Profiles2023-2025
Varun Kanade
University of Oxford
5Papers
4Last author
4Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers5
- Pause Tokens Strictly Increase the Expressivity of Constant-Depth TransformersAuthor 2 of 2NeurIPS 2025 · 14 citations
- Selective Omniprediction and Fair AbstentionAuthor 2 of 2NeurIPS 2025 · 3 citations
- Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete FunctionsAuthor 5 of 5ICLR 2024 · 77 citations
- Separations in the Representational Capabilities of Transformers and Recurrent ArchitecturesAuthor 4 of 4NeurIPS 2024 · 36 citations
- Simplicity Bias in Transformers and their Ability to Learn Sparse Boolean FunctionsAuthor 3 of 4ACL 2023 · 7 citations
