DrawScore: evaluating generated CAD models via engineering-drawing projections
Haorui Chen · hf · 2026-11-19
DrawScore is a new metric for automatic CAD generation that evaluates models through engineering-drawing projections with a saturating penalty along curves.
- It exposes small local errors that volume- or surface-based metrics miss;
- It can also serve as a reinforcement learning reward or a test-time verifier.
More from Research
- NBER's fereg beats reghdfe by up to 204x; author built it with Codex and Claude — daveholtz · 2026-11-19
- Self-trained pixel RL agent beats Elden Ring boss Margit in simulated gym — Sl33py_4est · 2026-11-19
- CriticNPU: Mechanistic critic halves simulator calls for sample-efficient NPU design-space exploration — Yicheng He · 2026-11-19
- VAE Rhythm: Detecting AI Videos via Periodic 3D Causal VAE Decoding Traces — showlab · 2026-11-19
- Tideway: phase-aware unified memory pool speeds MoE inference on consumer devices by ~2x — DukeUniversity · 2026-11-19
- Tencent's ECA adds evidence-based credit assignment to search-augmented LLM RL, +4.8pts on QA — tencent · 2026-11-19