SOTAVerified

Natural Language Inference

Natural language inference (NLI) is the task of determining whether a "hypothesis" is true (entailment), false (contradiction), or undetermined (neutral) given a "premise".

Example:

| Premise | Label | Hypothesis | | --- | ---| --- | | A man inspects the uniform of a figure in some East Asian country. | contradiction | The man is sleeping. | | An older and younger man smiling. | neutral | Two men are smiling and laughing at the cats playing on the floor. | | A soccer game with multiple males playing. | entailment | Some men are playing a sport. |

Approaches used for NLI include earlier symbolic and statistical approaches to more recent deep learning approaches. Benchmark datasets used for NLI include SNLI, MultiNLI, SciTail, among others. You can get hands-on practice on the SNLI task by following this d2l.ai chapter.

Further readings:

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 1961 papers

TitleStatusHype
LRCTI: A Large Language Model-Based Framework for Multi-Step Evidence Retrieval and Reasoning in Cyber Threat Intelligence Credibility Verification—0
DS@GT at CheckThat! 2025: Evaluating Context and Tokenization Strategies for Numerical Fact VerificationCode0
ARAG: Agentic Retrieval Augmented Generation for Personalized Recommendation—0
When Does Meaning Backfire? Investigating the Role of AMRs in NLI—0
Thunder-NUBench: A Benchmark for LLMs' Sentence-Level Negation Understanding—0
Explainable Compliance Detection with Multi-Hop Natural Language Inference on Assurance Case Structure—0
Theorem-of-Thought: A Multi-Agent Framework for Abductive, Deductive, and Inductive Reasoning in Language ModelsCode0
A MISMATCHED Benchmark for Scientific Natural Language InferenceCode0
CLATTER: Comprehensive Entailment Reasoning for Hallucination Detection—0
GATE: General Arabic Text Embedding for Enhanced Semantic Textual Similarity with Matryoshka Representation Learning and Hybrid Loss Training—0
Faithful and Robust LLM-Driven Theorem Proving for NLI Explanations—0
Drop Dropout on Single-Epoch Language Model PretrainingCode0
Preemptive Hallucination Reduction: An Input-Level Approach for Multimodal Language Model—0
LiTEx: A Linguistic Taxonomy of Explanations for Understanding Within-Label Variation in Natural Language InferenceCode0
S2LPP: Small-to-Large Prompt Prediction across LLMs—0
Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning—0
Learning to Reason via Mixture-of-Thought for Logical ReasoningCode1
DeFTX: Denoised Sparse Fine-Tuning for Zero-Shot Cross-Lingual Transfer—0
No Gold Standard, No Problem: Reference-Free Evaluation of Taxonomies—0
Boosting Neural Language Inference via Cascaded Interactive Reasoning—0
Document Attribution: Examining Citation Relationships using Large Language Models—0
Parameter-Efficient Transformer EmbeddingsCode0
Pushing the boundary on Natural Language Inference—0
Grounded in Context: Retrieval-Based Method for Hallucination Detection—0
SALAD: Improving Robustness and Generalization through Contrastive Learning with Structure-Aware and LLM-Driven Augmented Data—0
Cross-Document Cross-Lingual NLI via RST-Enhanced Graph Fusion and Interpretability Prediction—0
MedHal: An Evaluation Dataset for Medical Hallucination Detection—0
Negation: A Pink Elephant in the Large Language Models' Room?—0
HausaNLP at SemEval-2025 Task 3: Towards a Fine-Grained Model-Aware Hallucination Detection—0
Don't Fight Hallucinations, Use Them: Estimating Image Realism using NLI over Atomic FactsCode0
Am I eligible? Natural Language Inference for Clinical Trial Patient Recruitment: the Patient's Point of ViewCode0
Neutralizing Bias in LLM Reasoning using Entailment GraphsCode0
Collaboration is all you need: LLM Assisted Safe Code Translation—0
Introducing Verification Task of Set Consistency with Set-Consistency Energy Networks—0
ESNLIR: A Spanish Multi-Genre Dataset with Causal Relationships—0
Patient Trajectory Prediction: Integrating Clinical Notes with TransformersCode0
Giving AI Personalities Leads to More Human-Like Reasoning—0
Neuro-Symbolic Contrastive Learning for Cross-domain Inference—0
Beyond English: The Impact of Prompt Translation Strategies across Languages and Tasks in Multilingual LLMs—0
MorphNLI: A Stepwise Approach to Natural Language Inference Using Text Morphing—0
Does Training on Synthetic Data Make Models Less Robust?—0
Discourse-Driven Evaluation: Unveiling Factual Inconsistency in Long Document Summarization—0
Self-Rationalization in the Wild: A Large Scale Out-of-Distribution Evaluation on NLI-related tasksCode0
FactCG: Enhancing Fact Checkers with Graph-Based Multi-Hop DataCode1
TrustDataFilter:Leveraging Trusted Knowledge Base Data for More Effective Filtering of Unknown Information—0
A Study of the Plausibility of Attention between RNN Encoders in Natural Language Inference—0
Academic Case Reports Lack Diversity: Assessing the Presence and Diversity of Sociodemographic and Behavioral Factors related to Post COVID-19 Condition—0
Zero-shot and Few-shot Learning with Instruction-following LLMs for Claim Matching in Automated Fact-checking—0
Exploring Robustness of Multilingual LLMs on Real-World Noisy DataCode0
Entailed Between the Lines: Incorporating Implication into NLICode0
Show:102550
← PrevPage 1 of 40Next →
All datasetsSNLIRTEMultiNLIQNLIANLI testWNLILiDiRusRCBTERRaCommitmentBankSciTailFarsTail

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1UnitedSynT5 (3B)% Test Accuracy94.7—Unverified
2UnitedSynT5 (335M)% Test Accuracy93.5—Unverified
3EFL (Entailment as Few-shot Learner) + RoBERTa-large% Test Accuracy93.1—Unverified
4Neural Tree Indexers for Text Understanding% Test Accuracy93.1—Unverified
5RoBERTa-large + self-explaining layer% Test Accuracy92.3—Unverified
6RoBERTa-large+Self-Explaining% Test Accuracy92.3—Unverified
7CA-MTL% Test Accuracy92.1—Unverified
8SemBERT% Test Accuracy91.9—Unverified
9MT-DNN-SMARTLARGEv0% Test Accuracy91.7—Unverified
10MT-DNN-SMART_100%ofTrainingDataDev Accuracy91.6—Unverified
#ModelMetricClaimedVerifiedStatus
1Vega v2 6B (KD-based prompt transfer)Accuracy96—Unverified
2PaLM 540B (fine-tuned)Accuracy95.7—Unverified
3Turing NLR v5 XXL 5.4B (fine-tuned)Accuracy94.1—Unverified
4ST-MoE-32B 269B (fine-tuned)Accuracy93.5—Unverified
5DeBERTa-1.5BAccuracy93.2—Unverified
6MUPPET Roberta LargeAccuracy92.8—Unverified
7DeBERTaV3largeAccuracy92.7—Unverified
8T5-XXL 11B (fine-tuned)Accuracy92.5—Unverified
9T5-XXL 11BAccuracy92.5—Unverified
10UL2 20B (fine-tuned)Accuracy92.1—Unverified
#ModelMetricClaimedVerifiedStatus
1UnitedSynT5 (3B)Matched92.6—Unverified
2Turing NLR v5 XXL 5.4B (fine-tuned)Matched92.6—Unverified
3T5-XXL 11B (fine-tuned)Matched92—Unverified
4T5Matched92—Unverified
5T5-11BMismatched91.7—Unverified
6T5-3BMatched91.4—Unverified
7ALBERTMatched91.3—Unverified
8Adv-RoBERTa ensembleMatched91.1—Unverified
9DeBERTa (large)Matched91.1—Unverified
10SMARTRoBERTaDev Matched91.1—Unverified