Showing 2026Show all
3 papers · 1 filter
cs.AI2026
Persona Following Is Not Selective Control: The Neutrality Gap in LLM User Simulation
Jiashen Ren, Wenlin Zhang, Bohan Zhang +5
Persona prompting is widely used to construct user simulations with large language models (LLMs), yet it relies on a largely untested assumption: specifying one user attribute shou…
cs.LG2026
Weightless Fine-Tuning: Personalizing LLMs via Logit-Space Transport
Bohan Zhang, Anqi Ni, Yixin Wang +1
Supervised fine-tuning (SFT) is a standard approach for adapting LLMs to a target distribution, but in settings such as personalization, where each author requires separate weight…
cs.HC2026
Who Owns the Text? Design Patterns for Preserving Authorship in AI-Assisted Writing
Bohan Zhang, Chengke Bu, Paramveer S. Dhillon
AI writing assistants can reduce effort and improve fluency, but they may also weaken writers' sense of authorship. We study this tension with an ownership-aware co-writing editor…