mlsanityData to 7 Oct 2026

Probing via Prompting

arXiv 2207.01736 ↗First posted 4 Jul 2022cs.CLcited by 7

Not on the map. Why?

Abstract (arXiv)

Probing is a popular method to discern what linguistic information is contained in the representations of pre-trained language models. However, the mechanism of selecting the probe model has recently been subject to intense debate, as it is not clear if the probes are merely extracting information or modeling the linguistic property themselves. To address this challenge, this paper introduces a novel model-free approach to probing, by formulating probing as a prompting task. We conduct experiments on five probing tasks and show that our approach is comparable or better at extracting information than diagnostic probes while learning much less on its own. We further combine the probing via prompting approach with attention head pruning to analyze where the model stores the linguistic information in its architecture. We then examine the usefulness of a specific linguistic property for pre-training by removing the heads that are essential to that property and evaluating the resulting model's performance on language modeling.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 54 of its 67 references that are in the map. First posted on arXiv 4 Jul 2022.

Builds on

54 of 67 references.

References · 54

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  4. 2020Language Models are Few-Shot Learnerscited by 32,199
  5. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  6. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  7. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  8. 2016Pointer Sentinel Mixture Modelscited by 2,992
  9. 2018Deep contextualized word representationscited by 2,285
  10. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
  11. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  12. 2019Language Models as Knowledge Bases?cited by 1,922
38 more
  1. 2016Understanding intermediate layers using linear classifier probescited by 1,279
  2. 2020Making Pre-trained Language Models Better Few-shot Learnerscited by 1,255
  3. 2019BERT Rediscovers the Classical NLP Pipelinecited by 1,046
  4. 2019What Does BERT Look At? An Analysis of BERT's Attentioncited by 994
  5. 2019How Can We Know What Language Models Know?cited by 953
  6. 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 934
  7. 2021Towards a Unified View of Parameter-Efficient Transfer Learningcited by 863
  8. 2019Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Prunedcited by 844
  9. 2019Are Sixteen Heads Really Better than One?cited by 803
  10. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 697
  11. 2017Learning Efficient Convolutional Networks through Network Slimmingcited by 673
  12. 2020It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learnerscited by 668
  13. 2019Designing and Interpreting Probes with Control Taskscited by 608
  14. 2017Learning Sparse Neural Networks through L0 Regularizationcited by 571
  15. 2017Learning multiple visual domains with residual adapterscited by 553
  16. 2018What you can cram into a single vector: Probing sentence embeddings for linguistic propertiescited by 452
  17. 2017ThiNet: A Filter Level Pruning Method for Deep Neural Network Compressioncited by 438
  18. 2019What do you learn from context? Probing for sentence structure in contextualized word representationscited by 436
  19. 2019Linguistic Knowledge and Transferability of Contextual Representationscited by 382
  20. 2021Learning How to Ask: Querying LMs with Mixtures of Soft Promptscited by 376
  21. 2021Towards Understanding and Mitigating Social Biases in Language Modelscited by 294
  22. 2021Factual Probing Is [MASK]: Learning vs. Learning to Recallcited by 291
  23. 2018Analysis Methods in Neural Language Processing: A Surveycited by 276
  24. 2020AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Promptscited by 246
  25. 2017End-to-end Neural Coreference Resolutioncited by 240
  26. 2020Information-Theoretic Probing with Minimum Description Lengthcited by 230
  27. 2021WARP: Word-level Adversarial ReProgrammingcited by 220
  28. 2016Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Taskscited by 216
  29. 2018Stress Test Evaluation for Natural Language Inferencecited by 211
  30. 2020Information-Theoretic Probing for Linguistic Structurecited by 204
  31. 2021Template-Based Named Entity Recognition Using BARTcited by 153
  32. 2018Dissecting Contextual Word Embeddings: Architecture and Representationcited by 146
  33. 2020Probing the Probing Paradigm: Does Probing Accuracy Entail Task Relevance?cited by 144
  34. 2018Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Informationcited by 97
  35. 2021BERTese: Learning to Speak to BERTcited by 82
  36. 2021Low-Complexity Probing via Finding Subnetworkscited by 63
  37. 2022Probing for the Usage of Grammatical Numbercited by 57
  38. 2018Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Taskscited by 56

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 3; 2024: 1.

7 papers.

  1. 2023Explainability for Large Language Models: A Surveycited by 331
  2. 2022Prompting PaLM for Translation: Assessing Strategies and Performancecited by 144
  3. 2022Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networkscited by 126
  4. 2022Prompting Language Models for Linguistic Structurecited by 39
  5. 2023Unsupervised Contrast-Consistent Ranking with Language Modelscited by 13
  6. 2023What do Language Models know about word senses? Zero-Shot WSD with Language Models and Domain Inventoriescited by 9
  7. 2024Linguistic Knowledge Can Enhance Encoder-Decoder Models (If You Let It)cited by 0