Research map / 3D vision / Gaussian splatting
Enhancing Embodied Reasoning and Grounding by Novel View Synthesis
Yu-Ji Kim, Dahye Lee, Kim Jun-Seong, Nam Hyeon-Woo, GeonU Kim, Yongjin Kwon, Yu-Chiang Frank Wang, Jaesung Choe, Tae-Hyun Oh
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 |
|---|---|---|---|
| use visibility score instead of 3D-Mem view selection (ablation) | SplatReasoner · EM-EQA 184 subset | LLM-Match: 45.4 -> 48.2 | Improved |
| use visibility score instead of volume score (ablation) | SplatReasoner · EM-EQA 184 subset | LLM-Match: 47.8 -> 48.2 | Improved |
| add novel view synthesis and final view selection (ablation) | SplatReasoner · EM-EQA 184 subset | LLM-Match: 48.2 -> 50.5 | Improved |
| use text-based LLM judge instead of image-based VLM judge (ablation) | SplatReasoner · EM-EQA 184 subset | LLM-Match: 47.1 -> 50.5 | Improved |
| use text-only judge instead of image+text (ablation) | SplatReasoner · EM-EQA 184 subset | LLM-Match: 46.9 -> 50.5 | Improved |
| add verification stage (ablation) | SplatReasoner · EM-EQA 1,623 | LLM-Match: 54.5 -> 57.8 | Improved |
| SplatReasoner vs 3D-Mem (GPT-4o) | SplatReasoner · EM-EQA | LLM-Match: 54.6 -> 57.8 | Improved |
| SplatReasoner vs 3D-Mem (Qwen3-VL-8B) | SplatReasoner · EM-EQA | LLM-Match: 50.8 -> 51.6 | Improved |
| SplatReasoner vs Dr. Splat (sentence) on referring segmentation Acc@8 | SplatReasoner · 3D referring segmentation | Acc@8: 45.21 vs 45.14 | No clear effect |
| SplatReasoner vs Dr. Splat (category) on Multi3DRefer | SplatReasoner · Multi3DRefer | Acc@8: 18.33 -> 29.59 | Improved |
| ReferSplat vs others on grounding | ReferSplat · ScanRefer | 3D mIoU: 3.14 vs 9.44 (Dr. Splat sentence) | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- 3D-Mem — 3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning (CVPR 2025) · better on 4 of 4 shared results
- ReferSplat — ReferSplat: Referring Segmentation in 3D Gaussian Splatting (ICML 2025) · better on 1 of 1 shared result
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
- VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction (CVPR 2026 · similarity 0.89)
- Abstract 3D Perception for Spatial Intelligence in Vision-Language Models (CVPR 2026 · similarity 0.89)
- Seeing from Another Perspective: Evaluating Multi-View Understanding in MLLMs (AAAI 2026 · similarity 0.89)
- GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models (ICLR 2026 · similarity 0.89)
- Pursuing Minimal Sufficiency in Spatial Reasoning (ICLR 2026 · similarity 0.88)
- S$^2$-MLLM: Boosting Spatial Reasoning Capability of MLLMs for 3D Visual Grounding with Structural Guidance (CVPR 2026 · 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.
