paper-with-me

홈 › Papers

CUCo: An Agentic Framework for Compute and Communication Co-design

2026-03-02 · Yoga Sri Varshan Varadharajan, Bodun Hu, Saurabh Agarwal, Aditya Akella arxiv

Computation and communication in distributed LLM training and inference are traditionally optimized in isolation; expert-crafted systems such as DeepEP, FLUX, and TokenWeave show the potential of co-design but require deep systems expertise and hardware-specific tuning; CUCo is an agentic framework that automates compute-communication co-design of CUDA kernels by combining a structured design-space formalization with a correctness-first fast-path agent for reliable baselines and an evolution-driven slow-path agent for high-performance strategies, achieving up to 1.57x speedup across four multi-GPU workloads and discovering a two-stream overlap strategy on a DeepSeek-V3 MoE layer that hides dispatch behind local compute at an LLM inference cost under $10 per workload.

📄 PDF Abstract BibTeX arXiv:2603.02376

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MCUCoder: Adaptive Bitrate Learned Video Compression for IoT Devices

2024-11-29 · Ali Hojjat, Janek Haberer, Olaf Landsiedel

The rapid growth of camera-based IoT devices demands the need for efficient video compression, particularly for edge applications where devices face hardware constraints, often with only 1 or 2 MB of RAM and unstable int…

MS-SSIMSSIMVideo Compression

Agentic Design Patterns: A System-Theoretic Framework

2026-01-27 · Minh-Dung Dao, Quy Minh Le, Hoang Thanh Lam, Duc-Trong Le 외 arxiv

With the development of foundation model (FM), agentic AI systems are getting more attention, yet their inherent issues like hallucination and poor reasoning, coupled with the frequent ad-hoc nature of system design, lea…

From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communications

2025-05-28 · Feibo Jiang, Cunhua Pan, Li Dong, Kezhi Wang 외

With the advent of 6G communications, intelligent communication systems face multiple challenges, including constrained perception and response capabilities, limited scalability, and low adaptability in dynamic environme…

Intelligent Communication

Enhancing Large Language Models for Secure Code Generation: A Dataset-driven Study on Vulnerability Mitigation

2023-10-25 · Jiexin Wang, Liuwen Cao, Xitong Luo, Zhiping Zhou 외

Large language models (LLMs) have brought significant advancements to code generation, benefiting both novice and experienced developers. However, their training using unsanitized data from open-source repositories, like…

Code GenerationCode Repair

Agency and Morality as part of Text Entry AI Assistant Personas

2024-12-11 · Andreas Komninos

This paper discusses the need to move away from an instrumental view of text composition AI assistants under direct control of the user, towards a more agentic approach that is based on a value rationale. Based on an ana…