Workshops

Transformers, parallel computation, and logarithmic depth

Visit talk page
Workshops

Capabilities and limitations of Transformer in sequential reasoning

Visit talk page
Workshops

The Parallelism Tradeoff: Understanding Transformer Expressivity Through Circuit Complexity

Visit talk page
Workshops

Emergence and grokking in "simple" architectures

Visit talk page
Workshops

Iterated Models: Expressive Power, Learning, and Chain of Thought

Visit talk page
Workshops

The emergence of clusters in self-attention dynamics

Visit talk page
Workshops

Computational Benefits and Limitations of Transformers and State-Space Models

Visit talk page
Workshops

Transformer Expressivity and Formal Logic

Visit talk page
Workshops

Interpretability Agents

Visit talk page
Workshops

Language Generation in the Limit

Visit talk page