Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalization (2025 · 12 citations)
LLMs often guess who they are talking to from indirect hints, a behavior called implicit personalization. This can lower answer quality for some groups. The authors test how much stereotypes drive these guesses during multi-turn chats. They built template-based synthetic English conversations that separately control…
\textitOne Persona, \textitMany Cues, \textitDifferent Results: How Sociodemographic Cues Impact LLM Personalization (2026)
Studies of bias in LLM personalization usually tell the model who the user is through a single signal, or "persona cue," such as a name or a direct statement of the user's identity.
Actionable Interpretability for Churn Classification: A Text Bottleneck Model Case Study at a Major Telecom Provider (2026)
Telecom operators need to know why a customer wants to cancel, not only whether they do. Black-box LLM classifiers return only a label, which gives retention teams nothing to act on. This industry case study, carried out by the University of Amsterdam and KPN, applies the Text Bottleneck Model (TBM) of Ludan et al.…

Co-authors here

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

Venues

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

Papers of our library cite their works 139 times (of the 149,439 papers whose bibliographies we read); 43 citing papers with a page here →.

Updated 1 November 2026