AI Paper Atlas

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

Training Transitive and Commutative Multimodal Transformers with LoReTTa

Manuel Tran, Yashin Dicente Cid, Amal Lahiani, Fabian J Theis, Tingying Peng, Eldad Klaiman

NeurIPS 2023 · 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 transitive modeling (LoReTTa) on top of C2M3, pretrained on (T,I),(T,A)C2M3 · SVL-MNIST unseen pair (I,A)perplexity: 104.27 -> 5.04Improved
add transitive modeling, pretrained on (A,I),(A,T)C2M3 · SVL-MNIST unseen pair (I,T)perplexity: 9.70 -> 3.13Improved
add transitive modeling, pretrained on (I,T),(I,A)C2M3 · SVL-MNIST unseen pair (T,A)perplexity: 30.88 -> 11.21Improved
add transitive modeling, pretrained on (T,I),(T,A)C2M3 · SVL-MNIST probe accuracy, unseen (I,A)accuracy: 72.9 -> 80.9Improved
add transitive modeling, pretrained on (I,T),(I,A)C2M3 · SVL-MNIST probe accuracy, unseen (T,A)accuracy: 71.2 -> 84.0Improved
add transitive modeling, pretrained on (A,I),(A,T)C2M3 · SVL-MNIST probe accuracy, unseen (I,T,A)accuracy: 89.9 -> 91.6Improved
compare LoReTTa with full trimodal C2M3 upper baselineLoReTTa · SVL-MNIST probe accuracy (I,T,A)accuracy: 88.5 -> 91.6Improved
add commutative bimodal pretraining (C2M3 (I,T) vs CM3 (I))CM3 · SVL-MNIST probe accuracy, modality Iaccuracy: 79.7 -> 82.9Improved
add transitive modeling to BERT (T-BERT)BERT · TCGA-OMICS unseen (I,R)c-index: 0.594 -> 0.623Mixed
add transitive modeling to GPT (T-GPT)GPT · TCGA-OMICS unseen (M,I,R)c-index: 0.588 -> 0.622Improved
apply LoReTTa instead of C2M3C2M3 · TCGA-OMICS unseen (I,R)c-index: 0.571 -> 0.623Improved
apply LoReTTa instead of C2M3C2M3 · TCGA-OMICS modality Rc-index: 0.599 -> 0.563Hurt

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