Refining Sentence Embedding Model through Ranking Sentences Generation with Large Language Models

Liyang He, Chenglong Liu, Rui Li, Zhenya Huang, Shulan Ruan +2 more
2/19/2025
cs.CL

No official code found. Best implementation: hly1998/RankingSentenceGeneration (83 stars, last commit 2025-07-14).

Abstract

Sentence embedding is essential for many NLP tasks, with contrastive learning methods achieving strong performance using annotated datasets like NLI. Yet, the reliance on manual labels limits scalability. Recent studies leverage large language models (LLMs) to generate sentence pairs, reducing annotation dependency. However, they overlook ranking information crucial for fine-grained semantic distinctions. To tackle this challenge, we propose a method for controlling the generation direction of LLMs in the latent space. Unlike unconstrained generation, the controlled approach ensures meaningful semantic divergence. Then, we refine exist sentence embedding model by integrating ranking information and semantic information. Experiments on multiple benchmarks demonstrate that our method achieves new SOTA performance with a modest cost in ranking sentence synthesis.

View on arXivView PDF

Code Implementations(1)

hly1998/RankingSentenceGenerationCommunity100%

The code for Refining Sentence Embedding Model through Ranking Sentences Generation with Large Language Models (Finding of ACL2025)

836Shell, PythonFeb 19, 20251 years ago

Similar Papers

Improving Sentence Embeddings with Automatic Generation of Training Data Using Few-shot Examples

86% match1 repo12024

Unveiling Key Aspects of Fine-Tuning in Sentence Embeddings: A Representation Rank Analysis

84% match2024

Optimizing Sentence Embedding with Pseudo-Labeling and Model Ensembles: A Hierarchical Framework for Enhanced NLP Tasks

84% match2025

Narrowing the Gap between Supervised and Unsupervised Sentence Representation Learning with Large Language Model

84% match1 repo52023

Enhancing Unsupervised Sentence Embeddings via Knowledge-Driven Data Augmentation and Gaussian-Decayed Contrastive Learning

84% match2024

SemPA: Improving Sentence Embeddings of Large Language Models through Semantic Preference Alignment

84% match1 repo42026

Cite this paper

@article{he2025refining,
  title  = {Refining Sentence Embedding Model through Ranking Sentences Generation with Large Language Models},
  author = {Liyang He and Chenglong Liu and Rui Li and Zhenya Huang and Shulan Ruan and Jun Zhou and Enhong Chen},
  year   = {2025},
  eprint = {2502.13656},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2502.13656v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2502.13656.svg)](https://codeofpaper.com/paper/2502.13656)

Discussion