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.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Long Text Generation | Stylized Feedback Generation Benchmark | R-124.2 | 8 | |
| Long Text Generation | Hotel Experience | ROUGE-10.277 | 8 | |
| Long Text Generation | Amazon Review | ROUGE-121 | 8 | |
| Short-text generation | Stylized Feedback Generation Benchmark | ROUGE-1 Score0.191 | 8 | |
| Short-text generation | Amazon Review | ROUGE-117.7 | 8 | |
| Short-text generation | Hotel Experience | ROUGE-113.3 | 8 | |
| Long Text Generation | Personalization Benchmark Long Text | R-1 Score0.233 | 4 | |
| Short-text generation | Personalization Benchmark Short Text | R-10.157 | 4 |