MetaKernelBench: Measuring GPU Kernel Knowledge Transfer Beyond Code
cs.LG
Submitted: 2026-10-04
Updated: 2026-10-04
Code: https://github.com/harbor-framework/harbor
Project page: https://yige24.github.io/MetaKernelBench
Sources
- DICE: Diffusion Large Language Models Excel at Generating CUDA Kernels
- AVO: Agentic Variation Operators for Autonomous Evolutionary Search
- CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning
- daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization
- Towards Robust Agentic CUDA Kernel Benchmarking, Verification, and Optimization
- SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
- KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization
- Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks
- Astra: A Multi-Agent System for GPU Kernel Performance Optimization
- AgentKernelArena: Generalization-Aware Benchmarking of GPU Kernel Optimization Agents
- KLineage: Recovering the Missing When of Kernel Optimization by Deoptimizing Experts
- KernelBench-Verified: Do LLM-Generated Kernels Actually Beat PyTorch?
- CudaForge: An Agent Framework with Hardware Feedback for CUDA Kernel Optimization
- MobileKernelBench: Can LLMs Write Efficient Kernels for Mobile Devices?
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks