ESMC: MLLM-Based Embedding Selection for Explainable Multiple Clustering
This work addresses the problem of inflexible clustering for users needing multiple or tailored results, though it is incremental as it builds on existing MLLM and clustering techniques.
The paper tackles the limitation of deep clustering methods that provide only one clustering result per dataset by proposing a method using multi-modal large language models (MLLMs) to achieve user-driven clustering based on semantic requirements, with experiments showing competitive performance on diverse datasets and metrics.
Typical deep clustering methods, while achieving notable progress, can only provide one clustering result per dataset. This limitation arises from their assumption of a fixed underlying data distribution, which may fail to meet user needs and provide unsatisfactory clustering outcomes. Our work investigates how multi-modal large language models (MLLMs) can be leveraged to achieve user-driven clustering, emphasizing their adaptability to user-specified semantic requirements. However, directly using MLLM output for clustering has risks for producing unstructured and generic image descriptions instead of feature-specific and concrete ones. To address these issues, our method first discovers that MLLMs' hidden states of text tokens are strongly related to the corresponding features, and leverages these embeddings to perform clusterings from any user-defined criteria. We also employ a lightweight clustering head augmented with pseudo-label learning, significantly enhancing clustering accuracy. Extensive experiments demonstrate its competitive performance on diverse datasets and metrics.