Research map / Architectures and efficiency / RNNs and normalization
AdaTune: Adaptive Tensor Program Compilation Made Efficient
Menghao Li, Minjia Zhang, Chi Wang, Mingqin Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add adaptive evaluator (CV-based early termination of hardware measurement) (ablation) | AdaTune · ResNet-18 on CPU, per-iteration cost | hardware measurement time: 2.5× reduction | Improved |
| add adaptive evaluator on top of RFEI + CSA (ablation) | AdaTune · 4 tuning workloads | wall-clock search time | Improved |
| replace XGBoost with random forest + expected improvement surrogate (ablation) | AutoTVM baseline · search effectiveness per iteration | iterations to find best plan | Mixed |
| replace fixed epsilon-greedy with contextual simulated annealing (ablation) | AdaTune · search effectiveness per iteration | search effectiveness | Mixed |
| use AdaTune instead of AutoTVM under same time budget | Batched GEMM (GPU) · tuning | best GFLOPS: up to 115% higher GFLOPS | Improved |
| use AdaTune instead of AutoTVM for end-to-end optimization | ResNet-18 on GPU · optimization time | hours: 22.6 hours -> 9.6 hours | Improved |
| use AdaTune instead of AutoTVM for final inference time | ResNet-18 · inference speed | inference speed: up to 4.6% faster | Improved |
Closest papers by meaning
- DynaTune: Dynamic Tensor Program Optimization in Deep Neural Network Compilation (ICLR 2021 · similarity 0.86)
- Learning to Optimize Tensor Programs (NeurIPS 2018 · similarity 0.82)
- Einsum Benchmark: Enabling the Development of Next-Generation Tensor Execution Engines (NeurIPS 2024 · similarity 0.79)
- COGNATE: Acceleration of Sparse Tensor Programs on Emerging Hardware using Transfer Learning (ICML 2025 · similarity 0.78)
- Navigating Scaling Laws: Compute Optimality in Adaptive Model Training (ICML 2024 · similarity 0.77)
- QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Code (NeurIPS 2025 · similarity 0.77)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
