Prateek Jain

Spark Transformer: Reactivating Sparsity in Transformer FFN and Attention

Masked Generative Nested Transformers with Decode Time Scaling

Spark Transformer: How Many FLOPs is a Token Worth?

Does Safety Training of LLMs Generalize to Semantically Related Natural Prompts?

Matryoshka Quantization

Time-Reversal Provides Unsupervised Feedback to LLMs

MatFormer: Nested Transformer for Elastic Inference

Dual-Encoders for Extreme Multi-label Classification

LLM Augmented LLMs: Expanding Capabilities through Composition

EHI: End-to-end learning of Hierarchical Index for Efficient Dense Retrieval

MatFormer: Nested Transformer for Elastic Inference

Mixture of Nested Experts: Adaptive Processing of Visual Tokens

End-to-End Neural Network Compression via $\frac{\ell_1}{\ell_2}$ Regularized Latency Surrogates