Tag
This paper presents SAIL, a method for disentangling and representing human skill as interpretable multi-dimensional embeddings to improve behavior prediction and AI coaching, focusing on robustness and generalizability.
This paper proposes MIDAS, a unified framework for incomplete multimodal sentiment analysis that uses mutual information disentanglement and uncertainty-aware fusion to robustly represent and integrate modalities under missing-data conditions.
This paper proposes a unified definition of uncertainty as pointwise posterior risk and introduces a theory-backed benchmark using semi-synthetic datasets to directly compute oracle epistemic and aleatoric uncertainty, enabling fine-grained evaluation beyond proxy tasks.
The paper proposes NULLs (Natively Unlearnable LLMs), a model class that isolates source-specific contributions in sparsely activated sinks while sharing backbone neurons, enabling clean unlearning of individual data sources without retraining and preserving general language capabilities.