Research map / Large language models / Agents, code and math
DeepDSL: A Compilation-based Domain-Specific Language for Deep Learning
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 |
|---|---|---|---|
| Use DeepDSL instead of Caffe | Alexnet · Alexnet | runtime | Helped |
| Use DeepDSL instead of Caffe | Overfeat · Overfeat | runtime | Helped |
| Use DeepDSL instead of Caffe | GoogLeNet · GoogLeNet | runtime | Helped |
| Use DeepDSL instead of Caffe | Vgg · Vgg | runtime | Hurt |
| Use DeepDSL instead of Caffe | ResNet-50 · ResNet-50 | runtime | Hurt |
| Use DeepDSL instead of TensorFlow | DeepDSL · Alexnet, GoogLeNet, ResNet-50 | runtime | Helped |
| Use DeepDSL instead of TensorFlow | DeepDSL · Overfeat, Vgg | runtime | Hurt |
| Use DeepDSL instead of Caffe | DeepDSL · Alexnet, GoogLeNet, ResNet-50 | GPU memory | Helped |
| Use DeepDSL instead of Caffe | DeepDSL · Overfeat, Vgg | GPU memory | Hurt |
| Use DeepDSL instead of TensorFlow | DeepDSL · Alexnet, Overfeat, GoogLeNet, ResNet-50 | GPU memory | Helped |
| Use DeepDSL instead of TensorFlow | DeepDSL · Vgg | GPU memory | Hurt |
| Switch from runtime-efficient mode to memory-efficient mode, deallocating tensor objects promptly (ablation) | DeepDSL · Alexnet, Overfeat, GoogLeNet, Vgg, ResNet-50 training | GPU memory: 10 to 30\% less memory | Helped |
Closest papers by meaning
- PyTorch: An Imperative Style, High-Performance Deep Learning Library (NeurIPS 2019 · similarity 0.85)
- VL2Lite: Task-Specific Knowledge Distillation from Large Vision-Language Models to Lightweight Networks (CVPR 2025 · similarity 0.82)
- TorchTitan: One-stop PyTorch native solution for production ready LLM pretraining (ICLR 2025 · similarity 0.82)
- VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration (ICLR 2025 · similarity 0.82)
- Scalable Neural Network Kernels (ICLR 2024 · similarity 0.82)
- DMax: Aggressive Parallel Decoding for dLLMs (NeurIPS 2026 · 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.