English
Related papers

Related papers: NanoCP: Request-Level Dynamic Context Parallelism …

200 papers

Standard Mixture-of-Experts (MoE) transformers route tokens to expert subnetworks within each layer, but the layer structure itself remains monolithic. We introduce Mixture of Layers (MoL), which replaces full-width transformer blocks…

Machine Learning · Computer Science 2026-05-12 Ivan Ternovtsii , Yurii Bilak

Sparsely Mixture of Experts (MoE) has received great interest due to its promising scaling capability with affordable computational overhead. MoE converts dense layers into sparse experts, and utilizes a gated routing network to make…

Computation and Language · Computer Science 2022-07-20 Yuan Xie , Shaohan Huang , Tianyu Chen , Furu Wei

Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive language generation due to their potential for parallel decoding and global refinement of the entire sequence. To unlock this potential, DLM…

Machine Learning · Computer Science 2026-04-20 Xiang Xia , Wuyang Zhang , Jiazheng Liu , Cheng Yan , Yanyong Zhang

Transformer-based models have emerged as a leading architecture for natural language processing, natural language generation, and image generation tasks. A fundamental element of the transformer architecture is self-attention, which allows…

Machine Learning · Computer Science 2025-07-01 Venmugil Elango

The Model Context Protocol (MCP) (MCP Community, 2025) has emerged as a widely used framework for enabling LLM-based agents to communicate with external tools and services. The original MCP implementation (Anthropic, 2024) relies on a Large…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-26 Meenakshi Amulya Jayanti , X. Y. Han

Mixture of Experts (MoE) has become a mainstream architecture for building Large Language Models (LLMs) by reducing per-token computation while enabling model scaling. It can be viewed as partitioning a large Feed-Forward Network (FFN) at…

Machine Learning · Computer Science 2025-08-27 Weilin Cai , Le Qin , Shwai He , Junwei Cui , Ang Li , Jiayi Huang

Large Language Model (LLM) inference services demand exceptionally high availability and low latency, yet multi-GPU Tensor Parallelism (TP) makes them vulnerable to single-GPU failures. We present AnchorTP, a state-preserving elastic TP…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-18 Wendong Xu , Chujie Chen , He Xiao , Kuan Li , Jing Xiong , Chen Zhang , Wenyong Zhou , Chaofan Tao , Yang Bai , Bei Yu , Ngai Wong

Mixture-of-Experts (MoE) architectures achieve parameter efficiency through conditional computation, yet contemporary designs suffer from two fundamental limitations: structural parameter isolation that causes catastrophic forgetting, and…

Machine Learning · Computer Science 2026-01-21 Yuxing Gan , Ziyu Lei

Retrieval-Augmented Generation (RAG) systems suffer from severe time-to-first-token (TTFT) bottlenecks due to long input sequences. Existing KV cache reuse methods face a fundamental trade-off: prefix caching requires identical prefixes…

Machine Learning · Computer Science 2026-05-22 Bin Yang , Qiuyu Leng , Jun Zeng , Zhenhua Wu

Recent papers in neural machine translation have proposed the strict use of attention mechanisms over previous standards such as recurrent and convolutional neural networks (RNNs and CNNs). We propose that by running traditionally stacked…

Computation and Language · Computer Science 2018-10-31 Julian Richard Medina , Jugal Kalita

In LLM serving, reusing the KV cache of prompts across requests is critical for reducing TTFT and serving costs. Cache-affinity scheduling, which co-locates requests with the same prompt prefix to maximize KV cache reuse, often conflicts…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-09 Ying Yuan , Pengfei Zuo , Bo Wang , Zhangyu Chen , Zhipeng Tan , Zhou Yu

Sparse Mixture-of-Experts (MoE) models scale capacity by routing each token to a small subset of experts. However, their routers exhibit a fundamental trade-off: strong load balancing can suppress expert specialization, while aggressive…

Machine Learning · Computer Science 2026-05-12 Gleb Molodtsov , Alexander Miasnikov , Aleksandr Beznosikov

Extending large language models (LLMs) to process longer inputs is crucial for a wide range of applications. However, the substantial computational cost of transformers and limited generalization of positional encoding restrict the size of…

Computation and Language · Computer Science 2025-06-11 Howard Yen , Tianyu Gao , Danqi Chen

Attention mechanisms have become a core component of deep learning models, with Channel Attention and Spatial Attention being the two most representative architectures. Current research on their fusion strategies primarily bifurcates into…

Computer Vision and Pattern Recognition · Computer Science 2026-01-26 Zhongming Liu , Bingbing Jiang

Efficiently processing long sequences with Transformer models usually requires splitting the computations across accelerators via context parallelism. The dominant approaches in this family of methods, such as Ring Attention or DeepSpeed…

Machine Learning · Computer Science 2026-02-25 Ravi Ghadia , Maksim Abraham , Sergei Vorobyov , Max Ryabinin

This paper introduces a novel network topology that seamlessly integrates dynamic inference cost with a top-down attention mechanism, addressing two significant gaps in traditional deep learning models. Drawing inspiration from human…

Computer Vision and Pattern Recognition · Computer Science 2024-11-21 André Peter Kelm , Niels Hannemann , Bruno Heberle , Lucas Schmidt , Tim Rolff , Christian Wilms , Ehsan Yaghoubi , Simone Frintrop

Mixture-of-Experts layers achieve compute efficiency through weight sparsity: each token activates only a subset of experts. Data sparsity, where each expert processes only a subset of tokens, offers a complementary axis. Expert-choice…

Machine Learning · Computer Science 2026-01-23 Maciej Kilian , Oleg Mkrtchyan , Luke Zettlemoyer , Akshat Shrivastava , Armen Aghajanyan

Mixture-of-Experts (MoE) architectures enable efficient scaling of large language models by activating only a subset of parameters per input. However, existing MoE models suffer from two critical limitations: (1) inefficient token-to-expert…

Computation and Language · Computer Science 2025-10-10 Jing Li , Zhijie Sun , Dachao Lin , Xuan He , Binfan Zheng , Yi Lin , Rongqian Zhao , Xin Chen

Long-context large language models (LLMs) face constraints due to the quadratic complexity of the self-attention mechanism. The mainstream sequence parallelism (SP) method, Ring Attention, attempts to solve this by distributing the query…

Machine Learning · Computer Science 2025-10-10 Yida Wang , Ke Hong , Xiuhong Li , Yuanchao Xu , Wenxun Wang , Guohao Dai , Yu Wang

Sparse Mixture of Experts (MoE) models offer a scalable and efficient architecture for training large neural networks by activating only a subset of parameters ("experts") for each input. A learned router computes a distribution over these…

Machine Learning · Computer Science 2025-10-14 Nabil Omi , Siddhartha Sen , Ali Farhadi