Search papers, labs, and topics across Lattice.
6
0
9
Achieve more factual radiology reports by unifying neuro-symbolic reasoning with active uncertainty minimization, outperforming existing encoder-decoder and retrieval-augmented methods.
Unlocking superior multimodal sentiment analysis, TSD reveals that disentangling features into common, pairwise, and private subspaces dramatically boosts performance.
By explicitly modeling a multi-level semantic hierarchy and carefully controlling information exchange between modalities, CLCR achieves state-of-the-art results in multimodal learning tasks ranging from emotion recognition to action recognition.
By adversarially synthesizing graph structures and self-correcting node labels, AdvSynGNN achieves state-of-the-art robustness against structural noise and heterophily in graph neural networks.
Achieve practical privacy-utility trade-offs in multimodal sentiment analysis with Missing-by-Design (MBD), a framework that surgically removes modality-specific information without full retraining.
Achieve a better trade-off between search precision and carbon footprint with GaiaFlow, a framework that uses semantic-guided diffusion tuning to make neural search systems more sustainable.
