AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Adapting Point Cloud Analysis via Multimodal Bayesian Distribution Learning

Xingyu Zhu, Liang Yi, Shuo Wang, Wenbo Zhu, Yongliang Wu, Beier Zhu, Hanwang Zhang

CVPR 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 0 citations (OpenAlex)

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
add geometric distribution learning to fixed text prototypes (ablation)BayesMM on ULIP-2 · S-OBJaccuracy: 42.00 -> 46.47Improved
add geometric distribution learning to fixed text prototypes (ablation)BayesMM on ULIP-2 · Omni3Daccuracy: 26.58 -> 26.63No clear effect
add textual distribution learning instead (ablation)BayesMM on ULIP-2 · S-OBJaccuracy: 42.00 -> 52.50Improved
add textual distribution learning instead (ablation)BayesMM on ULIP-2 · Omni3Daccuracy: 26.58 -> 30.79Improved
add geometric distribution and Bayesian weighting on top of textual distribution (ablation)BayesMM on ULIP-2 · S-OBJaccuracy: 52.50 -> 53.02Improved
add geometric distribution and Bayesian weighting on top of textual distribution (ablation)BayesMM on ULIP-2 · Omni3Daccuracy: 30.79 -> 31.12Improved
vary alpha^2 weighting strengthBayesMM · ModelNet-Caccuracy: minimal impactNo clear effect
vary beta^2 variance scaleBayesMM · ModelNet-Caccuracy: ULIP stays around 59.4%, ULIP-2 near 65.2%No clear effect
apply BayesMM instead of Point-Cache (Hierarchical)ULIP · ModelNet-C avgaccuracy: 55.02 -> 59.42Improved
apply BayesMM to zero-shot backboneULIP · ModelNet-C avgaccuracy: 48.60% to 59.42% (+10.82)Improved
apply BayesMM to zero-shot backboneUni3D · Generalization avgaccuracy: 54.09 -> 57.72Improved
apply BayesMM instead of Point-Cache (Hierarchical)Uni3D · Sim-to-real avgaccuracy: 69.07 -> 70.94Mixed

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