ULLME: A Unified Framework for Large Language Model Embeddings with Generation-Augmented Learning

TL;DR

ULLME introduces a versatile framework for large language model embeddings, combining bidirectional attention support and a novel fine-tuning method called GRL, significantly improving passage embedding quality across diverse LLM architectures.

Contribution

The paper presents ULLME, a flexible, plug-and-play framework supporting various LLMs and fine-tuning strategies, along with GRL, a new method leveraging generative abilities for better embeddings.

Findings

Strong performance on the Massive Text Embedding Benchmark

Supports multiple LLM architectures from 1.5B to 8B parameters

Open-source implementation available for practical use

Abstract

Large Language Models (LLMs) excel in various natural language processing tasks, but leveraging them for dense passage embedding remains challenging. This is due to their causal attention mechanism and the misalignment between their pre-training objectives and the text ranking tasks. Despite some recent efforts to address these issues, existing frameworks for LLM-based text embeddings have been limited by their support for only a limited range of LLM architectures and fine-tuning strategies, limiting their practical application and versatility. In this work, we introduce the Unified framework for Large Language Model Embedding (ULLME), a flexible, plug-and-play implementation that enables bidirectional attention across various LLMs and supports a range of fine-tuning strategies. We also propose Generation-augmented Representation Learning (GRL), a novel fine-tuning method to boost LLMs for text embedding tasks. GRL enforces consistency between representation-based and generation-based relevance scores, leveraging LLMs' powerful generative abilities for learning passage embeddings. To showcase our framework's flexibility and effectiveness, we release three pre-trained models from ULLME with different backbone architectures, ranging from 1.5B to 8B parameters, all of which demonstrate strong performance on the Massive Text Embedding Benchmark. Our framework is publicly available at: https://github.com/nlp-uoregon/ullme. A demo video for ULLME can also be found at https://rb.gy/ws1ile.