Research map / Point clouds / Segmentation and detection
Language-Driven 6-DoF Grasp Detection Using Negative Prompt Guidance
Toan Nguyen, Minh Nhat Nhat Vu, Baoru Huang, An Dinh Vuong, Quan Vuong, Ngan Le, Thieu Vo, Anh Nguyen
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 |
|---|---|---|---|
| add negative prompt guidance (learned negative prompt embedding) to the diffusion sampler (ablation) | LGrasp6D · Grasp-Anything-6D | CR: 0.5459 -> 0.6694 | Improved |
| add negative prompt guidance (ablation) | LGrasp6D · Grasp-Anything-6D | EMD: 0.6262 -> 0.4013 | Improved |
| add negative prompt guidance (ablation) | LGrasp6D · Grasp-Anything-6D | CFR: 0.7336 -> 0.7706 | Improved |
| add negative prompt guidance (ablation) | LGrasp6D · Grasp-Anything-6D | inference time (s): 1.4328 -> 1.4832 | No clear effect |
| add negative prompt guidance (ablation) | LGrasp6D · Contact-GraspNet (cross-dataset) | CR: 0.5208 -> 0.6420 | Improved |
| add negative prompt guidance (ablation) | LGrasp6D · Real robot, cluttered | success rate: 0.36 -> 0.42 | Improved |
| reduce DDIM sampling steps from 200 to 10 (ablation) | LGrasp6D · Grasp-Anything-6D | CR: 0.6694 -> 0.5611 | Hurt |
| reduce DDIM sampling steps from 200 to 50 (ablation) | LGrasp6D · Grasp-Anything-6D | inference time (s): 1.4832 -> 0.3991 | Improved |
| reduce DDIM sampling steps from 200 to 20 (ablation) | LGrasp6D · Grasp-Anything-6D | CR: 0.6694 -> 0.6425; time 1.4832 -> 0.1464 | Mixed |
| use LGrasp6D instead of LGD (2D language-driven grasp detector) | real robot grasping · Real robot, single object | success rate: 0.43 vs 0.43 | No clear effect |
Closest papers by meaning
- Language-driven Grasp Detection (CVPR 2024 · similarity 0.92)
- RealVLG-R1: A Large-Scale Real-World Visual-Language Grounding Benchmark for Robotic Perception and Manipulation (CVPR 2026 · similarity 0.89)
- Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge (CVPR 2024 · similarity 0.89)
- Active Perception for Grasp Detection via Neural Graspness Field (NeurIPS 2024 · similarity 0.87)
- Object-Centric Prompt-Driven Vision-Language-Action Model for Robotic Manipulation (CVPR 2025 · similarity 0.87)
- Exploring the Limits of Vision-Language-Action Manipulation in Cross-task Generalization (NeurIPS 2025 · similarity 0.86)
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.
