AI Paper Atlas

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

CVPR 2026 · 3D vision › 3D generation · approach: Text-to-3D scene generation

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 changedWhereMeasuredEffect
fine-tune TRELLIS-text with bias-guided enriched text-3D pairsTRELLIS-text · MATE-3D Basicoverall quality score: 7.39 -> 8.19Improved
fine-tune with enriched pairsTRELLIS-text · MATE-3D Groupedoverall quality score: 5.73 -> 6.64Improved
fine-tune with enriched pairsTRELLIS-text · T3Bench multiple objectsaverage score: 28.5 -> 37.5Improved
fine-tune with enriched pairsTRELLIS-text · T3Bench single objectaverage score: 44.8 -> 50.2Improved
remove gradient guidance when generating enriched prompts (ablation)TRELLIS-text w/ ours · MATE-3D Fantasticoverall quality score: 6.95 -> 3.24Hurt
use only least-important word wl (ablation)TRELLIS-text w/ ours · MATE-3D Fantasticoverall quality score: 6.95 -> 5.23Hurt
add least-important word wl in addition to wh (ablation)TRELLIS-text w/ ours · MATE-3D Fantasticoverall quality score: 6.43 -> 6.95Improved
augment One-2-3-45++ prompts without bias guidance (ablation)One-2-3-45++ · MATE-3D Basicoverall quality score: 7.79 -> 7.86No clear effect
select hard samples by word gradient instead of evaluation score (50%) (ablation)TRELLIS-text w/ ours · MATE-3D Basicoverall quality score: 7.87 -> 7.92Improved
use 100% samples with 4 enriched prompts instead of 1 (ablation)TRELLIS-text w/ ours · MATE-3D Spatialoverall quality score: 6.63 -> 6.92Improved
halve training samples via gradient-based hard sampling (ablation)TRELLIS-text w/ ours · MATE-3D Basicoverall quality score: 8.07 -> 7.92Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy