🧑‍🔬 Michael Saxon

🏛️ Affiliations

🎓University of WashingtonSeattle · United States2024–2026 · 4 paperscurrent🎓University of Southern CaliforniaLos Angeles · United States2026 · 3 paperscurrent🎓University of California, Santa BarbaraSanta Barbara · United States2020–2025 · 38 papersEducation since 2020 · Doctor of Philosophy, Electrical and Computer Engineering🎓Arizona State UniversityTempe · United States2014–2020 · 6 papersEducation 2014–2018 · Bachelor's of Science in Engineering, Electrical, Computer, and Energy EngineeringEducation 2018–2020 · Master's of Science, Electrical, Computer, and Energy Engineering

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 more institutions 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. Automatically Correcting Large Language Models: Surveying the Landscape of Diverse Automated Correction Strategies (2024 · 140 citations)
   Large language models still make characteristic mistakes. They invent facts, reach conclusions that do not follow from their own reasoning, and sometimes produce toxic or biased text. Fixing these flaws with human feedback (e.g., RLHF) is expensive and cannot happen in real time. This TACL survey therefore focuses on…
2. Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning (2023 · 116 citations)
   In-context learning in LLMs depends heavily on which few-shot demonstrations are chosen. Existing theories of why it works have mostly been tested on synthetic data and toy Transformers. This paper treats real-world LLMs as implicit latent variable models, similar to topic models. In this view, the prompt lets the…
3. PECO: Examining Single Sentence Label Leakage in Natural Language Inference Datasets through Progressive Evaluation of Cluster Outliers (2023 · 9 citations)
   NLI is meant to be a pairwise task: the relation between premise and hypothesis should depend on both sentences together.
4. CAIRE: Cultural Attribution of Images with Retrieval (2026 · 1 citation)
   Text-to-image systems often depict cultures unevenly, favouring Western imagery or producing stereotyped and inaccurate pictures.

All works

7 works of theirs in our library, the most cited first; 4 with a page here, the others link their original until we summarise them.

1. Automatically Correcting Large Language Models: Surveying the Landscape of Diverse Automated Correction StrategiesTransactions of the Association for Computational Linguistics (TACL)Liangming Pan, Michael Saxon, Wenda Xu, et al.2024 · 140 citations
2. Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context LearningConference on Neural Information Processing Systems (NeurIPS)Xinyi Wang, Wanrong Zhu, Michael Saxon, et al.2023 · 116 citations
3. Self-Supervised Knowledge Assimilation for Expert-Layman Text Style Transfer ↗arXiv:2110.02950 · no summary yetWenda Xu et al.2021 · 11 citations
4. PECO: Examining Single Sentence Label Leakage in Natural Language Inference Datasets through Progressive Evaluation of Cluster OutliersConference of the European Chapter of the Association for Computational Linguistics (EACL)Michael Saxon, Xinyi Wang, Wenda Xu, et al.2023 · 9 citations
5. Lost in Translation? Translation Errors and Challenges for Fair Assessment of Text-to-Image Models on Multilingual Concepts ↗Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers) · no summary yetMichael Saxon et al.2024 · 5 citations
6. CAIRE: Cultural Attribution of Images with RetrievalConference of the European Chapter of the Association for Computational Linguistics (EACL)Arnav Yayavaram, Siddharth Yayavaram, Simran Khanuja, et al.2026 · 1 citation
7. CAIRe: Cultural Attribution of Images by Retrieval-Augmented Evaluation ↗arXiv (Cornell University) · no summary yetArnav Yayavaram et al.2025

🤝 Co-authors here

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

• William Yang Wang ×3
• Xinyi Wang ×3
• Wenda Xu ×2
• Deepak Nathani
• Graham Neubig
• Liangming Pan
• Simran Khanuja
• Wanrong Zhu

Venues

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

• Conference of the European Chapter of the Association for Computational Linguistics (EACL) ×2: EACL 2026, EACL 2023
• Conference on Neural Information Processing Systems (NeurIPS): NeurIPS 2023
• Transactions of the Association for Computational Linguistics (TACL): Transactions of the Association for Computational Linguistics (TACL) 2024
• Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)

🔗 Cited by

Papers of our library cite their works 156 times (of the 149,439 papers whose bibliographies we read); 28 citing papers with a page here →.

Updated 31 October 2026

↑ back to top

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