arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.22600cs.LGcs.AI
收藏

Transformer收敛到不变的算法核心

Transformers converge to invariant algorithmic cores

Joshua S. Schiffman

AI总结:

研究从Transformer偶然行为转向必然操作,用算法核心提取(ACE)隔离子空间等,发现不同Transformer的功能冗余及语言模型主谓语一致的控制轴,揭示其有简单共享计算结构,针对不变量利于理解和控制机制。

摘要:

训练选择的是行为而非电路,单一训练的神经网络可能只反映某次训练的偶然情况。本文通过提取算法核心,将关注点从Transformer的表现转移到其必然执行的操作上。算法核心提取(ACE)被用于隔离子空间、因果验证并恢复其中实现的算法。研究发现不同Transformer有功能冗余,加速从记忆到泛化的转变,还揭示了语言模型中主谓语一致受单一可操控轴控制,这表明Transformer有更简单的共享计算结构,针对不变量而非参数化有助于理解和控制其机制。

英文摘要:

Training selects for behavior, not circuitry: many weight configurations can implement the same function. Studying any single trained neural network thus risks describing accidents of one training run rather than the computation itself. This work shifts focus from what transformers happen to do to what they must do by extracting algorithmic cores, compact subspaces that are necessary and sufficient for a task and that recur across independently trained models. Here, Algorithmic Core Extraction (ACE) is introduced to isolate these subspaces, causally validate them, and recover the algorithms they implement across settings ranging from synthetic tasks to large-scale pretrained models. Markov-chain transformers embed three-dimensional cores in nearly orthogonal subspaces yet recover identical transition spectra. Modular-addition transformers form compact cyclic cores at grokking that later inflate under continued regularization, redundantly distributing the same computation across many functionally equivalent modes. This functional redundancy is found to accelerate the transition from memorization to generalization, yielding an inverse scaling law for grokking time. In six language models spanning more than two orders of magnitude in scale (GPT-2 Small/Medium/Large, LLaMA-3.1, Gemma-2, and Qwen2.5), subject-verb agreement is governed by a single, steerable axis that aligns across architectures. Flipping this axis inverts grammatical number throughout open-ended generation. Together these results suggest that beneath the apparent complexity of trained transformers lies a simpler, shared computational structure, and that targeting invariants rather than parameterizations may offer a more tractable path to mechanistic understanding and control. Code: https://github.com/joshseth/cores

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑