Research map / Vision-language and multimodal / Human and embodied modeling
CAESAR: An Embodied Simulator for Generating Multimodal Referring Expression Datasets
Md Mofijul Islam, Reza Manuel Mirzaiee, Alexi Gladstone, Haley N Green, Tariq Iqbal
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 gaze cue to verbal and visual input (V+NH -> V+G) (ablation) | CLIP · CAESAR-XL | accuracy: 64.78 -> 83.01 | Improved |
| add pointing gesture instead of gaze (V+NH -> V+P) (ablation) | ViLT · CAESAR-XL | accuracy: 64.86 -> 84.43 | Improved |
| use both gaze and pointing instead of gaze only (ablation) | CLIP · CAESAR-XL | accuracy: 83.01 -> 75.21 | Hurt |
| use both gaze and pointing instead of gaze only (ablation) | Late Fusion · CAESAR-XL | accuracy: 81.51 -> 76.00 | Hurt |
| use both gaze and pointing instead of gaze only (ablation) | Dual-Encoder · CAESAR-XL | accuracy: 72.26 -> 74.87 | Improved |
| verbal utterance only, no nonverbal cues (ablation) | BERT · CAESAR-XL | accuracy: 50.00 | Hurt |
| top view instead of exo view (ablation) | Late Fusion · CAESAR-XL | accuracy: 76.51 -> 88.37 | Improved |
| top view instead of exo view (ablation) | Dual-Encoder · CAESAR-XL | accuracy: 68.84 -> 71.04 | Improved |
| ego view instead of exo view (ablation) | ViLT · CAESAR-XL | accuracy: 62.47 -> 85.43 | Improved |
| all three views instead of best single view (ablation) | Late Fusion · CAESAR-XL | accuracy: 88.37 -> 76.00 | Hurt |
| all three views instead of best single view (ablation) | CLIP · CAESAR-XL | accuracy: 78.97 -> 75.21 | Hurt |
| SUM fusion instead of CONCAT (ablation) | CLIP · CAESAR-XL | accuracy: 75.21 -> 82.85 | Improved |
Closest papers by meaning
- A Large-Scale Human-Centric Benchmark for Referring Expression Comprehension in the LMM Era (NeurIPS 2024 · similarity 0.88)
- The Language of Motion: Unifying Verbal and Non-verbal Language of 3D Human Motion (CVPR 2025 · similarity 0.88)
- MultiPLY: A Multisensory Object-Centric Embodied Large Language Model in 3D World (CVPR 2024 · similarity 0.86)
- ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis (CVPR 2024 · similarity 0.86)
- Chain of Generation: Multi-Modal Gesture Synthesis via Cascaded Conditional Control (AAAI 2024 · similarity 0.85)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.85)
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.
