Research map / Vision-language and multimodal / Robot action and navigation
SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models
Haowen Liu, Shaoxiong Yao, Haonan Chen, Jiawei Gao, Jiayuan Mao, Jia-Bin Huang, Yilun Du
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 simulation-enabled VLM planning (SIMPACT) over pi0.5 VLA | pi0.5 · Seven real-world manipulation tasks | success rate: 0% -> 80% on non-toppling push | Improved |
| add simulation-enabled VLM planning over VoxPoser/MOKA | VoxPoser, MOKA · Real-world shape dough | success rate: 0% -> 80% | Improved |
| add VLM action sampler instead of Gaussian sampling (ablation) | SIMPACT · Non-toppling push | success rate: 0% -> 80% | Improved |
| add VLM action sampler instead of Gaussian sampling (ablation) | SIMPACT · Shape rope | success rate: 0% -> 90% | Improved |
| add simulation rollout context (ablation) | SIMPACT · Bowl stacking | success rate: 0% -> 60% | Improved |
| add simulation rollout context (ablation) | SIMPACT · Avoid obstacle | success rate: 0% -> 80% | Improved |
| add simulation rollout context (ablation) | SIMPACT · Sweeping | success rate: 20% -> 70% | Improved |
| add iterative VLM optimizer (ablation) | SIMPACT · Non-toppling push | success rate: 30% -> 80% | Improved |
| add iterative VLM optimizer (ablation) | SIMPACT · Shape dough | success rate: 80% -> 80% | No clear effect |
| add iterative VLM optimizer (ablation) | SIMPACT · Sweeping | success rate: 70% -> 70% | No clear effect |
| increase initial samples from 3 to 10 (ablation) | SIMPACT · Shape rope | success rate: 40% -> 90% | Improved |
| increase initial samples from 10 to 20 (ablation) | SIMPACT · Bowl stacking | success rate: 60% -> 20% | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MOKA — MokA: Multimodal Low-Rank Adaptation for MLLMs (NeurIPS 2025) · better on 7 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PAC Bench: Do Foundation Models Understand Prerequisites for Executing Manipulation Policies? (NeurIPS 2025 · similarity 0.91)
- OmniManip: Towards General Robotic Manipulation via Object-Centric Interaction Primitives as Spatial Constraints (CVPR 2025 · similarity 0.89)
- Localizing, Structuring, and Rendering: Bridging 3D and 2D Vision-Language-Action Models for Robotic Manipulation (CVPR 2026 · similarity 0.89)
- RoboCerebra: A Large-scale Benchmark for Long-horizon Robotic Manipulation Evaluation (NeurIPS 2025 · similarity 0.89)
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025 · similarity 0.88)
- DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge (NeurIPS 2025 · similarity 0.88)
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.
