2 citations · 2 across the 3 of their papers we have counts for
3 papers
Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding
Hayate Iso, Tiyasa Mitra, Sudipta Mondal +15
RL post-training of frontier language models is increasingly bottlenecked by autoregressive rollout generation, making rollout acceleration a central systems challenge. Many existi…
Duality between imperfect resources and measurements for propagating entanglement in networks
Sudipta Mondal, Pritam Halder, Aditi Sen De
We propose a measurement-based entanglement propagation strategy for networks in which all nodes except two are initially occupied by a suitably chosen single-qubit system and the…
GNNIE: GNN Inference Engine with Load-balancing and Graph-Specific Caching
Sudipta Mondal, Susmita Dey Manasi, Kishor Kunal +2
Graph neural networks (GNN) analysis engines are vital for real-world problems that use large graph models. Challenges for a GNN hardware platform include the ability to (a) host a…