Research map / 3D vision / 3D generation
Multimodal Semantic Bias Mitigation for Diverse Text-To-3D Generation
Yukuan Min, Muli Yang, Jinhao Zhang, Yuxuan Wang, Yihang Zhu, Jiexi Yan, Cheng Deng
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 |
|---|---|---|---|
| fine-tune TRELLIS-text with bias-guided enriched text-3D pairs | TRELLIS-text · MATE-3D Basic | overall quality score: 7.39 -> 8.19 | Improved |
| fine-tune with enriched pairs | TRELLIS-text · MATE-3D Grouped | overall quality score: 5.73 -> 6.64 | Improved |
| fine-tune with enriched pairs | TRELLIS-text · T3Bench multiple objects | average score: 28.5 -> 37.5 | Improved |
| fine-tune with enriched pairs | TRELLIS-text · T3Bench single object | average score: 44.8 -> 50.2 | Improved |
| remove gradient guidance when generating enriched prompts (ablation) | TRELLIS-text w/ ours · MATE-3D Fantastic | overall quality score: 6.95 -> 3.24 | Hurt |
| use only least-important word wl (ablation) | TRELLIS-text w/ ours · MATE-3D Fantastic | overall quality score: 6.95 -> 5.23 | Hurt |
| add least-important word wl in addition to wh (ablation) | TRELLIS-text w/ ours · MATE-3D Fantastic | overall quality score: 6.43 -> 6.95 | Improved |
| augment One-2-3-45++ prompts without bias guidance (ablation) | One-2-3-45++ · MATE-3D Basic | overall quality score: 7.79 -> 7.86 | No clear effect |
| select hard samples by word gradient instead of evaluation score (50%) (ablation) | TRELLIS-text w/ ours · MATE-3D Basic | overall quality score: 7.87 -> 7.92 | Improved |
| use 100% samples with 4 enriched prompts instead of 1 (ablation) | TRELLIS-text w/ ours · MATE-3D Spatial | overall quality score: 6.63 -> 6.92 | Improved |
| halve training samples via gradient-based hard sampling (ablation) | TRELLIS-text w/ ours · MATE-3D Basic | overall quality score: 8.07 -> 7.92 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- One-2-3-45++ — One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization (NeurIPS 2023) · better on 3 of 3 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
- CG-MLLM: Captioning and Generating 3D content via Multi-modal Large Language Models (ICML 2026 · similarity 0.89)
- Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation (ICLR 2024 · similarity 0.89)
- MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation (CVPR 2025 · similarity 0.89)
- DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization (ICML 2025 · similarity 0.88)
- VP3D: Unleashing 2D Visual Prompt for Text-to-3D Generation (CVPR 2024 · similarity 0.88)
- Progressive3D: Progressively Local Editing for Text-to-3D Content Creation with Complex Semantic Prompts (ICLR 2024 · 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.
