Code-Optimise: Self-Generated Preference Data for Correctness and Efficiency.
Proceedings of the Findings of the Association for Computational Linguistics: NAACL 2025, Albuquerque, New Mexico, USA, April 29, 2025
Efficient Online Inference of Vision Transformers by Training-Free Tokenization.
CoRR, 2024
Are Compressed Language Models Less Subgroup Robust?
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, 2023
Multi-word Tokenization for Sequence Compression.
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: EMNLP 2023, 2023
Fast Vocabulary Transfer for Language Model Compression.
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: EMNLP 2022 - Industry Track, Abu Dhabi, UAE, December 7, 2022