# STARK: Strategic Team of Agents for Refining Kernels

arXiv: https://arxiv.org/abs/2510.16996 · HTML: https://mlsanity.com/paper/2510.16996/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 13 of its 28 references that are in the map.
- First posted on arXiv 19 Oct 2025.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 13 of 13)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [SWE-smith: Scaling Data for Software Engineering Agents](https://mlsanity.com/paper/2504.21798/) (2025)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [Ansor: Generating High-Performance Tensor Programs for Deep Learning](https://mlsanity.com/paper/2006.06762/) (2020)
- [Can ChatGPT replace StackOverflow? A Study on Robustness and Reliability of Large Language Model Code Generation](https://mlsanity.com/paper/2308.10335/) (2023)
- [MLGO: a Machine Learning Guided Compiler Optimizations Framework](https://mlsanity.com/paper/2101.04808/) (2021)
- [CLTune: A Generic Auto-Tuner for OpenCL Kernels](https://mlsanity.com/paper/1703.06503/) (2017)
- [BaCO: A Fast and Portable Bayesian Compiler Optimization Framework](https://mlsanity.com/paper/2212.11142/) (2022)
- [Kernel Launcher: C++ Library for Optimal-Performance Portable CUDA Applications](https://mlsanity.com/paper/2303.12374/) (2023)
- [Offline Imitation Learning from Multiple Baselines with Applications to Compiler Optimization](https://mlsanity.com/paper/2403.19462/) (2024)

## Built on by (most cited first, 25 of 25)

- [CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation](https://mlsanity.com/paper/2602.24286/) (2026)
- [KernelFoundry: Hardware-aware evolutionary GPU kernel optimization](https://mlsanity.com/paper/2603.12440/) (2026)
- [KernelBenchX: A Comprehensive Benchmark for Evaluating LLM-Generated GPU Kernels](https://mlsanity.com/paper/2605.04956/) (2026)
- [Towards Automated Kernel Generation in the Era of LLMs](https://mlsanity.com/paper/2601.15727/) (2026)
- [CUDABench: Benchmarking LLMs for Text-to-CUDA Generation](https://mlsanity.com/paper/2603.02236/) (2026)
- [The Correctness Illusion in LLM-Generated GPU Kernels](https://mlsanity.com/paper/2606.20128/) (2026)
- [Towards Cold-Start Drafting and Continual Refining: A Value-Driven Memory Approach with Application to NPU Kernel Synthesis](https://mlsanity.com/paper/2603.10846/) (2026)
- [AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization](https://mlsanity.com/paper/2603.23566/) (2026)
- [PerfAgent: Profiler-Guided Iterative Refinement for Repository-Level Code Optimization](https://mlsanity.com/paper/2607.19653/) (2026)
- [CUCo: An Agentic Framework for Compute and Communication Co-design](https://mlsanity.com/paper/2603.02376/) (2026)
- [Xe-Forge: Multi-Stage LLM-Powered Kernel Optimization for Intel GPU](https://mlsanity.com/paper/2605.26118/) (2026)
- [Rudder: Steering Prefetching in Distributed GNN Training using LLM Agents](https://mlsanity.com/paper/2602.23556/) (2026)
- [PassNet: Scaling Large Language Models for Graph Compiler Pass Generation](https://mlsanity.com/paper/2605.29357/) (2026)
- [From Tokens to Regions: CUDA-Sensitive Instruction Tuning for GPU Kernel Generation](https://mlsanity.com/paper/2606.16231/) (2026)
- [Hawk: Harnessing Hardware-Aware Knowledge for High-Performance NPU Kernel Generation](https://mlsanity.com/paper/2607.01590/) (2026)
- [SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD](https://mlsanity.com/paper/2607.20145/) (2026)
- [The New Compiler Stack: A Survey on the Synergy of LLMs and Compilers](https://mlsanity.com/paper/2601.02045/) (2026)
- [Step-TP: A Grounded, Step-Level Dataset with Chain-of-Thought Reasoning for LLM-Guided Tensor Program Optimization](https://mlsanity.com/paper/2605.25954/) (2026)
- [Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation](https://mlsanity.com/paper/2605.26720/) (2026)
- [Arbor: Tree Search as a Cognition Layer for Autonomous Agents](https://mlsanity.com/paper/2606.12563/) (2026)
- [Test-Input Generation for Tensor Programs: What Actually Finds Kernel Bugs](https://mlsanity.com/paper/2606.27396/) (2026)
- [KernelBrain: Coarse-to-Fine, Budget-Aware Search for Agentic GPU Kernel Optimization](https://mlsanity.com/paper/2608.02611/) (2026)
- [CUDA-Harness: Harnessing Agentic CUDA Kernel Generation and Optimization from Natural Language](https://mlsanity.com/paper/2609.00058/) (2026)
- [KernelZero: Co-Evolving Proposer and Coder for Continuously Improved GPU Kernel Generation](https://mlsanity.com/paper/2609.33074/) (2026)
- [GPUPhysBench: Benchmarking Coding Agents for Correct and Efficient GPU Physics Simulation](https://mlsanity.com/paper/2609.35639/) (2026)

## Abstract (arXiv)

The efficiency of GPU kernels is central to the progress of modern AI, yet optimizing them remains a difficult and labor-intensive task due to complex interactions between memory hierarchies, thread scheduling, and hardware-specific characteristics. While recent advances in large language models (LLMs) provide new opportunities for automated code generation, existing approaches largely treat LLMs as single-shot generators or naive refinement tools, limiting their effectiveness in navigating the irregular kernel optimization landscape. We introduce an LLM agentic framework for GPU kernel optimization that systematically explores the design space through multi-agent collaboration, grounded instruction, dynamic context management, and strategic search. This framework mimics the workflow of expert engineers, enabling LLMs to reason about hardware trade-offs, incorporate profiling feedback, and refine kernels iteratively. We evaluate our approach on KernelBench, a benchmark for LLM-based kernel optimization, and demonstrate substantial improvements over baseline agents: our system produces correct solutions where baselines often fail, and achieves kernels with up to 16x faster runtime performance. These results highlight the potential of agentic LLM frameworks to advance fully automated, scalable GPU kernel optimization.