---
title: "Peter Hase — all works — Elamur Research"
description: "All 33 works of Peter Hase in the Elamur Research library, the most cited first: our summaries where we have them, else the original."
canonical: "https://research.elamur.ai/free/en/authors/peter-hase/works/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/peter-hase/works/)

[home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / [Peter Hase](https://md.research.elamur.ai/free/en/authors/peter-hase/) / all works

# Peter Hase — all works

33 works · [← Peter Hase](https://md.research.elamur.ai/free/en/authors/peter-hase/)

1. **[Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?](https://md.research.elamur.ai/free/en/papers/hase-2020-evaluating-explainable-ai-which-algorithmic/) [Annual Meeting of the Association for Computational Linguistics (ACL)](https://md.research.elamur.ai/free/en/venues/acl/)**Peter Hase, Mohit Bansal*2020 · 318 citations*
2. **[Rethinking machine unlearning for large language models ↗](https://doi.org/10.1038/s42256-025-00985-0) Nature Machine Intelligence · no summary yet**Sijia Liu et al.*2025 · 197 citations*
3. **[GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models](https://md.research.elamur.ai/free/en/papers/prasad-2023-grips-gradient-free-edit-based/) [Conference of the European Chapter of the Association for Computational Linguistics (EACL)](https://md.research.elamur.ai/free/en/venues/eacl/)**Archiki Prasad, Peter Hase, Xiang Zhou, et al.*2023 · 184 citations*
4. **[Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models ↗](https://arxiv.org/abs/2301.04213) arXiv:2301.04213 · no summary yet**Peter Hase et al.*2023 · 137 citations*
5. **[Interpretable Image Recognition with Hierarchical Prototypes ↗](https://doi.org/10.1609/hcomp.v7i1.5265) Proceedings of the AAAI Conference on Human Computation and Crowdsourcing · no summary yet**Peter Hase et al.*2019 · 131 citations*
6. **[Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?](https://md.research.elamur.ai/free/en/papers/hase-2020-leakage-adjusted-simulatability-can-models/) [Findings of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/findings/)**Peter Hase, Shiyue Zhang, Harry Xie, et al.*2020 · 126 citations*
7. **[Reasoning Models Don't Always Say What They Think ↗](https://arxiv.org/abs/2505.05410) arXiv:2505.05410 · no summary yet**Yanda Chen et al.*2025 · 106 citations*
8. **[FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging ↗](https://aclanthology.org/2021.emnlp-main.808/) Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · no summary yet**Han Guo et al.*2021 · 86 citations*
9. **[When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data ↗](https://aclanthology.org/2022.lnls-1.4/) Proceedings of the First Workshop on Learning with Natural Language Supervision · no summary yet**Peter Hase, Mohit Bansal*2022 · 80 citations*
10. **[Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs ↗](https://arxiv.org/abs/2111.13654) arXiv:2111.13654 · no summary yet**Peter Hase et al.*2021 · 60 citations*
11. **[Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks ↗](https://arxiv.org/abs/2309.17410) arXiv (Cornell University) · no summary yet**Vaidehi Patil, Peter Hase, Mohit Bansal*2023 · 56 citations*
12. **[Methods for Measuring, Updating, and Visualizing Factual Beliefs in Language Models](https://md.research.elamur.ai/free/en/papers/hase-2023-methods-measuring-updating-visualizing-factual/) [Conference of the European Chapter of the Association for Computational Linguistics (EACL)](https://md.research.elamur.ai/free/en/venues/eacl/)**Peter Hase, Mona Diab, Asli Celikyilmaz, et al.*2023 · 51 citations*
13. **[Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees ↗](https://arxiv.org/abs/2209.10492) arXiv (Cornell University) · no summary yet**Swarnadeep Saha et al.*2022 · 23 citations*
14. **[The Unreasonable Effectiveness of Easy Training Data for Hard Tasks ↗](https://aclanthology.org/2024.acl-long.378/) Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · no summary yet**Peter Hase et al.*2024 · 21 citations*
15. **[Can Language Models Teach Weaker Agents? Teacher Explanations Improve Students via Personalization ↗](https://arxiv.org/abs/2306.09299) arXiv:2306.09299 · no summary yet**Swarnadeep Saha, Peter Hase, Mohit Bansal*2023 · 20 citations*
16. **[The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanations ↗](https://arxiv.org/abs/2106.00786) arXiv (Cornell University) · no summary yet**Peter Hase, Harry Xie, Mohit Bansal*2021 · 18 citations*
17. **[Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations ↗](https://aclanthology.org/2022.emnlp-main.137/) Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · no summary yet**Swarnadeep Saha et al.*2022 · 17 citations*
18. **[Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs? ↗](https://arxiv.org/abs/2406.19354) arXiv (Cornell University) · no summary yet**Peter Hase et al.*2024 · 15 citations*
19. **[System-1.x: Learning to Balance Fast and Slow Planning with Language Models ↗](https://arxiv.org/abs/2407.14414) arXiv (Cornell University) · no summary yet**Swarnadeep Saha et al.*2024 · 13 citations*
20. **[LACIE: Listener-Aware Finetuning for Calibration in Large Language Models ↗](https://doi.org/10.52202/079017-1364) neural information processing systems · no summary yet**Elias Stengel-Eskin, Peter Hase, Mohit Bansal*2024 · 12 citations*
21. **[LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models ↗](https://arxiv.org/abs/2405.21028) arXiv (Cornell University) · no summary yet**Elias Stengel-Eskin, Peter Hase, Mohit Bansal*2024 · 12 citations*
22. **[VisFIS: Visual Feature Importance Supervision with Right-For-The-Right-Reason Objectives ↗](https://doi.org/10.52202/068431-1241) neural information processing systems · no summary yet**Zhuofan Ying, Peter Hase, Mohit Bansal*2022 · 11 citations*
23. **[Teaching Models to Balance Resisting and Accepting Persuasion ↗](https://aclanthology.org/2025.naacl-long.412/) Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · no summary yet**Elias Stengel-Eskin, Peter Hase, Mohit Bansal*2025 · 9 citations*
24. **[Can Language Models Teach? Teacher Explanations Improve Student Performance via Personalization ↗](https://doi.org/10.52202/075280-2746) neural information processing systems · no summary yet**Swarnadeep Saha, Peter Hase, Mohit Bansal*2023 · 8 citations*
25. **[Shall I Compare Thee to a Machine-Written Sonnet? An Approach to Algorithmic Sonnet Generation ↗](https://arxiv.org/abs/1811.05067) arXiv (Cornell University) · no summary yet**Benhardt, John et al.*2018 · 7 citations*
26. **[Are language models rational? The case of coherence norms and belief revision ↗](https://arxiv.org/abs/2406.03442) arXiv:2406.03442 · no summary yet**Thomas Hofweber et al.*2024 · 6 citations*
27. **[Unlearning Sensitive Information in Multimodal LLMs: Benchmark and Attack-Defense Evaluation ↗](https://arxiv.org/abs/2505.01456) arXiv (Cornell University) · no summary yet**Vaidehi Patil et al.*2025 · 3 citations*
28. **[Unsupervised Elicitation of Language Models ↗](https://arxiv.org/abs/2506.10139) arXiv (Cornell University) · no summary yet**Jiaxin Wen et al.*2025 · 2 citations*
29. **[Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization ↗](https://arxiv.org/abs/2608.31079) arXiv (Cornell University) · no summary yet**Camila Blank et al.*2026 · 1 citation*
30. **[Counterfactual Simulation Training for Chain-of-Thought Faithfulness ↗](https://arxiv.org/abs/2602.20710) arXiv (Cornell University) · no summary yet**Peter Hase, Christopher E. Potts*2026 · 1 citation*
31. **[Outcome Rewards Do Not Guarantee Verifiable or Causally Important Reasoning ↗](https://arxiv.org/abs/2604.22074) arXiv (Cornell University) · no summary yet**Qinan Yu et al.*2026 · 1 citation*
32. **[Adaptive Contextual Perception: How To Generalize To New Backgrounds and Ambiguous Objects ↗](https://doi.org/10.52202/075280-1951) neural information processing systems · no summary yet**Zhuofan Ying, Peter Hase, Mohit Bansal*2023 · 1 citation*
33. **[Low-Cost Algorithmic Recourse for Users With Uncertain Cost Functions ↗](https://arxiv.org/abs/2111.01235) arXiv (Cornell University) · no summary yet**Prateek Yadav, Peter Hase, Mohit Bansal*2021*