1
Building Better Activation Oracles
改进激活预言机训练，解决幻觉和模糊性，可提升残差流激活的可解释性研究价值。
arXiv:2606.02609v1 Announce Type: cross Abstract: Activation Oracles (AOs) are promising methods for interpreting residual stream activations. However…
2
Bottom-up Policy Optimization: Your Language Model Policy Secretly Contains Internal Policies
将LLM策略分解为内部层与模块策略，揭示底层优化新路径，前沿RL研究突破。
arXiv:2512.19673v3 Announce Type: replace-cross Abstract: Existing reinforcement learning (RL) approaches treat large language models (LLMs) as a unif…
3
Dynamics of the Transformer Residual Stream: Coupling Spectral Geometry to Network Topology
将Transformer深度视为离散时间，揭示残差流中的谱几何与网络拓扑耦合机制，为理解大模型计算传播提供新视角。
arXiv:2605.14258v1 Announce Type: cross Abstract: Large language models are remarkably capable, yet how computation propagates through their layers re…
4
Residual Stream Duality in Modern Transformer Architectures
揭示Transformer残差流不仅是优化通道，更是模型表示核心——两轴视角（序列位置×层深度）重构设计空间，自注意力与残差路径协同新理解。
arXiv:2603.16039v2 Announce Type: replace-cross Abstract: Recent work has made clear that the residual pathway is not mere optimization plumbing; it i…
5
You can predict LLM output sensitivity in closed form
一个独特视角：从残差流扰动角度解释LLM输出稳定区域的几何形状与学习动态。
Article URL: https://noahgolmant.com/blog/stable-regions-residual-stream/ Comments URL: https://news.ycombinator.com/item?id=48182506 Points: 2 # Comm…