TRACE: TRansformer-based Attribution using Contrastive Embeddings in LLMs

TL;DR

This paper introduces TRACE, a transformer-based attribution framework utilizing contrastive embeddings to improve source attribution accuracy in large language models, addressing transparency and accountability challenges.

Contribution

We propose TRACE, a novel contrastive learning-based attribution method for LLMs, filling a gap in source attribution techniques for natural language processing.

Findings

TRACE significantly improves source attribution accuracy

TRACE demonstrates high efficiency across various settings

The framework enhances LLM transparency and trustworthiness

Abstract

The rapid evolution of large language models (LLMs) represents a substantial leap forward in natural language understanding and generation. However, alongside these advancements come significant challenges related to the accountability and transparency of LLM responses. Reliable source attribution is essential to adhering to stringent legal and regulatory standards, including those set forth by the General Data Protection Regulation. Despite the well-established methods in source attribution within the computer vision domain, the application of robust attribution frameworks to natural language processing remains underexplored. To bridge this gap, we propose a novel and versatile TRansformer-based Attribution framework using Contrastive Embeddings called TRACE that, in particular, exploits contrastive learning for source attribution. We perform an extensive empirical evaluation to demonstrate the performance and efficiency of TRACE in various settings and show that TRACE significantly improves the ability to attribute sources accurately, making it a valuable tool for enhancing the reliability and trustworthiness of LLMs.