Research map / Vision-language and multimodal / Human and embodied modeling
Multi-Task Domain Adaptation for Language Grounding with 3D Objects
Penglei Sun, Yaoxian Song, Xinglin Pan, Peijie Dong, Xiaofei Yang, Qiang Wang, Zhixu Li, Tiefeng Li, Xiaowen Chu
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 |
|---|---|---|---|
| replace frozen or fully trained domain encoder with LoRA-style domain adapter (ablation) | DA4LG · SNARE validation (All) | accuracy: 84.9 -> 86.8 | Improved |
| train full domain-specific encoder instead of adapter (ablation) | DA4LG · SNARE validation (All) | accuracy: 86.8 -> 80.8 | Hurt |
| add VLC contrastive auxiliary task (ablation) | DA4LG · SNARE validation (All) | accuracy: 81.9 -> 85.4 | Improved |
| add VGC caption auxiliary task (ablation) | DA4LG · SNARE validation (All) | accuracy: 81.9 -> 85.0 | Improved |
| add both VLC and VGC to LGR (ablation) | DA4LG · SNARE validation (All) | accuracy: 81.9 -> 86.8 | Improved |
| add domain-specific encoder alongside vision encoder (ablation) | DA4LG · SNARE validation (All) | accuracy: 82.9 -> 86.8 | Improved |
| initialize domain encoder from CLIP-B pre-training instead of scratch (ablation) | DA4LG · SNARE validation (All) | accuracy: 81.6 -> 86.8 | Improved |
| use larger ViT-L as domain-specific encoder instead of ViT-B (ablation) | DA4LG · SNARE validation (All) | accuracy: 86.8 -> 85.0 | No clear effect |
| use domain adaptation in single-view setting versus multi-view baselines | DA4LG · SNARE validation (All) | accuracy: 82.0 -> 83.8 | Improved |
| apply DA4LG zero-shot to simulation benchmark | DA4LG · Simulation-SNARE (multi-view) | accuracy: 61.4 -> 63.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- MAGiC — MAGIC: Multimodal relAtional Graph adversarIal inferenCe for Diverse and Unpaired Text-Based Image Captioning (AAAI 2022) · better on 2 of 4 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
- Cross-Modal Match for Language Conditioned 3D Object Grounding (AAAI 2024 · similarity 0.90)
- From Thousands to Billions: 3D Visual Language Grounding via Render-Supervised Distillation from 2D VLMs (ICML 2025 · similarity 0.89)
- Unifying 2D and 3D Vision-Language Understanding (ICML 2025 · similarity 0.88)
- Towards CLIP-driven Language-free 3D Visual Grounding via 2D-3D Relational Enhancement and Consistency (CVPR 2024 · similarity 0.88)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.88)
- AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring (AAAI 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.
