Research Notes

On surprise, human–AI teaming, and temporal networks.

Unleashing the Power of Absence and Surprise

Can a machine be surprised? Learning via Surprisability teaches AI systems to form expectations and adjust when reality diverges.

1 min read · November 03, 2025

2025 · surprisability interpretable-ml · research
Lab Celebration: Eran's RecSys 2025 Paper!

PAIRSAT combines scarce satisfaction labels with abundant preference data to estimate user satisfaction in dialogue systems.

2 min read · September 14, 2025

2025 · recommender-systems human-ai-teaming · announcements
Paper Alert: What Travelers Say vs. What They Rate

Across 137,000 hotel reviews, textual sentiment often diverges from star ratings — especially for solo travelers.

2 min read · May 04, 2025

2025 · reviews emotions e-commerce · research
Congratulations to Roi Alfassi — Part of His Master's Thesis Highlighted at HAI-GEN, IUI'2025!

Online Storytelling Spaces — participants' perceptions of overt and covert AI agents.

1 min read · May 04, 2025

2025 · human-ai-teaming storytelling · announcements
From Feelings to Fingerprints: A Talk at the Department of Communication Colloquium

Two methods for analyzing online discourse — emotional signatures, and personal signatures via Learning via Surprisability.

2 min read · April 29, 2025

2025 · talks emotions surprisability · announcements