Compiler-Grounded Hierarchical Diagnosis for LLM-Based Triton Kernel Optimization

arXiv:2607.23089v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have enabled automated kernel generation and optimization, but most existing approaches rely on surface signals such as compilation feedback and profiling metrics. These signals reveal that a kernel is slow, but not why the backend compiler fails to realize a profitable optimization, especially on emerging accelerators such as NPUs. We therefore formulate kernel optimization as a progressive cross-layer diagnosis problem that links runtime symptoms to IR structure and compiler behavior before rewriting source. Based on this insight, we present our system, a compiler-grounded and hierarchical optimization framework for Triton kernels. the system escalates from lightweight pattern triage and profiling diagnosis to IR attribution and compiler-grounded analysis only when deeper evidence is needed, then proposes evidence-backed source-level rewrites. We implement the system on Triton for Ascend NPUs and evaluate it on 37 successfully converted entries from a standardized NPUKernelBench-derived Ascend 950 benchmark. Across these entries, the system attains a geometric-mean speedup of 4.35imes and a median speedup of 2.73imes from the initial to optimized Triton kernel; 22/37 exceed 2imes and 13/37 exceed 5imes. The complete distribution ranges from near-baseline entries to large wins, motivating transparent reporting of the current system's scope and limitations.

Related

• Exploring LLM-based Verilog Code Generation with Data-Efficient Fine-Tuning and Testbench Automation
• Hierarchical Experimentalist Agents
• From Natural Language to Verified Code: Toward AI Assisted Problem-to-Code Generation with Dafny-Based Formal Verification

Source: arXiv cs.AI | 2026-07-28