🧑‍🔬 Thomas Icard

🏛️ Affiliations

🎓Stanford UniversityStanford · United States2008–2026 · 70 paperscurrent

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 1 more institution on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations (2023 · 81 citations)
   Causal abstraction tests whether a neural network faithfully implements a simpler, interpretable algorithm.
2. Interpretability at Scale: Identifying Causal Mechanisms in Alpaca (2023 · 81 citations)
   Causal-abstraction methods test whether a neural network actually implements a human-readable algorithm, but until now they had only been applied to small, task-specific models.

All works

14 works of theirs in our library, the most cited first; 2 with a page here, the others link their original until we summarise them.

1. Holistic Evaluation of Language Models ↗arXiv:2211.09110 · Published in Transactions on Machine Learning Research (TMLR), 2023 · no summary yetPercy Liang et al.2022 · 1,001 citations
2. On Pearl's Hierarchy and the Foundations of Causal Inference ↗Probabilistic and Causal Inference: The Works of Judea Pearl, ACM Books · no summary yetElias Bareinboim et al.2022 · 181 citations
3. Causal Abstractions of Neural Networks ↗arXiv:2106.02997 · no summary yetAtticus Geiger et al.2021 · 151 citations
4. Finding Alignments Between Interpretable Causal Variables and Distributed Neural RepresentationsarXiv:2303.02536Atticus Geiger, Zhengxuan Wu, Christopher Potts, et al.2023 · 81 citations
5. Interpretability at Scale: Identifying Causal Mechanisms in AlpacaConference on Neural Information Processing Systems (NeurIPS)Zhengxuan Wu, Atticus Geiger, Thomas Icard, et al.2023 · 81 citations
6. Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretability ↗arXiv:2301.04709 · no summary yetAtticus Geiger et al.2023 · 65 citations
7. Inducing Causal Structure for Interpretable Neural Networks ↗arXiv (Cornell University) · no summary yetAtticus Geiger et al.2021 · 53 citations
8. Language models cannot reliably distinguish belief from knowledge and fact ↗Nature Machine Intelligence · no summary yetMirac Süzgün et al.2025 · 28 citations
9. Causal Distillation for Language Models ↗Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · no summary yetZhengxuan Wu et al.2022 · 25 citations
10. Belief in the Machine: Investigating Epistemological Blind Spots of Language Models ↗arXiv (Cornell University) · no summary yetMirac Süzgün et al.2024 · 12 citations
11. A Reply to Makelov et al. (2023)'s "Interpretability Illusion" Arguments ↗arXiv (Cornell University) · no summary yetZhengxuan Wu et al.2024 · 5 citations
12. Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution Behaviors ↗arXiv (Cornell University) · no summary yetHuang, Jing et al.2025 · 3 citations
13. Transcoder Adapters for Reasoning-Model Diffing ↗arXiv (Cornell University) · no summary yetNathan Hu et al.2026
14. Evaluating Commercial AI Chatbots as News Intermediaries ↗arXiv (Cornell University) · no summary yetMirac Süzgün et al.2026

🤝 Co-authors here

4 co-authors with a page, the most papers together first.

• Atticus Geiger ×2
• Christopher E. Potts ×2
• Noah D. Goodman ×2
• Zhengxuan Wu ×2

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• Conference on Neural Information Processing Systems (NeurIPS): NeurIPS 2023
• Nature Machine Intelligence
• Probabilistic and Causal Inference: The Works of Judea Pearl, ACM Books
• Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

🔗 Cited by

Papers of our library cite their works 2,296 times (of the 149,439 papers whose bibliographies we read); 494 citing papers with a page here →.

Updated 2 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/thomas-icard/