Reliability-Aware Probabilistic Fusion for Multimodal Learning with Incomplete Evidence
We introduce a reliability-aware Gaussian Product-of-Experts (PoE) in which an input-dependent reliability weight tempers each modality's latent precision, while missing experts are omitted.
ProofPaper ↗
Key points
- Multimodal inputs may be missing or degraded, raising distinct questions of evidence availability and trust.
- We characterize how reliability and availability affect posterior information and expert influence.
- Across two clinical multimodal tasks, the full framework remains competitive with complete inputs and achieves its clearest mean-AUROC advantage under modality absence, particularly on HAIM-MIMIC-MM.
- Ablations show a substantial gain from missing-aware training and an additional HAIM improvement from reliability weighting under matched PoE, missing-aware, and consistency training.
Sources (1)
- [1]Reliability-Aware Probabilistic Fusion for Multimodal Learning with Incomplete EvidencearXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 21, 12:04 PM
We introduce a reliability-aware Gaussian Product-of-Experts (PoE) in which an input-dependent reliability weight tempers each modality's latent precision, while missing experts are omitted.
Multimodal inputs may be missing or degraded, raising distinct questions of evidence availability and trust.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 21, 2026Mistral Advances AI for Engineering with NVIDIA
- Oct 21, 2026OpenAI Frontier Models Arrive on Google Cloud and Model Garden
- Oct 20, 2026Voxtral 3
- Oct 20, 2026[AINews] DeepSeek DeepInfer: Open Source SOTA Inference running 155T toks/day/cluster making $22m/day at ~80% margin for a 81 day payback
- Oct 20, 2026Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds
- Oct 19, 2026Introducing Clairic, a new System One class of extraction models