AI Paper Atlas

Research map / Large language models / Alignment and preferences

P-GenRM: Personalized Generative Reward Model with Test-time User-based Scaling

Pinyi Zhang, Ting-En Lin, Yuchuan Wu, Jingyang Chen, Zongqi Wang, Hua Yang, Xu Ze, Fei Huang, Yongbin Li, Kai Zhang

ICLR 2026 · Large language models › Alignment and preferences · approach: Reward model design and preference models · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add prototype-level scaling to individual-level scaling (Ind-16, Pro-8 vs Ind-32) (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 75.59 -> 75.92Improved
add test-time user-based scaling (Ind-16, Pro-8)P-GenRM 8B · PRISM-Personalizedaccuracy (%): 65.32 -> 68.06Improved
use prototype-level scaling alone (Ind-0, Pro-8) (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 72.68 -> 66.90Hurt
increase prototype-level samples from 8 to 16 with Ind-16 (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 75.92 -> 72.59Hurt
add curriculum learning (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 71.07 -> 72.68Improved
add process reward in RL (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 70.22 -> 71.07Improved
add outcome reward in RL (ablation)P-GenRM 8B · PRISM-Personalizedaccuracy (%): 60.94 -> 63.82Improved
add RL stage after SFT (ablation)P-GenRM 8B · PRISM-Personalizedaccuracy (%): 57.08 -> 63.82Improved
add SFT (persona-guided scoring induction) stage (ablation)P-GenRM 8B · Chatbot Arena-Personalizedaccuracy (%): 56.37 -> 66.76Improved
replace static SynthesizeMe persona with adaptive PSI (ablation)o3 judge · PRISM-Personalizedaccuracy (%): 58.49 -> 63.87Improved
replace SynthesizeMe with PSI (ablation)Qwen3-8B judge · Chatbot Arena-Personalizedaccuracy (%): 62.57 -> 64.22Improved
increase prototypes from 0 to 50 (Ind-8, Pro-4) (ablation)P-GenRM 8B · PRISM-Personalizedaccuracy (%): 65.32 -> 67.54Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy