DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings
Not on the map. Why?
Abstract (arXiv)
We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between the original sentence and an edited sentence, where the edited sentence is obtained by stochastically masking out the original sentence and then sampling from a masked language model. We show that DiffSCE is an instance of equivariant contrastive learning (Dangovski et al., 2021), which generalizes contrastive learning and learns representations that are insensitive to certain types of augmentations and sensitive to other "harmful" types of augmentations. Our experiments show that DiffCSE achieves state-of-the-art results among unsupervised sentence representation learning methods, outperforming unsupervised SimCSE by 2.3 absolute points on semantic textual similarity tasks.
Key facts. Cited by 132 papers in the map (data through 14 Oct 2026). Builds on 26 of its 41 references that are in the map. First posted on arXiv 21 Apr 2022.
Builds on
26 of 41 references.
References · 26
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
- 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
- 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
- 2020Exploring Simple Siamese Representation Learningcited by 2,866
- 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,173
- 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,405
- 2017SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluationcited by 1,001
- 2017Supervised Learning of Universal Sentence Representations from Natural Language Inference Datacited by 432
14 more
- 2015Skip-Thought Vectorscited by 322
- 2018SentEval: An Evaluation Toolkit for Universal Sentence Representationscited by 294
- 2020DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representationscited by 251
- 2021ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfercited by 231
- 2021Whitening Sentence Representations for Better Semantics and Faster Retrievalcited by 202
- 2018An efficient framework for learning sentence representationscited by 198
- 2021On Feature Decorrelation in Self-Supervised Learningcited by 160
- 2021COCO-LM: Correcting and Contrasting Text Sequences for Language Model Pretrainingcited by 126
- 2021Self-Guided Contrastive Learning for BERT Sentence Representationscited by 103
- 2016Learning Distributed Representations of Sentences from Unlabelled Datacited by 91
- 2020An Unsupervised Sentence Embedding Method by Mutual Information Maximizationcited by 91
- 2020On the Sentence Embeddings from Pre-trained Language Modelscited by 83
- 2021Equivariant Contrastive Learningcited by 66
- 2020Universal Sentence Representation Learning with Conditional Masked Language Modelcited by 34
Built on by
Citing papers per year: 2022: 18; 2023: 57; 2024: 32; 2025: 15; 2026: 10.
132 papers.
Sort:
By area:
- 2022RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encodercited by 118
- 2023AnglE-optimized Text Embeddingscited by 99
- 2023Scaling Sentence Embeddings with Large Language Modelscited by 85
- 2023Exploring the Benefits of Training Expert Language Models over Instruction Tuningcited by 80
- 2023RaSa: Relation and Sensitivity Aware Representation Learning for Text-based Person Searchcited by 54
- 2023BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddingscited by 32
- 2023Improving Contrastive Learning of Sentence Embeddings from AI Feedbackcited by 29
- 2022InfoCSE: Information-aggregated Contrastive Learning of Sentence Embeddingscited by 26
- 2023RankCSE: Unsupervised Sentence Representations Learning via Learning to Rankcited by 23
- 2022UnifieR: A Unified Retriever for Large-Scale Retrievalcited by 19
- 2022ContraCLM: Contrastive Learning For Causal Language Modelcited by 18
- 2024Question-Based Retrieval using Atomic Units for Enterprise RAGcited by 17
38 more
- 2024Meta-Task Prompting Elicits Embeddings from Large Language Modelscited by 16
- 2022Ranking-Enhanced Unsupervised Sentence Representation Learningcited by 15
- 2023SimCSE++: Improving Contrastive Learning for Sentence Embeddings from Two Perspectivescited by 15
- 2024BIRCO: A Benchmark of Information Retrieval Tasks with Complex Objectivescited by 15
- 2023Composition-contrastive Learning for Sentence Embeddingscited by 14
- 2025Formalizing Learning from Language Feedback with Provable Guaranteescited by 13
- 2022Don't Judge a Language Model by Its Last Layer: Contrastive Learning with Layer-Wise Attention Poolingcited by 12
- 2023A Comprehensive Survey of Sentence Representations: From the BERT Epoch to the ChatGPT Era and Beyondcited by 12
- 2023CoTEVer: Chain of Thought Prompting Annotation Toolkit for Explanation Verificationcited by 11
- 2023ACCENT: An Automatic Event Commonsense Evaluation Metric for Open-Domain Dialogue Systemscited by 11
- 2022Non-Linguistic Supervision for Contrastive Learning of Sentence Embeddingscited by 10
- 2023TimeDRL: Disentangled Representation Learning for Multivariate Time-Seriescited by 10
- 2022Differentiable Data Augmentation for Contrastive Sentence Representation Learningcited by 9
- 2024Understanding and Mitigating the Threat of Vec2Text to Dense Retrieval Systemscited by 9
- 2024Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Modelscited by 9
- 2024D2LLM: Decomposed and Distilled Large Language Models for Semantic Searchcited by 9
- 2022SDA: Simple Discrete Augmentation for Contrastive Sentence Representation Learningcited by 8
- 2023CoT-BERT: Enhancing Unsupervised Sentence Representation through Chain-of-Thoughtcited by 8
- 20242D Matryoshka Sentence Embeddingscited by 8
- 2024Hyper-CL: Conditioning Sentence Representations with Hypernetworkscited by 8
- 2022miCSE: Mutual Information Contrastive Learning for Low-shot Sentence Embeddingscited by 7
- 2023Logic Against Bias: Textual Entailment Mitigates Stereotypical Sentence Reasoningcited by 7
- 2023Improving Contrastive Learning of Sentence Embeddings with Focal-InfoNCEcited by 7
- 2023Learning Disentangled Semantic Spaces of Explanations via Invertible Neural Networkscited by 6
- 2023Efficient Domain Adaptation of Sentence Embeddings Using Adapterscited by 6
- 2023Length is a Curse and a Blessing for Document-level Semanticscited by 6
- 2023How Well Do Text Embedding Models Understand Syntax?cited by 6
- 2023Improving Self-supervised Molecular Representation Learning using Persistent Homologycited by 6
- 2024Fine-tuning CLIP Text Encoders with Two-step Paraphrasingcited by 6
- 2022Masked Autoencoders As The Unified Learners For Pre-Trained Sentence Representationcited by 5
- 2022AugCSE: Contrastive Sentence Embedding with Diverse Augmentationscited by 5
- 2023Clustering-Aware Negative Sampling for Unsupervised Sentence Representationcited by 5
- 2023Bridging Continuous and Discrete Spaces: Interpretable Sentence Representation Learning via Compositional Operationscited by 5
- 2023Revealing the Blind Spot of Sentence Encoder Evaluation by HEROScited by 5
- 2023AspectCSE: Sentence Embeddings for Aspect-based Semantic Textual Similarity Using Contrastive Learning and Structured Knowledgecited by 5
- 2023DebCSE: Rethinking Unsupervised Contrastive Sentence Embedding Learning in the Debiasing Perspectivecited by 5
- 2023On the Dimensionality of Sentence Embeddingscited by 5
- 2023Contrastive Learning-based Sentence Encoders Implicitly Weight Informative Wordscited by 5
50 most cited of 132 All 132 (JSON)
- 2026Cluster Validation Indices as Self-Supervised Objectives for Text Representation Learning4 Oct 2026
- 2026mEOL: Training-Free Instruction-Guided Multimodal Embedder for Vector Graphics and Image Retrieval18 Apr 2026
- 2026Bridging Linguistic Gaps: Cross-Lingual Mapping in Pre-Training and Dataset for Enhanced Multilingual LLM Performance12 Apr 2026
- 2026Pooling and Semantic Shift: The Fundamental Challenges in Long Text Embedding and Retrieval22 Mar 2026
- 2026ViCLSR: A Supervised Contrastive Learning Framework with Natural Language Inference for Natural Language Understanding Tasks22 Mar 2026
- 2026Towards Motion-aware Referring Image Segmentation18 Mar 2026
- 2026On Representation Redundancy in Large-Scale Instruction Tuning Data Selection14 Feb 2026
- 2026NEST: Nested Event Stream Transformer for Sequences of Multisets31 Jan 2026
- 2026Contextual Discrepancy-Aware Contrastive Learning for Robust Medical Time Series Diagnosis in Small-Sample Scenarios12 Jan 2026
- 2026SemPA: Improving Sentence Embeddings of Large Language Models through Semantic Preference Alignment8 Jan 2026
- 2025SAVeD: Semantic Aware Version Discovery21 Nov 2025
- 2025PatenTEB: A Comprehensive Benchmark and Model Family for Patent Text Embedding25 Oct 2025
38 more
- 2025Text2Token: Unsupervised Text Representation Learning with Token Target Prediction11 Oct 2025
- 2025Context-Aware Language Models for Forecasting Market Impact from Sequences of Financial News15 Sep 2025
- 2025SEA-BED: How Do Embedding Models Represent Southeast Asian Languages?17 Aug 2025
- 2025On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey28 Jul 2025
- 2025DALR: Dual-level Alignment Learning for Multimodal Sentence Representation Learning26 Jun 2025
- 2025Formalizing Learning from Language Feedback with Provable Guarantees12 Jun 2025
- 2025SAS-Bench: A Fine-Grained Benchmark for Evaluating Short Answer Scoring with Large Language Models12 May 2025
- 2025CSE-SFP: Enabling Unsupervised Sentence Representation Learning via a Single Forward Pass1 May 2025
- 2025Out-of-the-Box Conditional Text Embeddings from Large Language Models23 Apr 2025
- 2025Domain Adaptation for Japanese Sentence Embeddings with Contrastive Learning based on Synthetic Sentence Generation12 Mar 2025
- 2025OASIS: Order-Augmented Strategy for Improved Code Search11 Mar 2025
- 2025Refining Sentence Embedding Model through Ranking Sentences Generation with Large Language Models19 Feb 2025
- 2025Beyond Self-Consistency: Loss-Balanced Perturbation-Based Regularization Improves Industrial-Scale Ads Ranking5 Feb 2025
- 2024HNCSE: Advancing Sentence Embeddings via Hybrid Contrastive Learning with Hard Negatives19 Nov 2024
- 2024CmdCaliper: A Semantic-Aware Command-Line Embedding Model and Dataset for Security Research2 Nov 2024
- 2024UniHGKR: Unified Instruction-aware Heterogeneous Knowledge Retrievers26 Oct 2024
- 2024Self-Supervised Learning of Disentangled Representations for Multivariate Time-Series16 Oct 2024
- 2024Enhancing Unsupervised Sentence Embeddings via Knowledge-Driven Data Augmentation and Gaussian-Decayed Contrastive Learning19 Sep 2024
- 2024ConCSE: Unified Contrastive Learning and Augmentation for Code-Switched Embeddings28 Aug 2024
- 2024reCSE: Portable Reshaping Features for Sentence Embedding in Self-supervised Contrastive Learning9 Aug 2024
- 2024One Stone, Four Birds: A Comprehensive Solution for QA System Using Supervised Contrastive Learning12 Jul 2024
- 2024D2LLM: Decomposed and Distilled Large Language Models for Semantic Search25 Jun 2024
- 2024Space Decomposition for Sentence Embedding5 Jun 2024
- 2024Randomized Geometric Algebra Methods for Convex Neural Networks4 Jun 2024
- 2024Question-Based Retrieval using Atomic Units for Enterprise RAG20 May 2024
- 2024Unveiling Key Aspects of Fine-Tuning in Sentence Embeddings: A Representation Rank Analysis18 May 2024
- 2024SetCSE: Set Operations using Contrastive Learning of Sentence Embeddings25 Apr 2024
- 2024Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models18 Apr 2024
- 2024Event-enhanced Retrieval in Real-time Search9 Apr 2024
- 2024ContrastCAD: Contrastive Learning-based Representation Learning for Computer-Aided Design Models2 Apr 2024
- 2024AAdaM at SemEval-2024 Task 1: Augmentation and Adaptation for Multilingual Semantic Textual Relatedness1 Apr 2024
- 2024Making Sentence Embeddings Robust to User-Generated Content25 Mar 2024
- 2024Hyper-CL: Conditioning Sentence Representations with Hypernetworks14 Mar 2024
- 2024Surveying the Dead Minds: Historical-Psychological Text Analysis with Contextualized Construct Representation (CCR) for Classical Chinese1 Mar 2024
- 2024Meta-Task Prompting Elicits Embeddings from Large Language Models28 Feb 2024
- 2024IR2: Information Regularization for Information Retrieval25 Feb 2024
- 2024Fine-tuning CLIP Text Encoders with Two-step Paraphrasing23 Feb 2024
- 20242D Matryoshka Sentence Embeddings22 Feb 2024
50 newest of 132 All 132 (JSON)
- 2024Question-Based Retrieval using Atomic Units for Enterprise RAGcited by 17
- 2024Event-enhanced Retrieval in Real-time Searchcited by 1
- 2024Question-Based Retrieval using Atomic Units for Enterprise RAG20 May 2024
- 2024Event-enhanced Retrieval in Real-time Search9 Apr 2024
- 2022RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encodercited by 118
- 2023AnglE-optimized Text Embeddingscited by 99
- 2023Scaling Sentence Embeddings with Large Language Modelscited by 85
- 2023Exploring the Benefits of Training Expert Language Models over Instruction Tuningcited by 80
- 2023RaSa: Relation and Sensitivity Aware Representation Learning for Text-based Person Searchcited by 54
- 2023BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddingscited by 32
- 2022InfoCSE: Information-aggregated Contrastive Learning of Sentence Embeddingscited by 26
- 2023RankCSE: Unsupervised Sentence Representations Learning via Learning to Rankcited by 23
- 2022UnifieR: A Unified Retriever for Large-Scale Retrievalcited by 19
- 2022ContraCLM: Contrastive Learning For Causal Language Modelcited by 18
- 2024Meta-Task Prompting Elicits Embeddings from Large Language Modelscited by 16
- 2022Ranking-Enhanced Unsupervised Sentence Representation Learningcited by 15
12 most cited of 122 All 122 (JSON)
- 2026Cluster Validation Indices as Self-Supervised Objectives for Text Representation Learning4 Oct 2026
- 2026mEOL: Training-Free Instruction-Guided Multimodal Embedder for Vector Graphics and Image Retrieval18 Apr 2026
- 2026Bridging Linguistic Gaps: Cross-Lingual Mapping in Pre-Training and Dataset for Enhanced Multilingual LLM Performance12 Apr 2026
- 2026Pooling and Semantic Shift: The Fundamental Challenges in Long Text Embedding and Retrieval22 Mar 2026
- 2026ViCLSR: A Supervised Contrastive Learning Framework with Natural Language Inference for Natural Language Understanding Tasks22 Mar 2026
- 2026Towards Motion-aware Referring Image Segmentation18 Mar 2026
- 2026On Representation Redundancy in Large-Scale Instruction Tuning Data Selection14 Feb 2026
- 2026NEST: Nested Event Stream Transformer for Sequences of Multisets31 Jan 2026
- 2026Contextual Discrepancy-Aware Contrastive Learning for Robust Medical Time Series Diagnosis in Small-Sample Scenarios12 Jan 2026
- 2025SAVeD: Semantic Aware Version Discovery21 Nov 2025
- 2025PatenTEB: A Comprehensive Benchmark and Model Family for Patent Text Embedding25 Oct 2025
- 2025Text2Token: Unsupervised Text Representation Learning with Token Target Prediction11 Oct 2025
12 newest of 122 All 122 (JSON)