Research map / Large language models / Agents, code and math
Instead of Rewriting Foreign Code for Machine Learning, Automatically Synthesize Fast Gradients
William Moses, Valentin Churavy
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 |
|---|---|---|---|
| run optimization before AD (AD on optimized IR) instead of AD before optimization (ablation) | Enzyme · ML benchmark suite (ADBench, Euler, RK4, FFT, Bruss) | geometric mean speedup: 4.5x | Improved |
Closest papers by meaning
- Automatic differentiation in ML: Where we are and where we should be going (NeurIPS 2018 · similarity 0.87)
- IR-OptSet: An Optimization-Sensitive Dataset for Advancing LLM-Based IR Optimizer (NeurIPS 2025 · similarity 0.86)
- REASONING COMPILER: LLM-Guided Optimizations for Efficient Model Serving (NeurIPS 2025 · similarity 0.84)
- QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Code (NeurIPS 2025 · similarity 0.83)
- LPGD: A General Framework for Backpropagation through Embedded Optimization Layers (ICML 2024 · similarity 0.83)
- HYSYNTH: Context-Free LLM Approximation for Guiding Program Synthesis (NeurIPS 2024 · similarity 0.82)
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.
