Michael Saxon

Postdoctoral Scholar

mssaxon@uw.edu

Website ◆ CV

Specializations

Natural language processing
Multimodal models, including vision language models, speech, and diffusion systems
Multilingually and multiculturally representative systems

Research Areas

Biography

Michael Saxon is a Siegel Postdoctoral Fellow with the Tech Policy Lab and the Information School at the University of Washington. His research sits on the intersection of generative model benchmarking, multimodality, and AI ethics. He is particularly interested in difficult evaluation questions that arise in multimodal systems, and in developing methods to make systems performant and authentically user-responsive across languages and cultures. Saxon earned his bachelor’s in Electrical Engineering and master’s in Computer Engineering at Arizona State University, and his Ph.D. in Computer Science at the University of California, Santa Barbara, advised by William Wang.

Education

Ph D, Computer Science, University of California, Santa Barbara, 2025
MS, Computer Engineering, Arizona State University, 2020
BS, Electrical Engineering, Arizona State University, 2018

Awards

Neal Fenzi—Resonant Founder Fellowship - University of California, Santa Barbara, 2024
Rising Star in Generative AI - UMass Amherst Rising Stars Workshop, 2024
Outstanding Reviewer Award - ACL 2023, 2023
Center for Responsible Machine Learning Fellowship - University of California, Santa Barbara, 2020
Graduate Division Central Fellowship - University of California, Santa Barbara, 2020
National Science Foundation Graduate Research Fellowship - National Science Foundation (NSF GRFP), 2020

Publications and Contributions

Conference Paper

Can Vision Language Models Understand Mimes? (2025)
Findings of the Association for Computational Linguistics Authors: Justin Hyundong Cho, Spencer Lin, Tejas Srinivasan, Michael Saxon, Deuksin Kwon, Natali T. Chavez, Jonathan May
Conference Paper

Culture is Everywhere: A Call for Intentionally Cultural Evaluation (2025)
Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2025) Authors: Juhyun Oh, Inha Cha, Michael Saxon, Hyunseung Lim, Shaily Bhatt, Alice Oh
Conference Paper

Do You Know About My Nation? Investigating Multilingual Language Models’ Cultural Literacy Through Factual Knowledge (2025)
Conference on Empirical Methods in Natural Language Processing (EMNLP 2025) Authors: Eshaan Tanwar, Anwoy Chatterjee, Michael Saxon, Alon Albalak, William Yang Wang, Tanmoy Chakraborty
Conference Paper

TC-Bench: Benchmarking Temporal Compositionality in Text-to-Video and Image-to-Video Generation (2025)
Findings of the Association for Computational Linguistics Authors: Weixi Feng, Jiachen Li, Michael Saxon, Tsu-jui Fu, Wenhu Chen, William Yang Wang
Conference Paper

ThoughtTerminator: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models (2025)
Second Conference on Language Modeling (COLM 2025) Authors: Michael Saxon, Xiao Pu, Wenyue Hua, William Yang Wang
Conference Paper

VSP: Assessing the dual challenges of perception and reasoning in spatial planning tasks for VLMs (2025)
Proc. International Conference on Computer Vision (ICCV 2025) Authors: Qiucheng Wu, Handong Zhao, Michael Saxon, Trung Bui, William Yang Wang, Yang Zhang, Shiyu Chang
Journal Article, Academic Journal

Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies (2024)
Transactions of the Association for Computational Linguistics (TACL), 12(2024), pp. 484-506 Authors: Liangming Pan, Michael Saxon, Wenda Xu, Deepak Nathani, Xinyi Wang, William Yang Wang
Conference Paper

Benchmarks as Microscopes: A Call for Model Metrology (2024)
Proceedings of the Conference on Language Modeling (COLM 2024) Authors: Michael Saxon, Ari Holtzman, Peter West, William Yang Wang, Naomi Saphra
Conference Paper

Losing Visual Needles in Image Haystacks: Vision Language Models are Easily Distracted in Short and Long Contexts (2024)
Proceedings of the Empirical Methods in Natural Language Processing Conference (EMNLP 2024) Authors: Michael Saxon, Aditya Sharma, William Yang Wang
Conference Paper

Lost in Translation? Translation Errors and Challenges for Fair Assessment of Text-to-Image Models on Multilingual Concepts (2024)
Proceedings of the 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024) Authors: Michael Saxon, Yiran Luo, Sharon Levy, Chitta Baral, Yezhou Yang, William Yang Wang
Conference Paper

Who Evaluates the Evaluations? Assessing the Faithfulness and Consistency of Text-to-Image Evaluation Metrics with T2ISCORESCORE (2024)
Proceedings of the 38th Annual Conference on Neural Information Processing Systems (NeurIPS 2024) Authors: Michael Saxon, Mahsa Khoshnoodi, Fatima Jahara, Yujie Lu, Aditya Sharma, William Yang Wang
Conference Paper

Causal Balancing for Domain Generalization (2023)
Proceedings of the Eleventh International Conference on Learning Representations (ICLR 2023) Authors: Xinyi Wang, Michael Saxon, Jiachen Li, Hongyang Zhang, Kun Zhang, William Yang Wang
Conference Paper

CausalDialogue: Modeling Utterance-level Causality in Conversations (2023)
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023) Authors: Yi-lin Tuan, Alon Albalak, Wenda Xu, Michael Saxon, Connor Pryor, Lise Getoor, William Wang
Conference Workshop Paper

Data Augmentation for Diverse Voice Conversion in Noisy Environments (2023)
Interspeech 2023 Show and Tell Authors: Avani Tanna, Michael Saxon, Amr El Abbadi, William Yang Wang
Conference Paper

Large Language Models Are Implicitly Topic Models: Explaining and Finding Good Demonstrations for In-Context Learning (2023)
Proceedings of the 37th Annual Conference on Neural Information Processing Systems (NeurIPS 2023) Authors: Xinyi Wang, Wanrong Zhu, Michael Saxon, Mark Steyvers, William Yang Wang
Conference Paper

Let’s Think Frame by Frame with VIP: A Video Infilling and Prediction Dataset for Evaluating Video Chain-of-Thought (2023)
Proceedings of the Empirical Methods in Natural Language Processing Conference (EMNLP 2023) Authors: Vaishnavi Himakunthala, Andy Ouyang, Daniel Rose, Ryan He, Alex Mei, Yujie Lu, C. Sonar, Michael Saxon, William Yang Wang
Conference Paper

Multilingual Conceptual Coverage in Text-to-Image Models (2023)
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023) Authors: Michael Saxon, William Yang Wang
Conference Paper

PECO: Examining Single Sentence Label Leakage in Natural Language Inference Datasets (2023)
Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2023), pp. 3061-3074 Authors: Michael Saxon, Xinyi Wang, Wenda Xu, William Yang Wang
Conference Paper

WikiWhy: Answering and Explaining Cause-and-Effect Questions (2023)
Proceedings of the Eleventh International Conference on Learning Representations (ICLR 2023) Authors: Matthew Ho, Aditya Sharma, Justin Chang, Michael Saxon, Sharon Levy, Yujie Lu, William Yang Wang
Conference Paper

Not All Errors are Equal: Learning Text Generation Metrics using Stratified Error Synthesis (2022)
Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2022) Authors: Wenda Xu, Yilin Tuan, Yujie Lu, Michael Saxon, Lei Li, William Yang Wang
Conference Paper

Self-Supervised Knowledge Assimilation for Expert-Layman Style Transfer (2022)
Proceedings of the 36th edition of the Association for the Advancement of Artificial Intelligence Conference (AAAI 2022) Authors: Wenda Xu, Michael Saxon, Misha Sra, William Yang Wang
Conference Paper

Counterfactual Maximum Likelihood Estimation for Training Deep Networks (2021)
Proceedings of the Thirty-Fifth Annual Conference on Neural Information Processing Systems (NeurIPS 2021) Authors: Xinyi Wang, Wenhu Chen, Michael Saxon, William Yang Wang
Conference Paper

End-to-End Spoken Language Understanding for Generalized Voice Assistants (2021)
Interspeech 2021, pp. 4738-4742 Authors: Michael Saxon, Samridhi Choudhary, Joseph P. McKenna, Athanasios Mouchtaris
Conference Paper

Investigating Memorization of Conspiracy Theories in Text Generation (2021)
Findings of the Association for Computational Linguistics (ACL 2021) Authors: Sharon Levy, Michael Saxon, William Yang Wang
Conference Paper

Modeling Disclosive Transparency in NLP Application Descriptions (2021)
Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2021), pp. 2023-2037 Authors: Michael Saxon, Sharon Levy, Xinyi Wang, Alon Albalak, William Yang Wang
Journal Article, Academic Journal

Robust Estimation of Hypernasality in Dysarthria (2020)
IEEE Transactions on Audio, Speech, and Language Processing, pp. 2511-2522 Authors: Michael Saxon, Ayush Tripathi, Yishan Jiao, Julie Liss, Visar Berisha
Conference Paper

Semantic Complexity in End-to-End Spoken Language Understanding (2020)
Interspeech 2020, pp. 4273-4277 Authors: Michael Saxon, Joe McKenna, Samridhi Choudhary, Grant Strimel, Athanasios Mouchtaris
Conference Paper

UncommonVoice: A Crowdsourced Dataset of Dysphonic Speech (2020)
Interspeech 2020, pp. 2532-2536 Authors: Meredith Moore, P. Papreja, Michael Saxon, Visar Berisha, Sethuraman Panchanathan
Conference Paper

Objective Measures of Plosive Nasalization in Hypernasal Speech (2019)
Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (IEEE ICASSP 2019), pp. 6520-6524 Authors: Michael Saxon, Julie Liss, Visar Berisha
Conference Paper

Say what? A dataset for exploring the error patterns that two ASR engines make (2019)
Interspeech 2019, pp. 2528-2532 Authors: Meredith Moore, Michael Saxon, Hemanth Venkateswara, Visar Berisha, Sethuraman Panchanathan
Conference Workshop Paper

Word Pair Convolutional Model for Happy Moment Classification (2019)
AAAI AffCon Workshop 2019, pp. 111-119 Authors: Michael Saxon, S. Bhandari, L. Ruskin, G. Honda
Conference Paper

2D Grating Pitch Mapping of a through Silicon Via (TSV) and Solder Ball Interconnect Region Using Laser Diffraction (2016)
Proceedings of the IEEE 66th Electronic Components and Technology Conference (ECTC 2016), pp. 2222-2227 Authors: Todd Houghton, Michael Saxon, Zeming Song, Hoa Nyugen, Hanqing Jiang, Hongbin Yu

Presentations

How to nitpick multimodal evaluations (2025)
CVPR 2025 - Virtual
Multilingual multimodal evaluation: how and why (2025)
Google Translate Research - Mountain View, CA
Rigorous measurement in text-to-image systems (2024)
UMD CLIP Seminar - College Park MD
Rigorous measurement in text-to-image systems (2024)
Stanford SALT Group - Palo Alto, CA
Rigorous measurement in text-to-image systems (2024)
Georgetown University - Washington DC
Disparities in Text-to-Image Model Conceptual Knowledge Across Languages (2023)
2023 ACM Conference on Fairness, Accountability, and Transparency (FAccT) - Chicago, IL
Rigorous measurement in text-to-image systems (2023)
Arizona State University - Tempe, AZ
Rigorous measurement in text-to-image systems (2023)
USC Information Sciences Institute - Marina Del Rey, CA