activity
20202026
most citedPerformance-Aligned LLMs for Generating Fast Code

1 citations · 3 across the 20 of their papers we have counts for

collaborators
Showing cs.LGShow all

13 papers · 1 filter

cs.LG2026

Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training

Cunyang Wei, Siddharth Singh, Aishwarya Sarkar +7

Graph neural networks (GNNs) are widely used for learning on graph datasets derived from various real-world scenarios. Learning from extremely large graphs requires distributed tra…

cs.LG2026

Speculating Experts Accelerates Inference for Mixture-of-Experts

Vivan Madan, Prajwal Singhania, Abhinav Bhatele +2

Mixture-of-Experts (MoE) models have gained popularity as a means of scaling the capacity of large language models (LLMs) while maintaining sparse activations and reduced per-token…

cs.LG2025

Power Law Guided Dynamic Sifting for Efficient Attention

Nirav Koley, Prajwal Singhania, Abhinav Bhatele

Efficient inference on GPUs using large language models remains challenging due to memory bandwidth limitations, particularly during data transfers between High Bandwidth Memory (H…

cs.LG2025

Plexus: Taming Billion-edge Graphs with 3D Parallel Full-graph GNN Training

Aditya K. Ranjan, Siddharth Singh, Cunyang Wei +1

Graph neural networks (GNNs) leverage the connectivity and structure of real-world graphs to learn intricate properties and relationships between nodes. Many real-world graphs exce…

cs.LG2025

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach

Jonas Geiping, Sean McLeish, Neel Jain +6

We study a novel language model architecture that is capable of scaling test-time computation by implicitly reasoning in latent space. Our model works by iterating a recurrent bloc…

cs.LG2025

Democratizing AI: Open-source Scalable LLM Training on GPU-based Supercomputers

Siddharth Singh, Prajwal Singhania, Aditya Ranjan +9

Training and fine-tuning large language models (LLMs) with hundreds of billions to trillions of parameters requires tens of thousands of GPUs, and a highly scalable software stack.…