mlsanityData to 14 Oct 2026

Rainproof: An Umbrella To Shield Text Generators From Out-Of-Distribution Data

arXiv 2212.09171 ↗First posted 18 Dec 2022cs.CLcited by 21

Not on the map. Why?

Abstract (arXiv)

Implementing effective control mechanisms to ensure the proper functioning and security of deployed NLP models, from translation to chatbots, is essential. A key ingredient to ensure safe system behaviour is Out-Of-Distribution (OOD) detection, which aims to detect whether an input sample is statistically far from the training distribution. Although OOD detection is a widely covered topic in classification tasks, most methods rely on hidden features output by the encoder. In this work, we focus on leveraging soft-probabilities in a black-box framework, i.e. we can access the soft-predictions but not the internal states of the model. Our contributions include: (i) RAINPROOF a Relative informAItioN Projection OOD detection framework; and (ii) a more operational evaluation setting for OOD detection. Surprisingly, we find that OOD detection is not necessarily aligned with task-specific measures. The OOD detector may filter out samples well processed by the model and keep samples that are not, leading to weaker performance. Our results show that RAINPROOF provides OOD detection methods more aligned with task-specific performance metrics than traditional OOD detectors.

Key facts. Cited by 21 papers in the map (data through 14 Oct 2026). Builds on 61 of its 108 references that are in the map. First posted on arXiv 18 Dec 2022.

Builds on

61 of 108 references.

References · 61

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  4. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  5. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,984
  6. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  7. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,328
  8. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  9. 2020Energy-based Out-of-distribution Detectioncited by 1,299
  10. 2019PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarizationcited by 1,102
  11. 2022Red Teaming Language Models with Language Modelscited by 1,026
  12. 2019DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generationcited by 842
38 more
  1. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 722
  2. 2017DailyDialog: A Manually Labelled Multi-turn Dialogue Datasetcited by 652
  3. 2018MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversationscited by 643
  4. 2020Beyond English-Centric Multilingual Machine Translationcited by 561
  5. 2019Defending Against Neural Fake Newscited by 559
  6. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  7. 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
  8. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 345
  9. 2021On the Importance of Gradients for Detecting Distributional Shifts in the Wildcited by 339
  10. 2019HopSkipJumpAttack: A Query-Efficient Decision-Based Attackcited by 333
  11. 2018Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problemcited by 323
  12. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 221
  13. 2021A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detectioncited by 207
  14. 2020MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselinescited by 204
  15. 2019Facebook FAIR's WMT19 News Translation Task Submissioncited by 179
  16. 2019Sparse Sequence-to-Sequence Modelscited by 170
  17. 2020The Tatoeba Translation Challenge -- Realistic Data Sets for Low Resource and Multilingual MTcited by 128
  18. 2021Types of Out-of-Distribution Texts and How to Detect Themcited by 80
  19. 2021Contrastive Out-of-Distribution Detection for Pretrained Transformerscited by 71
  20. 2021Revisiting Mahalanobis Distance for Transformer-Based Out-of-Domain Detectioncited by 70
  21. 2019Affect-Driven Dialog Generationcited by 50
  22. 2019Out-of-distribution Detection in Classifiers via Generationcited by 49
  23. 2020Feature Space Singularity for Out-of-Distribution Detectioncited by 49
  24. 2022Open-environment Machine Learningcited by 48
  25. 2021Automatic Text Evaluation through the Lens of Wasserstein Barycenterscited by 38
  26. 2021A Novel Estimator of Mutual Information for Learning to Disentangle Textual Representationscited by 37
  27. 2020Hierarchical Pre-training for Sequence Labelling in Spoken Dialogcited by 36
  28. 2021Open-world Machine Learning: Applications, Challenges, and Opportunitiescited by 34
  29. 2021InfoLM: A New Metric to Evaluate Summarization & Data2Text Generationcited by 33
  30. 2020Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformerscited by 31
  31. 2021A statistical framework for efficient out of distribution detection in deep neural networkscited by 28
  32. 2020DQI: Measuring Data Quality in NLPcited by 24
  33. 2018Multi-Modal Emotion recognition on IEMOCAP Dataset using Deep Learningcited by 21
  34. 2020Heavy-tailed Representations, Text Polarity Classification & Data Augmentationcited by 20
  35. 2022Igeood: An Information Geometry Approach to Out-of-Distribution Detectioncited by 20
  36. 2021Adversarial Robustness via Fisher-Rao Regularizationcited by 19
  37. 2018Neural Sentence Embedding using Only In-domain Sentences for Out-of-domain Sentence Detection in Dialog Systemscited by 18
  38. 2021BERTTune: Fine-Tuning Neural Machine Translation with BERTScorecited by 18

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 4; 2024: 7; 2025: 5; 2026: 5.

21 papers.

Sort:

By area:

  1. 2024Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraphcited by 105
  2. 2024SaulLM-54B & SaulLM-141B: Scaling Up Domain Adaptation for the Legal Domaincited by 39
  3. 2025Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Homecited by 33
  4. 2025TokUR: Token-Level Uncertainty Estimation for Large Language Model Reasoningcited by 24
  5. 2025Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Modelscited by 18
  6. 2023Unsupervised Layer-wise Score Aggregation for Textual OOD Detectioncited by 15
  7. 2023Towards More Robust NLP System Evaluation: Handling Missing Scores in Benchmarkscited by 14
  8. 2024SAUP: Situation Awareness Uncertainty Propagation on LLM Agentcited by 14
  9. 2024Enhanced Hallucination Detection in Neural Machine Translation through Simple Detector Aggregationcited by 9
  10. 2023Transductive Learning for Textual Few-Shot Classification in API-based Embedding Modelscited by 4
  11. 2024WeiPer: OOD Detection using Weight Perturbations of Class Projectionscited by 4
  12. 2024What If the Input is Expanded in OOD Detection?cited by 4
9 more
  1. 2023Toward Stronger Textual Attack Detectorscited by 3
  2. 2026Semantic Calibration Prevails Where Token Confidence Fails: Benchmarking Long-Form Scientific QAcited by 3
  3. 2025Is Your Model Fairly Certain? Uncertainty-Aware Fairness Evaluation for LLMscited by 2
  4. 2024Towards Trustworthy Reranking: A Simple yet Effective Abstention Mechanismcited by 1
  5. 2025ALIEN: Aligned Entropy Head for Improving Uncertainty Estimation of LLMscited by 1
  6. 2026Towards Reliable Medical LLMs: Benchmarking and Enhancing Confidence Estimation of Large Language Models in Medical Consultationcited by 0
  7. 2026How Do Uncertainty Estimators Align with Hallucination Occurrence in LLMs?cited by 0
  8. 2026GUT: Quantifying and Optimizing the Reasoning Uncertainty of LLMs via Graph Complexitycited by 0
  9. 2026Characterizing Overconfident Failure in LLM-Based Code Generationcited by 0