Ellipsoid-Based Decision Boundaries for Open Intent Classification

TL;DR

The paper introduces EliDecide, an innovative method for open intent classification that uses learnable ellipsoid decision boundaries to better capture distributional variance and improve detection of unknown intents in dialogue systems.

Contribution

EliDecide is the first approach to learn ellipsoid decision boundaries with varying scales for open intent classification, enhancing flexibility and detection accuracy.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Demonstrates superior open intent detection capability.

Shows strong potential for generalization to diverse text classification tasks.

Abstract

Textual open intent classification is crucial for real-world dialogue systems, enabling robust detection of unknown user intents without prior knowledge and contributing to the robustness of the system. While adaptive decision boundary methods have shown great potential by eliminating manual threshold tuning, existing approaches assume isotropic distributions of known classes, restricting boundaries to balls and overlooking distributional variance along different directions. To address this limitation, we propose EliDecide, a novel method that learns ellipsoid decision boundaries with varying scales along different feature directions. First, we employ supervised contrastive learning to obtain a discriminative feature space for known samples. Second, we apply learnable matrices to parameterize ellipsoids as the boundaries of each known class, offering greater flexibility than spherical boundaries defined solely by centers and radii. Third, we optimize the boundaries via a novelly designed dual loss function that balances empirical and open-space risks: expanding boundaries to cover known samples while contracting them against synthesized pseudo-open samples. Our method achieves state-of-the-art performance on multiple text intent benchmarks and further on a question classification dataset. The flexibility of the ellipsoids demonstrates superior open intent detection capability and strong potential for generalization to more text classification tasks in diverse complex open-world scenarios.