Search

Satwik Bhattamishra

Discovering Interpretable Algorithms by Decompiling Transformers to RASP
Automata Learning and Identification of the Support of Language Models
Benefits and Limitations of Communication in Multi-Agent Reasoning
A Formal Framework for Understanding Length Generalization in Transformers
Separations in the Representational Capabilities of Transformers and Recurrent Architectures

Published with Wowchemy — the free, open source website builder that empowers creators.