Research map / Multi-agent systems and games / Human-AI and goal-conditioned
March in Chat: Interactive Prompting for Remote Embodied Referring Expression
Yanyuan Qiao, Yuankai Qi, Zheng Yu, Jing Liu, Qi Wu
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 goal-oriented static planning (GOSP) from LLM (ablation) | HM3D-DUET with HLI · REVERIE val unseen | SPL: 40.49 -> 42.46 | Improved |
| add scene-oriented dynamic planning (SODP) (ablation) | HM3D-DUET with HLI · REVERIE val unseen | SPL: 40.49 -> 42.94 | Improved |
| SODP instead of GOSP on object grounding (ablation) | HM3D-DUET with HLI · REVERIE val unseen | RGSPL: 28.24 -> 27.81 | Mixed |
| combine GOSP and SODP (ablation) | HM3D-DUET with HLI · REVERIE val unseen | SPL: 40.49 -> 43.60 | Improved |
| make plan dynamic with ROASP feedback instead of static plan (ablation) | MiC · REVERIE val unseen | SPL: 41.74 -> 42.94 | Improved |
| add dynamic demonstration selection (ablation) | SODP · Human study relevancy | relevancy: 1.41 -> 2.06 | Improved |
| add ROASP scene perception to SODP (ablation) | SODP · Human study rationality | rationality: 1.55 -> 1.93 | Improved |
| MiC on top of HM3D-DUET vs HM3D-DUET alone | HM3D-DUET · REVERIE test unseen | SPL: 38.88 -> 41.97 | Improved |
Closest papers by meaning
- General Scene Adaptation for Vision-and-Language Navigation (ICLR 2025 · similarity 0.88)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.88)
- ImagineNav: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination (ICLR 2025 · similarity 0.87)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.87)
- VELMA: Verbalization Embodiment of LLM Agents for Vision and Language Navigation in Street View (AAAI 2024 · similarity 0.87)
- PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs (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.
