Research map / Architectures and efficiency / RNNs and normalization
Ithemal: Accurate, Portable and Fast Basic Block Throughput Estimation using Deep Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace hand-written analytical estimation with learned Ithemal | Basic-block throughput estimation on Ivy Bridge · Ivy Bridge | average error: 0.181 -> 0.089 | Helped |
| Replace hand-written analytical estimation with learned Ithemal | Basic-block throughput estimation on Haswell · Haswell | average error: 0.209 -> 0.089 | Helped |
| Replace hand-written analytical estimation with learned Ithemal | Basic-block throughput estimation on Skylake · Skylake | average error: 0.167 -> 0.079 | Helped |
| Use a dependency-graph prediction layer instead of sequential instruction processing (ablation) | Ithemal hierarchical LSTM · Basic-block throughput prediction | validation loss | Hurt |
| Process all tokens sequentially without a separate instruction-level hierarchy (ablation) | Ithemal hierarchical LSTM · Basic-block throughput prediction | validation loss | Hurt |
Closest papers by meaning
- Bridging Layout and RTL: Knowledge Distillation based Timing Prediction (ICML 2025 · similarity 0.85)
- ArchPower: Dataset for Architecture-Level Power Modeling of Modern CPU Design (NeurIPS 2025 · similarity 0.82)
- Learning to Add, Multiply, and Execute Algorithmic Instructions Exactly with Neural Networks (NeurIPS 2025 · similarity 0.82)
- NoMAD-Attention: Efficient LLM Inference on CPUs Through Multiply-add-free Attention (NeurIPS 2024 · similarity 0.81)
- CircuitFusion: Multimodal Circuit Representation Learning for Agile Chip Design (ICLR 2025 · similarity 0.81)
- PCEval: A Benchmark for Evaluating Physical Computing Capabilities of Large Language Models (NeurIPS 2026 · similarity 0.81)
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.