7 papers
Cross-Model KV Cache Transfer in LLM Families: A Closed-Form Linear Mapping for Prefill Reuse
Taekyung Heo, Rasoul Shafipour, Ritchie Zhao +6
Production deployments often swap between different-sized models in a family for cost-quality cascading, mid-conversation switching, and routing, and each swap forces the receiver…
LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts
Venmugil Elango, Nidhi Bhatia, Roger Waleffe +13
Mixture of Experts (MoEs) have become a central component of many state-of-the-art open-source and proprietary large language models. Despite their widespread adoption, it remains…
Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens
Yanpeng Yu, Haiyue Ma, Krish Agarwal +10
Expert Parallelism (EP) permits Mixture of Experts (MoE) models to scale beyond a single GPU. To address load imbalance across GPUs in EP, existing approaches aim to balance the nu…
Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM Decoding
Nidhi Bhatia, Ankit More, Ritika Borkar +7
As LLMs scale to multi-million-token KV histories, real-time autoregressive decoding under tight Token-to-Token Latency (TTL) constraints faces growing pressure. Two core bottlenec…
Beyond the Buzz: A Pragmatic Take on Inference Disaggregation
Tiyasa Mitra, Ritika Borkar, Nidhi Bhatia +10
As inference scales to multi-node deployments, disaggregation - splitting inference into distinct phases - offers a promising path to improving the throughput-interactivity Pareto…
ResMoE: Space-efficient Compression of Mixture of Experts LLMs via Residual Restoration
Mengting Ai, Tianxin Wei, Yifan Chen +7
Mixture-of-Experts (MoE) Transformer, the backbone architecture of multiple phenomenal language models, leverages sparsity by activating only a fraction of model parameters for eac…