ICXML: An In-Context Learning Framework for Zero-Shot Extreme Multi-Label Classification
It addresses the lack of supervision in real-world XMC scenarios, enabling classification in extremely large label spaces without labeled data.
The paper tackles the problem of zero-shot extreme multi-label classification by introducing ICXML, a two-stage framework that uses in-context learning to generate candidate labels and rerank them, achieving state-of-the-art results on two public benchmarks.
This paper focuses on the task of Extreme Multi-Label Classification (XMC) whose goal is to predict multiple labels for each instance from an extremely large label space. While existing research has primarily focused on fully supervised XMC, real-world scenarios often lack supervision signals, highlighting the importance of zero-shot settings. Given the large label space, utilizing in-context learning approaches is not trivial. We address this issue by introducing In-Context Extreme Multilabel Learning (ICXML), a two-stage framework that cuts down the search space by generating a set of candidate labels through incontext learning and then reranks them. Extensive experiments suggest that ICXML advances the state of the art on two diverse public benchmarks.