“Transformers Learn to Compress Variable-order Markov Chains in-Context”, was recognized as the best paper at the Machine Learning and Compression Workshop of NeurIPS 2024. A reprint of this work can be found at preprint of this work at https://arxiv.org/abs/2410.05493.

NeurIPS 2024 was the thirty-eighth Annual Conference on Neural Information Processing Systems, and was held at the Vancouver Convention Center, December 10th – 15th.

Congratulations to Ruida Zhou (UCLA), Dr. Chao Tian (Texas A&M), and Dr. Suhas Diggavi (UCLA) on their accomplishments.