EditHero: First Long-Horizon Part-Level 3D Editing Benchmark Shows LLM Agents Win
Ruihan Yu · hf · 2026-10-05
Researchers introduce EditHero, the first benchmark for long-horizon, part-level 3D editing with natural-language instructions and target images for both geometry and texture, using a deterministic assembly engine and hand-reviewed edit sequences.
- It contrasts non-agentic methods (regenerating from a learned 3D representation, inferring what to keep) against LLM/VLM agents that edit via code inspecting the mesh and rewriting only required parts.
- Non-agentic methods often miss requested changes and disturb regions that should stay fixed; LLMs follow instructions more closely and all preserve unedited parts better, but each edit takes minutes.
- The assembly engine and edit sequences will be released for research on reliable iterative 3D editing.
More from Multimodal
- User has Claude Fable 5.5 make a Rocket League edit — imjustnewatai · 2026-10-05
- One Claude chat turns a 50-second poem into an 18-scene AI film — aziz4ai · 2026-10-05
- OctLLM Uses Sparse Octree Tokens to Hit 3D SOTA While Preserving Language Ability — Ran Dan · 2026-10-05
- DEFINE Decouples Accent from Voice in Zero-Shot TTS, Matching Two-Model Cascades with One Model — amaai-lab · 2026-10-05
- Diptych: AI-Assisted Reference Listening Lets Musicians Define What to Compare in Music Production — amaai-lab · 2026-10-05
- FrameMorrow Selects History Frames by Predicted Future Needs, Boosting 11 Long-Video Generators — NationalUniversityofSingapore · 2026-10-05