URL: Universal Referential Knowledge Linking via Task-instructed Representation Compression

TL;DR

This paper introduces a universal model for referential knowledge linking that leverages task-instructed representation compression and multi-view learning, effectively handling diverse real-world linking tasks beyond specific domains.

Contribution

It proposes a unified approach for diverse referential knowledge linking tasks using LLM-driven representation compression and constructs a new benchmark for evaluation.

Findings

The framework outperforms previous methods significantly.

Universal RKL is challenging for existing approaches.

The approach adapts well across different scenarios.

Abstract

Linking a claim to grounded references is a critical ability to fulfill human demands for authentic and reliable information. Current studies are limited to specific tasks like information retrieval or semantic matching, where the claim-reference relationships are unique and fixed, while the referential knowledge linking (RKL) in real-world can be much more diverse and complex. In this paper, we propose universal referential knowledge linking (URL), which aims to resolve diversified referential knowledge linking tasks by one unified model. To this end, we propose a LLM-driven task-instructed representation compression, as well as a multi-view learning approach, in order to effectively adapt the instruction following and semantic understanding abilities of LLMs to referential knowledge linking. Furthermore, we also construct a new benchmark to evaluate ability of models on referential knowledge linking tasks across different scenarios. Experiments demonstrate that universal RKL is challenging for existing approaches, while the proposed framework can effectively resolve the task across various scenarios, and therefore outperforms previous approaches by a large margin.