Multi-Personality Generation of LLMs at Decoding-time
Abstract
A novel framework for multi-personality generation in LLMs uses decoding-time combination and speculative chunk-level rejection sampling to efficiently generate high-quality responses with multiple personalization attributes.
Multi-personality generation for LLMs, enabling simultaneous embodiment of multiple personalization attributes, is a fundamental challenge. Existing retraining-based approaches are costly and poorly scalable, while decoding-time methods often rely on external models or heuristics, limiting flexibility and robustness. In this paper, we propose a novel Multi-Personality Generation (MPG) framework under the decoding-time combination paradigm. It flexibly controls multi-personality without relying on scarce multi-dimensional models or extra training, leveraging implicit density ratios in single-dimensional models as a "free lunch" to reformulate the task as sampling from a target strategy aggregating these ratios. To implement MPG efficiently, we design Speculative Chunk-level based Rejection sampling (SCR), which generates responses in chunks and parallelly validates them via estimated thresholds within a sliding window. This significantly reduces computational overhead while maintaining high-quality generation. Experiments on MBTI personality and Role-Playing demonstrate the effectiveness of MPG, showing improvements up to 16%-18%. Code and data are available at https://github.com/Libra117/MPG .
Models citing this paper 0
No model linking this paper
Datasets citing this paper 6
Browse 6 datasets citing this paperSpaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper