Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference Paper • 2412.13663 • Published Dec 18, 2024 • 125 • 10
Reducing the Footprint of Multi-Vector Retrieval with Minimal Performance Impact via Token Pooling Paper • 2409.14683 • Published Sep 23, 2024 • 11