Value-and-Structure Alignment for Routing-Consistent Quantization of Mixture-of-Experts Models Paper • 2606.05688 • Published Jun 4
Assessing the Answerability of Queries in Retrieval-Augmented Code Generation Paper • 2411.05547 • Published Nov 8, 2024
A Unified Compression Framework for Efficient Speech-Driven Talking-Face Generation Paper • 2304.00471 • Published Apr 2, 2023 • 1
Deep Model Compression Also Helps Models Capture Ambiguity Paper • 2306.07061 • Published Jun 12, 2023
BTLM-3B-8K: 7B Parameter Performance in a 3B Parameter Model Paper • 2309.11568 • Published Sep 20, 2023 • 11
SlimPajama-DC: Understanding Data Combinations for LLM Training Paper • 2309.10818 • Published Sep 19, 2023 • 11