Research

Current Projects

STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts
We present an interpretable inference-time compute method that searches over high-level reasoning patterns. Instead of relying on high-temperature sampling, STATe uses discrete textual interventions: a controller selects actions encoding reasoning choices, a generator produces steps conditioned on those choices, and an evaluator scores candidates to guide search. This structured approach yields greater response diversity, interpretable action sequences that predict output quality, and the ability to steer generation toward promising regions of the action space.
→ Preprint available here
→ Software release here.

Understanding Message Virality in Conspiratorial Social Media Groups
We use supervised topic modelling to identify thematic drivers of message virality within online conspiratorial communities.

Provocations for Human-AI Interaction: A Community Forum Case Study
We are exploring user preferences and concerns based on results of the Meta Community Forum 2023 hosted by the Stanford Deliberative Democracy Lab.

Published Work

Till Raphael Saenger, Musashi Hinck, Justin Grimmer, and Brandon M. Stewart. 2024. AutoPersuade: A Framework for Evaluating and Explaining Persuasive Arguments. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 16325–16342, Miami, Florida, USA. Association for Computational Linguistics.
→ Associated Python package: suntopicmodel

Till Raphael Saenger, Ethan Kapstein, and Ronnie Sircar. 2024. Estimating the Collapse of Afghanistan’s Economy Using Nightlights Data. In PLOS ONE, 19(12): e0315337.

Previous Work

Till Raphael Saenger. Exploring the Cyclical Predictability of Sector-Specific Premia. 2020.

Till Raphael Saenger. On the Limitations of Cryptocurrencies. 2019.