---
title: "BlackboxNLP Workshop on Analyzing and Interpreting Neural"
description: "All 43 papers of BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP 2021, the most cited first; those with our…"
canonical: "https://research.elamur.ai/free/en/venues/blackboxnlp/2021/papers/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/venues/blackboxnlp/2021/papers/)

[home](https://md.research.elamur.ai/free/en/) / [venues](https://md.research.elamur.ai/free/en/venues/) / [BlackboxNLP Workshop on Analyzing and…](https://md.research.elamur.ai/free/en/venues/blackboxnlp/) / [2021](https://md.research.elamur.ai/free/en/venues/blackboxnlp/2021/) / papers

# BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP 2021: all accepted papers

43 papers · the most cited first · [← BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP 2021](https://md.research.elamur.ai/free/en/venues/blackboxnlp/2021/)

1. **[Fine-Tuned Transformers Show Clusters of Similar Representations Across Layers ↗](https://aclanthology.org/2021.blackboxnlp-1.42/)**Jason Phang, Haokun Liu, Samuel R. Bowman*22 citations*
2. **[Perturbing Inputs for Fragile Interpretations in Deep Natural Language Processing ↗](https://aclanthology.org/2021.blackboxnlp-1.33/)**Sanchit Sinha, Hanjie Chen, Arshdeep Sekhon et al.*20 citations*
3. **[To what extent do human explanations of model behavior align with actual model behavior? ↗](https://aclanthology.org/2021.blackboxnlp-1.1/)**Grusha Prasad, Yixin Nie, Mohit Bansal et al.*18 citations*
4. **[Enhancing Interpretable Clauses Semantically using Pretrained Word Representation ↗](https://aclanthology.org/2021.blackboxnlp-1.19/)**Rohan Kumar Yadav, Lei Jiao, Ole-Christoffer Granmo et al.*17 citations*
5. **[Probing Language Models for Understanding of Temporal Expressions ↗](https://aclanthology.org/2021.blackboxnlp-1.31/)**Shivin Thukral, Kunal Kukreja, Christian Kavouras*17 citations*
6. **[Can Transformers Jump Around Right in Natural Language? Assessing Performance Transfer from SCAN ↗](https://aclanthology.org/2021.blackboxnlp-1.9/)**Rahma Chaabouni, Roberto Dessì, Eugene Kharitonov*15 citations*
7. **[Analyzing BERT's Knowledge of Hypernymy via Prompting ↗](https://aclanthology.org/2021.blackboxnlp-1.20/)**Michael Hanna, David Mareček*14 citations*
8. **[Do Language Models Know the Way to Rome? ↗](https://aclanthology.org/2021.blackboxnlp-1.40/)**Bastien Liétard, Mostafa Abdou, Anders Søgaard*14 citations*
9. **[A howling success or a working sea? Testing what BERT knows about metaphors ↗](https://aclanthology.org/2021.blackboxnlp-1.13/)**Paolo Pedinotti, Eliana Di Palma, Ludovica Cerini et al.*13 citations*
10. **[ALL Dolphins Are Intelligent and SOME Are Friendly: Probing BERT for Nouns' Semantic Properties and their Prototypicality ↗](https://aclanthology.org/2021.blackboxnlp-1.7/)**Marianna Apidianaki, Aina Garí Soler*13 citations*
11. **[How Does BERT Rerank Passages? An Attribution Analysis with Information Bottlenecks ↗](https://aclanthology.org/2021.blackboxnlp-1.39/)**Zhiying Jiang, Raphael Tang, Ji Xin et al.*11 citations*
12. **[The Language Model Understood the Prompt was Ambiguous: Probing Syntactic Uncertainty Through Generation ↗](https://aclanthology.org/2021.blackboxnlp-1.4/)**Laura Aina, Tal Linzen*11 citations*
13. **[Not All Models Localize Linguistic Knowledge in the Same Place: A Layer-wise Probing on BERToids' Representations ↗](https://aclanthology.org/2021.blackboxnlp-1.29/)**Mohsen Fayyaz, Ehsan Aghazadeh, Ali Modarressi et al.*10 citations*
14. **[Investigating Negation in Pre-trained Vision-and-language Models ↗](https://aclanthology.org/2021.blackboxnlp-1.27/)**Radina Dobreva, Frank Keller*9 citations*
15. **[On the Language-specificity of Multilingual BERT and the Impact of Fine-tuning ↗](https://aclanthology.org/2021.blackboxnlp-1.15/)**Marc Tanti, Lonneke van der Plas, Claudia Borg et al.*9 citations*
16. **[Does External Knowledge Help Explainable Natural Language Inference? Automatic Evaluation vs. Human Ratings](https://md.research.elamur.ai/free/en/papers/schuff-2021-does-external-knowledge-help-explainable/)**Hendrik Schuff, Hsiu-Yu Yang, Heike Adel et al.*8 citations · our summary*
17. **[Transferring Knowledge from Vision to Language: How to Achieve it and how to Measure it? ↗](https://aclanthology.org/2021.blackboxnlp-1.10/)**Tobias Norlund, Lovisa Hagström, Richard Johansson*8 citations*
18. **[Not all parameters are born equal: Attention is mostly what you need ↗](https://aclanthology.org/2021.blackboxnlp-1.28/)**Nikolay Bogoychev*7 citations*
19. **[Relating Neural Text Degeneration to Exposure Bias ↗](https://aclanthology.org/2021.blackboxnlp-1.16/)**Ting-Rui Chiang, Yun-Nung Chen*7 citations*
20. **[BERT Has Uncommon Sense: Similarity Ranking for Word Sense BERTology ↗](https://aclanthology.org/2021.blackboxnlp-1.43/)**Luke Gessler, Nathan Schneider*6 citations*
21. **[How Length Prediction Influence the Performance of Non-Autoregressive Translation? ↗](https://aclanthology.org/2021.blackboxnlp-1.14/)**Minghan Wang, Guo Jiaxin, Yuxia Wang et al.*6 citations*
22. **[Screening Gender Transfer in Neural Machine Translation ↗](https://aclanthology.org/2021.blackboxnlp-1.24/)**Guillaume Wisniewski, Lichao Zhu, Nicolas Ballier et al.*6 citations*
23. **[Assessing the Generalization Capacity of Pre-trained Language Models through Japanese Adversarial Natural Language Inference ↗](https://aclanthology.org/2021.blackboxnlp-1.26/)**Hitomi Yanaka, Koji Mineshima*5 citations*
24. **[How Familiar Does That Sound? Cross-Lingual Representational Similarity Analysis of Acoustic Word Embeddings ↗](https://aclanthology.org/2021.blackboxnlp-1.32/)**Badr Abdullah, Iuliia Zaitova, Tania Avgustinova et al.*5 citations*
25. **[Learning Mathematical Properties of Integers ↗](https://aclanthology.org/2021.blackboxnlp-1.30/)**Maria Ryskina, Kevin Knight*5 citations*
26. **[Discrete representations in neural models of spoken language ↗](https://aclanthology.org/2021.blackboxnlp-1.11/)**Bertrand Higy, Lieke Gelderloos, Afra Alishahi et al.*4 citations*
27. **[The Acceptability Delta Criterion: Testing Knowledge of Language using the Gradience of Sentence Acceptability ↗](https://aclanthology.org/2021.blackboxnlp-1.38/)**Héctor Javier Vázquez Martínez*4 citations*
28. **[What BERT Based Language Model Learns in Spoken Transcripts: An Empirical Study ↗](https://aclanthology.org/2021.blackboxnlp-1.25/)**Ayush Kumar, Mukuntha Narayanan Sundararaman, Jithendra Vepa*4 citations*
29. **[Word Equations: Inherently Interpretable Sparse Word Embeddings through Sparse Coding ↗](https://aclanthology.org/2021.blackboxnlp-1.12/)**Adly Templeton*4 citations*
30. **[Attacks against Ranking Algorithms with Text Embeddings: A Case Study on Recruitment Algorithms ↗](https://aclanthology.org/2021.blackboxnlp-1.36/)**Anahita Samadi, Debapriya Banerjee, Shirin Nilizadeh*3 citations*
31. **[Multi-Layer Random Perturbation Training for improving Model Generalization Efficiently ↗](https://aclanthology.org/2021.blackboxnlp-1.23/)**Lis Kanashiro Pereira, Yuki Taya, Ichiro Kobayashi*3 citations*
32. **[What Models Know About Their Attackers: Deriving Attacker Information From Latent Representations ↗](https://aclanthology.org/2021.blackboxnlp-1.6/)**Zhouhang Xie, Jonathan Brophy, Adam Noack et al.*3 citations*
33. **[An Investigation of Language Model Interpretability via Sentence Editing ↗](https://aclanthology.org/2021.blackboxnlp-1.34/)**Samuel Stevens, Yu Su*2 citations*
34. **[Exploratory Model Analysis Using Data-Driven Neuron Representations ↗](https://aclanthology.org/2021.blackboxnlp-1.41/)**Daisuke Oba, Naoki Yoshinaga, Masashi Toyoda*2 citations*
35. **[Test Harder than You Train: Probing with Extrapolation Splits ↗](https://aclanthology.org/2021.blackboxnlp-1.2/)**Jenny Kunz, Marco Kuhlmann*2 citations*
36. **[Efficient Explanations from Empirical Explainers ↗](https://aclanthology.org/2021.blackboxnlp-1.17/)**Robert Schwarzenberg, Nils Feldhus, Sebastian Möller*1 citation*
37. **[Interacting Knowledge Sources, Inspection and Analysis: Case-studies on Biomedical text processing ↗](https://aclanthology.org/2021.blackboxnlp-1.35/)**Parsa Bagherzadeh, Sabine Bergler*1 citation*
38. **[ProSPer: Probing Human and Neural Network Language Model Understanding of Spatial Perspective ↗](https://aclanthology.org/2021.blackboxnlp-1.8/)**Tessa Masis, Carolyn Jane Anderson*1 citation*
39. **[An in-depth look at Euclidean disk embeddings for structure preserving parsing ↗](https://aclanthology.org/2021.blackboxnlp-1.21/)**Federico Fancellu, Lan Xiao, Allan Jepson et al.
40. **[Controlled tasks for model analysis: Retrieving discrete information from sequences ↗](https://aclanthology.org/2021.blackboxnlp-1.37/)**Ionut-Teodor Sorodoc, Gemma Boleda, Marco Baroni
41. **[On the Limits of Minimal Pairs in Contrastive Evaluation ↗](https://aclanthology.org/2021.blackboxnlp-1.5/)**Jannis Vamvas, Rico Sennrich
42. **[Training Dynamic based data filtering may not work for NLP datasets ↗](https://aclanthology.org/2021.blackboxnlp-1.22/)**Arka Talukdar, Monika Dagar, Prachi Gupta et al.
43. **[Variation and generality in encoding of syntactic anomaly information in sentence embeddings ↗](https://aclanthology.org/2021.blackboxnlp-1.18/)**Qinxuan Wu, Allyson Ettinger
- [← back to](https://md.research.elamur.ai/free/en/venues/blackboxnlp/2021/) — BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP 2021