Research map / Video / Video-language understanding
EC2: Emergent Communication for Embodied Control
Yao Mu, Shunyu Yao, Mingyu Ding, Ping Luo, Chuang Gan
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 emergent language to pre-training (ablation) | EC² · MetaWorld, language instruction, 10 and 25 demos | success rate: 69.0±2.0% -> 72.4±2.4%; 78.0±2.4% -> 82.6±2.0% | Improved |
| add emergent language to pre-training (ablation) | EC² · Franka Kitchen, language instruction, 10 and 25 demos | success rate: 43.0±3.3% -> 48.4±2.5%; 54.0±2.6% -> 59.8±2.5% | Improved |
| add natural language to pre-training (ablation) | EC² · MetaWorld, video instruction, 10 and 25 demos | success rate: 72.6±2.4% -> 76.0±2.1%; 82.0±2.0% -> 85.0±2.5% | Improved |
| add natural language to pre-training (ablation) | EC² · Franka Kitchen, video instruction, 10 and 25 demos | success rate: 49.0±2.5% -> 53.6±2.5%; 61.3±2.4% -> 63.2±2.6% | Improved |
| replace emergent language with pre-trained video captions (ablation) | EC² · MetaWorld, video instruction, 10 and 25 demos | success rate: 73.0±2.2% -> 76.0±2.1%; 73.2±2.5% -> 85.0±2.5% | Mixed |
| replace emergent language with pre-trained video captions (ablation) | EC² · Franka Kitchen, video instruction, 10 and 25 demos | success rate: 48.6±2.4% -> 53.6±2.5%; 49.0±2.3% -> 63.2±2.6% | Improved |
Closest papers by meaning
- Pre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control (NeurIPS 2024 · similarity 0.88)
- EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents (ICML 2025 · similarity 0.87)
- Any2Policy: Learning Visuomotor Policy with Any-Modality (NeurIPS 2024 · similarity 0.87)
- Provable Ordering and Continuity in Vision-Language Pretraining for Generalizable Embodied Agents (NeurIPS 2025 · similarity 0.87)
- ESCA: Contextualizing Embodied Agents via Scene-Graph Generation (NeurIPS 2025 · similarity 0.87)
- RoboDreamer: Learning Compositional World Models for Robot Imagination (ICML 2024 · similarity 0.87)
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.
