Skip to yearly menu bar Skip to main content


Poster

RETSim: Resilient and Efficient Text Similarity

Marina Zhang · Owen Vallis · Aysegul Bumin · Tanay Vakharia · Elie Bursztein

Halle B #236
[ ] [ Project Page ]
Fri 10 May 1:45 a.m. PDT — 3:45 a.m. PDT

Abstract:

This paper introduces RETSim (Resilient and Efficient Text Similarity), a lightweight, multilingual deep learning model trained to produce robust metric embeddings for near-duplicate text retrieval, clustering, and dataset deduplication tasks. We demonstrate that RETSim is significantly more robust and accurate than MinHash and neural text embeddings, achieving new state-of-the-art performance on dataset deduplication, adversarial text retrieval benchmarks, and spam clustering tasks. Additionally, we introduce the W4NT3D benchmark (Wiki-40B 4dversarial Near-T3xt Dataset), enabling the evaluation of models on typo-laden near-duplicate text retrieval in a multilingual setting. RETSim and the W4NT3D benchmark are released under the MIT License at https://github.com/google/unisim.

Chat is not available.