AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

BIP: Bi-level Information Transfer and Completion Prompting for Visual Recognition with Missing Modalities

Haoran Fan, Xu Han, Xianglong Bao, Zheng Gao, Qi Fan, Yang Song, Jiaojiao Jiang

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 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 missing-type conditioning to HyperPrompt (ablation)HyperPrompt · MM-IMDbF1-Macro: 53.79 -> 54.32Improved
replace DCP correlated prompt with missing-type HyperPromptDCP · MM-IMDbF1-Macro: 53.41 -> 54.32Improved
add missing-type conditioning (CKA, Layer6) (ablation)HyperPrompt · MM-IMDb TXT-missingCKA similarity (%): 31.13 -> 52.01Improved
add missing-type conditioning (CKA, Layer4) (ablation)HyperPrompt · MM-IMDb TXT-missingCKA similarity (%): 68.34 -> 77.99Improved
add low-rank instance alignment (ablation)AlignPrompt · MM-IMDbF1-Macro: 53.89 -> 54.32Improved
add MI completion (ablation)AlignPrompt · MM-IMDbF1-Macro: 53.47 -> 54.32Improved
add both alignment and MI completion (ablation)AlignPrompt · MM-IMDbF1-Macro: 53.01 -> 54.32Improved
add HyperPrompt to AlignPrompt (ablation)BIP · MM-IMDb (50% image missing rate)F1-Macro: 55.89 -> 57.39Improved
prompt completion instead of feature reconstruction (Knowledge Bridger)AlignPrompt · MM-IMDb (50% image missing rate)F1-Macro: 51.95 -> 55.89Improved
prompt completion instead of feature reconstructionAlignPrompt · MM-IMDb image-missing validationlogit KL to full-modality prediction: 75.5% of samples lower KLImproved
replace DCP with BIP on tri-modal data (double-missing)BIP · CMU-MOSI double-missingACC: 70.14 -> 72.96Improved
increase trainable parameters from 2.8M to 7.9M (ablation)BIP · MM-IMDb testF1-Macro (%): 53.18 -> 54.32Improved

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