DEC) methods may…
揭秘数据混合对模型缩放的影响规律，为AI训练中的最优数据配比提供理论解释。
arXiv:2606.08167v1 Announce Type: new Abstract: Recent research has established empirical scaling laws to predict model performance on multi-domain da…
从理论层面形式化语言反馈学习，提供可证明的保证，为机器学习新范式奠定基础。
arXiv:2506.10341v2 Announce Type: replace Abstract: Interactively learning from observation and language feedback is an increasingly studied area driv…
探讨统计决策理论中引入反事实损失的新框架，为因果推断与机器学习决策提供理论支撑。
arXiv:2505.08908v3 Announce Type: replace-cross Abstract: Many researchers apply classical statistical decision theory to evaluate treatment choices a…
群组公平约束下最优传输新框架，ICML 2026 Spotlight论文，理论深度与实践价值兼备。
arXiv:2601.07144v3 Announce Type: replace-cross Abstract: Ensuring fairness in matching algorithms is a key challenge in allocating scarce resources a…
用潜变量模型解释大模型缩放定律，为架构与基准激增提供理论框架。
arXiv:2512.06553v2 Announce Type: replace-cross Abstract: We propose a statistical framework built on latent variable modeling for scaling laws of lar…
二元脉冲神经网络如何从结构上实现因果建模，这篇论文给出了理论框架与可行路径
arXiv:2604.27007v2 Announce Type: replace Abstract: We provide a causal analysis of Binary Spiking Neural Networks (BSNNs) to explain their behavior. …
提出用秩统计量近似f-散度的新方法，理论简洁且计算高效，为信息论和机器学习提供实用工具。
arXiv:2601.22784v2 Announce Type: replace-cross Abstract: We introduce a rank-statistic approximation of $f$-divergences that avoids explicit density-…
揭示学习峰值分布时，训练时间与模型性能呈现普适的1/3幂律缩放，ICML 2026前沿理论发现。
arXiv:2602.03685v2 Announce Type: replace Abstract: Training large language models (LLMs) is computationally expensive, partly because the loss exhibi…