2026
-
自私的基因：第 4 章 基因机器
TLDR: Bodies and brains are gene-built action machines: genes set up the machinery, but fast behavior has to be delegated to perception, memory, and decision systems.
-
自私的基因：第 5 章 进犯行为
TLDR: Aggression is not simply about being harsher; stable strategies depend on injury cost, resource value, opponent behavior, and the wider population mix.
-
自私的基因：第 6 章 基因种族
TLDR: Kin selection explains why helping relatives can still serve selfish genes, because the same gene may be preserved through another body.
-
CS336: Lecture 3 - LM Architecture and Hyperparameters
TLDR: LM architecture is a stack of trade-offs across normalization, activations, attention, positional encoding, hyperparameters, stability, and inference cost.
-
CS336: Lecture 4 - Mixture of Experts
TLDR: MoE scales parameter count through sparse expert routing, but the real work is balancing tokens, capacity, communication cost, and specialization.
-
Yuandong Tian talks: search quality is action-space quality
TLDR: More rollouts are not enough. Search becomes powerful when the action space, representation, evaluator, and memory make good trajectories easier to find.
-
Concordia: LLM agents as social simulation actors
Concordia is useful because it treats LLM agents as situated social actors with memory, roles, norms, partial observations, and a world state mediated by a Game Master.
-
Compression Is All You Need: measuring mathematical progress
TLDR: A mathematical abstraction is valuable when it compresses downstream work: proofs become shorter, repeated patterns disappear, and the library becomes easier to extend.