SOTAVerified

Semantic Textual Similarity

Semantic textual similarity deals with determining how similar two pieces of texts are. This can take the form of assigning a score from 1 to 5. Related tasks are paraphrase or duplicate identification.

Image source: Learning Semantic Textual Similarity from Conversations

Papers

Showing 150 of 2381 papers

TitleStatusHype
Rethinking the Sample Relations for Few-Shot ClassificationCode7
LLM.int8(): 8-bit Matrix Multiplication for Transformers at ScaleCode5
2D Matryoshka Sentence EmbeddingsCode4
AlignScore: Evaluating Factual Consistency with a Unified Alignment FunctionCode4
One Embedder, Any Task: Instruction-Finetuned Text EmbeddingsCode4
MTEB: Massive Text Embedding BenchmarkCode4
Automatically Interpreting Millions of Features in Large Language ModelsCode3
ERNIE 2.0: A Continual Pre-training Framework for Language UnderstandingCode3
ERNIE: Enhanced Representation through Knowledge IntegrationCode3
BERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingCode3
InstructSAM: A Training-Free Framework for Instruction-Oriented Remote Sensing Object RecognitionCode2
FinMTEB: Finance Massive Text Embedding BenchmarkCode2
Reasoning to Attend: Try to Understand How <SEG> Token WorksCode2
Squeezed Attention: Accelerating Long Context Length LLM InferenceCode2
Large Continual Instruction AssistantCode2
Linguistic Minimal Pairs Elicit Linguistic Similarity in Large Language ModelsCode2
beeFormer: Bridging the Gap Between Semantic and Interaction Similarity in Recommender SystemsCode2
Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented GenerationCode2
Weakly-supervised Audio Separation via Bi-modal Semantic SimilarityCode2
EAGLE: Eigen Aggregation Learning for Object-Centric Unsupervised Semantic SegmentationCode2
PeFoMed: Parameter Efficient Fine-tuning of Multimodal Large Language Models for Medical ImagingCode2
BeLLM: Backward Dependency Enhanced Large Language Model for Sentence EmbeddingsCode2
SEINE: Short-to-Long Video Diffusion Model for Generative Transition and PredictionCode2
AnglE-optimized Text EmbeddingsCode2
DiffCSE: Difference-based Contrastive Learning for Sentence EmbeddingsCode2
PromptBERT: Improving BERT Sentence Embeddings with PromptsCode2
SimCSE: Simple Contrastive Learning of Sentence EmbeddingsCode2
Top2Vec: Distributed Representations of TopicsCode2
DeBERTa: Decoding-enhanced BERT with Disentangled AttentionCode2
Exploring the Limits of Transfer Learning with a Unified Text-to-Text TransformerCode2
ALBERT: A Lite BERT for Self-supervised Learning of Language RepresentationsCode2
Inv-Entropy: A Fully Probabilistic Framework for Uncertainty Quantification in Language ModelsCode1
IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response TheoryCode1
Label-Guided In-Context Learning for Named Entity RecognitionCode1
The Avengers: A Simple Recipe for Uniting Smaller Language Models to Challenge Proprietary GiantsCode1
Smoothie: Smoothing Diffusion on Token Embeddings for Text GenerationCode1
R2MED: A Benchmark for Reasoning-Driven Medical RetrievalCode1
One-Step Offline Distillation of Diffusion-based Models via Koopman ModelingCode1
ELITE: Embedding-Less retrieval with Iterative Text ExplorationCode1
CDF-RAG: Causal Dynamic Feedback for Adaptive Retrieval-Augmented GenerationCode1
High Temporal Consistency through Semantic Similarity Propagation in Semi-Supervised Video Semantic Segmentation for Autonomous FlightCode1
SGC-Net: Stratified Granular Comparison Network for Open-Vocabulary HOI DetectionCode1
Mitigating Hallucinations in Large Vision-Language Models by Adaptively Constraining Information FlowCode1
MedFILIP: Medical Fine-grained Language-Image Pre-trainingCode1
DiffSim: Taming Diffusion Models for Evaluating Visual SimilarityCode1
DuSSS: Dual Semantic Similarity-Supervised Vision-Language Model for Semi-Supervised Medical Image SegmentationCode1
Familiarity: Better Evaluation of Zero-Shot Named Entity Recognition by Quantifying Label Shifts in Synthetic Training DataCode1
Vid-Morp: Video Moment Retrieval Pretraining from Unlabeled Videos in the WildCode1
RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable DataCode1
Semantic-Aware Resource Management for C-V2X Platooning via Multi-Agent Reinforcement LearningCode1
Show:102550
← PrevPage 1 of 48Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SMARTRoBERTaDev Pearson Correlation92.8Unverified
2DeBERTa (large)Accuracy92.5Unverified
3SMART-BERTDev Pearson Correlation90Unverified
4MT-DNN-SMARTPearson Correlation0.93Unverified
5StructBERTRoBERTa ensemblePearson Correlation0.93Unverified
6Mnet-SimPearson Correlation0.93Unverified
7XLNet (single model)Pearson Correlation0.93Unverified
8T5-11BPearson Correlation0.93Unverified
9ALBERTPearson Correlation0.93Unverified
10RoBERTaPearson Correlation0.92Unverified
#ModelMetricClaimedVerifiedStatus
1AnglE-UAESpearman Correlation84.54Unverified
2ST5-XXLSpearman Correlation82.63Unverified
3ST5-LargeSpearman Correlation81.83Unverified
4ST5-XLSpearman Correlation81.66Unverified
5ST5-BaseSpearman Correlation81.14Unverified
6MPNet-multilingualSpearman Correlation80.73Unverified
7SGPT-5.8B-nliSpearman Correlation80.53Unverified
8MPNetSpearman Correlation80.28Unverified
9MiniLM-L12Spearman Correlation79.8Unverified
10SimCSE-BERT-supSpearman Correlation79.12Unverified
#ModelMetricClaimedVerifiedStatus
1MT-DNN-SMARTAccuracy93.7Unverified
2ALBERTAccuracy93.4Unverified
3RoBERTa (ensemble)Accuracy92.3Unverified
4BigBirdF191.5Unverified
5StructBERTRoBERTa ensembleAccuracy91.5Unverified
6FLOATER-largeAccuracy91.4Unverified
7SMARTAccuracy91.3Unverified
8RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned)Accuracy91Unverified
9RoBERTa-large 355M + Entailment as Few-shot LearnerF191Unverified
10SpanBERTAccuracy90.9Unverified
#ModelMetricClaimedVerifiedStatus
1PromCSE-RoBERTa-large (0.355B)Spearman Correlation0.82Unverified
2PromptEOL+CSE+LLaMA-30BSpearman Correlation0.82Unverified
3PromptEOL+CSE+OPT-13BSpearman Correlation0.82Unverified
4SimCSE-RoBERTalargeSpearman Correlation0.82Unverified
5PromptEOL+CSE+OPT-2.7BSpearman Correlation0.81Unverified
6SentenceBERTSpearman Correlation0.75Unverified
7SRoBERTa-NLI-baseSpearman Correlation0.74Unverified
8SRoBERTa-NLI-largeSpearman Correlation0.74Unverified
9Dino (STS/̄🦕)Spearman Correlation0.74Unverified
10SBERT-NLI-largeSpearman Correlation0.74Unverified
#ModelMetricClaimedVerifiedStatus
1AnglE-LLaMA-7BSpearman Correlation0.91Unverified
2AnglE-LLaMA-7B-v2Spearman Correlation0.91Unverified
3PromptEOL+CSE+LLaMA-30BSpearman Correlation0.9Unverified
4PromptEOL+CSE+OPT-13BSpearman Correlation0.9Unverified
5PromptEOL+CSE+OPT-2.7BSpearman Correlation0.9Unverified
6PromCSE-RoBERTa-large (0.355B)Spearman Correlation0.89Unverified
7Trans-Encoder-BERT-large-bi (unsup.)Spearman Correlation0.89Unverified
8Trans-Encoder-BERT-large-cross (unsup.)Spearman Correlation0.88Unverified
9Trans-Encoder-RoBERTa-large-cross (unsup.)Spearman Correlation0.88Unverified
10SimCSE-RoBERTa-largeSpearman Correlation0.87Unverified