Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Sparse Personalized Text Generation with Multi-Trajectory Reasoning

About

As Large Language Models (LLMs) advance, personalization has become a key mechanism for tailoring outputs to individual user needs. However, most existing methods rely heavily on dense interaction histories, making them ineffective in cold-start scenarios where such data is sparse or unavailable. While external signals (e.g., content of similar users) can offer a potential remedy, leveraging them effectively remains challenging: raw context is often noisy, and existing methods struggle to reason over heterogeneous data sources. To address these issues, we introduce PAT (Personalization with Aligned Trajectories), a reasoning framework for cold-start LLM personalization. PAT first retrieves information along two complementary trajectories: writing-style cues from stylistically similar users and topic-specific context from preference-aligned users. It then employs a reinforcement learning-based, iterative dual-reasoning mechanism that enables the LLM to jointly refine and integrate these signals. Experimental results across real-world personalization benchmarks show that PAT consistently improves generation quality and alignment under sparse-data conditions, establishing a strong solution to the cold-start personalization problem.

Bo Ni, Haowei Fu, Qinwen Ge, Franck Dernoncourt, Samyadeep Basu, Nedim Lipka, Seunghyun Yoon, Yu Wang, Nesreen K. Ahmed, Subhojyoti Mukherjee, Puneet Mathur, Ryan A. Rossi, Tyler Derr• 2026

Related benchmarks

TaskDatasetResultRank
Long Text GenerationStylized Feedback Generation Benchmark
R-124.2
8
Long Text GenerationHotel Experience
ROUGE-10.277
8
Long Text GenerationAmazon Review
ROUGE-121
8
Short-text generationStylized Feedback Generation Benchmark
ROUGE-1 Score0.191
8
Short-text generationAmazon Review
ROUGE-117.7
8
Short-text generationHotel Experience
ROUGE-113.3
8
Long Text GenerationPersonalization Benchmark Long Text
R-1 Score0.233
4
Short-text generationPersonalization Benchmark Short Text
R-10.157
4
Showing 8 of 8 rows

Other info

Follow for update