Ask, Don't Equip: Question-Driven Coding Agents for Deep Kernel Optimization
cs.LG
Submitted: 2026-11-02
Updated: 2026-11-02
Code: https://github.com/NVIDIA/cutlass
Terminology
Sources
- From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents
- Kevin: Multi-Turn RL for Generating CUDA Kernels
- K-Search: LLM Kernel Generation via Co-Evolving Intrinsic World Model
- AVO: Agentic Variation Operators for Autonomous Evolutionary Search
- CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
- STARK: Strategic Team of Agents for Refining Kernels
- Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
- EvoEngineer: Mastering Automated CUDA Kernel Code Evolution with Large Language Models
- Autocomp: A Powerful and Portable Code Optimizer for Tensor Accelerators
- Another Turn, Better Output? A Turn-Wise Analysis of Iterative LLM Prompting
- ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
- Towards Robust Agentic CUDA Kernel Benchmarking, Verification, and Optimization
- TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization
- CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning
- KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta
- SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
- Dr. Kernel: Reinforcement Learning Done Right for Triton Kernel Generations
- Agentic Auto-Scheduling: An Experimental Study of LLM-Guided Loop Optimization
- AlphaEvolve: A coding agent for scientific and algorithmic discovery
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis