arXiv:2606. 03398v1 Announce Type: cross Abstract: Formal languages have proven to be effective conduits to understand the inner mechanisms of transformers.
arXiv:2604. 22128v2 Announce Type: replace-cross Abstract: When trained on tasks requiring an understanding of hierarchical structure, transformers have been found to represent this hierarchy in distinct ways: in the geometry of the residual stream, and in stack-like attention patterns maintaining a last-in, first-out ordering.
arXiv:2607. 01218v1 Announce Type: cross Abstract: Transformers use the same forward computation stream to both predict the next token and store useful state for future token predictions.
arXiv:2511. 17864v3 Announce Type: replace Abstract: Recent research has established that the impact of context in a vanilla transformer can be represented implicitly by forming a token-dependent, rank-1 patch to its MLP weights.
arXiv:2607. 18570v1 Announce Type: cross Abstract: Discourse relations provide document structure, critical to language understanding and enabling language model performance and ethicality.
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:2608. 04980v1 Announce Type: cross Abstract: We show that tiny transformers can profitably employ a simple form of Chain of Thought, which we call protoreasoning, allowing us to study step-by-step reasoning on ~1M-parameter models and opening up opportunities for much more detailed experimentation and analysis than is feasible for larger models.
arXiv:2607. 13047v1 Announce Type: new Abstract: Parameter decomposition (PD) decomposes neural networks into interpretable computational components that faithfully reflect the original network's operations.
arXiv:2606. 20959v2 Announce Type: replace Abstract: Language models may encode both outdated facts and their newer replacements.
arXiv:2602. 06774v2 Announce Type: replace Abstract: State Space Models (SSMs) have emerged as an efficient alternative to the Transformer architecture.
arXiv:2606. 17522v1 Announce Type: cross Abstract: Deep neural networks are widely believed to derive their expressive power from their ability to form \textbf{hierarchical representations}, capturing progressively more abstract and compositional features across layers.
arXiv:2502. 20681v3 Announce Type: replace-cross Abstract: Transformers may exhibit two-stage training dynamics during the real-world training process.