Ananda Theertha Suresh

CafeQ: Calibration-free Quantization via Learned Transformations and Adaptive Rounding

Private Set Union with Multiple Contributions

InfAlign: Inference-aware language model alignment

Block Verification Accelerates Speculative Decoding

Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe

Theoretical guarantees on the best-of-n alignment policy

Asymmetric Embedding Models for Hierarchical Retrieval: Provable Constructions and a Pretrain-Finetune Recipe

Concept forgetting via label annealing