rsadhukhan ranonrkm

• Carnegie Mellon University

• Pittsburgh, PA
• ranonrkm.github.io

Pinned Loading

1. Forked from microsoft/DiskANN

   Scalable graph based indices for approximate nearest neighbor search

   C++ 1

3. [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding

   Python 145 12

4. Kinetics: Rethinking Test-Time Scaling Laws

   Python 86 3

5. Speculative decoding for high-throughput long-context inference

   JavaScript

6. MagicDec: Breaking the Latency-Throughput Tradeoff for Long Contexts with Speculative Decoding

   JavaScript