Talks
| Title | Institute | Date | Video | Slides |
|---|---|---|---|---|
| Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference | ICML 2024 | 25 Jul 2024 | Link | Link |
| McGill NLP | 22 Apr 2024 | |||
| ServiceNow | 4 Apr 2024 | |||
| University of Edinburgh | 4 Apr 2024 | |||
| Efficient Transformers with Dynamic Token Pooling | ACL 2023 | 11 Jul 2023 | Link | Link |
| University of Cambridge (LTL) | 2 Mar 2023 | |||
| University of Edinburgh | 7 Nov 2022 | |||
| Hierarchical Transformers Are More Efficient Language Models | Harvard CMSA | 8 Dec 2021 | Link | Link |
| DeepMind | 12 Nov 2021 | |||
| ML in PL 2021 Conference | 5 Nov 2021 |