- Rating
- 1
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
The paper proposes hierarchical diagnosis for LLM-based Triton kernel optimization. It progresses from profiling to IR and compiler analysis to guide source rewrites, with evaluation on 37 Ascend NPU kernels.
Critical strengths
The real-hardware speedups are useful, and the case studies connect compiler behavior to concrete source changes.
Critical weaknesses
-
There is no comparison with profiling-only or always-deep analysis. The benefit of the hierarchy itself is therefore unclear.
-
Token use and total optimization time are not reported. It is unclear whether using deep analysis only when needed actually saves resources.
-
The agent decides when to use deeper analysis, but the paper does not test how well these decisions work.
-
The main speedup is measured on the same five input cases used during optimization. It is unclear whether these gains hold on new inputs.
Questions for authors
-
With the same model, toolchain, and budget, does the hierarchy perform better than profiling-only and always-deep analysis?
-
How many tokens and how much time does each method need to reach the same speedup?
-
Does the decision rule work better than simply using deeper analysis after a few failed attempts?
-
Do the gains hold on new input shapes and sizes? How often is the optimized kernel slower than the initial kernel on these inputs?