Publications

KVFlow: Efficient Prefix Caching for Accelerating LLM-Based Multi-Agent Workflows Permalink

Published in The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025

Scaling Multi-Agent Simulation

Published in Submitted to The 9th Annual Conference on Machine Learning and Systems, 2025

This paper is under review to MLSys2026

Unlocking Long-Term Dependencies in Spiking Neural Networks with a Recurrent LIF Memory Module

Published in Submitted to The Fourteenth International Conference on Learning Representations, 2025

This paper is under review to ICLR2026