Towards LLM-guided Efficient and Interpretable Multi-linear Tensor Network Rank Selection
This work addresses the challenge of making tensor network decompositions accessible and interpretable for users without specialized domain expertise, though it appears incremental by combining existing LLM capabilities with a known bottleneck in tensor analysis.
The authors tackled the problem of rank selection in tensor network models for higher-order data analysis by proposing a framework that uses large language models (LLMs) to guide the process, resulting in enhanced interpretability and strong generalization on financial datasets.
We propose a novel framework that leverages large language models (LLMs) to guide the rank selection in tensor network models for higher-order data analysis. By utilising the intrinsic reasoning capabilities and domain knowledge of LLMs, our approach offers enhanced interpretability of the rank choices and can effectively optimise the objective function. This framework enables users without specialised domain expertise to utilise tensor network decompositions and understand the underlying rationale within the rank selection process. Experimental results validate our method on financial higher-order datasets, demonstrating interpretable reasoning, strong generalisation to unseen test data, and its potential for self-enhancement over successive iterations. This work is placed at the intersection of large language models and higher-order data analysis.