PPL
Marco Cuturi
Papers (5)
Locking Pretrained Weights via Deep Low-Rank Residual DistillationDynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic MixturesNectar: Neural Estimation of Cached-Token Attention via RegressionScaling Categorical Flow MapsHyperTransport: Amortized Conditioning of T2I Generative Models