We present a theoretical framework to explain the emergence of inductive reasoning abilities in Transformer language models. While previous works on Transformer learning dynamics have so far been mostly tied to specific tasks, we study a generalized class of inductive tasks that unifies several synthetic tasks known in the literature, including in-context n-grams and multi-hop reasoning.
arXiv:2607. 11875v1 Announce Type: cross Abstract: We present a theoretical framework to explain the emergence of inductive reasoning abilities in Transformer language models.
arXiv:2510. 25013v2 Announce Type: replace-cross Abstract: Mechanistic interpretability aims to reverse-engineer large language models (LLMs) into human-understandable computational circuits.
arXiv:2601. 15158v4 Announce Type: replace-cross Abstract: Transformers trained via Reinforcement Learning (RL) with outcome-based supervision can spontaneously develop the ability to generate intermediate reasoning steps (Chain-of-Thought).
arXiv:2604. 07822v2 Announce Type: replace-cross Abstract: We study implicit reasoning, i.
arXiv:2502. 20681v3 Announce Type: replace-cross Abstract: Transformers may exhibit two-stage training dynamics during the real-world training process.
arXiv:2410. 24050v3 Announce Type: replace Abstract: Large-scale pretraining of transformers has been central to the success of foundation models.
arXiv:2607. 15178v1 Announce Type: cross Abstract: Transformer reasoning is limited by autoregressive decoding, which repeat edly compresses rich hidden computation through token space and makes it difficult for intermediate reasoning states to persist across time.
arXiv:2602. 22600v2 Announce Type: replace-cross Abstract: Training selects for behavior, not circuitry: many weight configurations can implement the same function.
arXiv:2607. 19405v1 Announce Type: new Abstract: The CoTFormer architecture formalizes Chain-of-Thought as a form of recurrent latent computation, preserving intermediate states as attendable representations to mimic explicit reasoning traces.
arXiv:2607. 07316v1 Announce Type: new Abstract: This article offers a comprehensive overview of mechanistic interpretability, an emerging field that seeks to reverse-engineer the internal algorithms of modern neural networks.
arXiv:2510. 18315v2 Announce Type: replace-cross Abstract: We investigate how embedding dimension affects the emergence of an internal "world model" in a transformer trained with reinforcement learning to perform bubble-sort-style adjacent swaps.
arXiv:2509. 24653v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) excel at multi-hop reasoning in distribution, yet fail on unseen compositions, a phenomenon known as the curse of two-hop reasoning.