# Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents

arXiv: https://arxiv.org/abs/2603.01438 · HTML: https://mlsanity.com/paper/2603.01438/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 32 of its 41 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 30 of 32)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Character-LLM: A Trainable Agent for Role-Playing](https://mlsanity.com/paper/2310.10158/) (2023)
- [From Persona to Personalization: A Survey on Role-Playing Language Agents](https://mlsanity.com/paper/2404.18231/) (2024)
- [RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models](https://mlsanity.com/paper/2310.00746/) (2023)
- [CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation](https://mlsanity.com/paper/2401.01275/) (2024)
- [Controlled Decoding from Language Models](https://mlsanity.com/paper/2310.17022/) (2023)
- [OASIS: Open Agent Social Interaction Simulations with One Million Agents](https://mlsanity.com/paper/2411.11581/) (2024)
- [Decoding-Time Language Model Alignment with Multiple Objectives](https://mlsanity.com/paper/2406.18853/) (2024)
- [SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents](https://mlsanity.com/paper/2403.08715/) (2024)
- [Unveiling the Truth and Facilitating Change: Towards Agent-based Large-scale Social Movement Simulation](https://mlsanity.com/paper/2402.16333/) (2024)
- [Decoding-time Realignment of Language Models](https://mlsanity.com/paper/2402.02992/) (2024)
- [Don't throw away your value model! Generating more preferable text with Value-Guided Monte-Carlo Tree Search decoding](https://mlsanity.com/paper/2309.15028/) (2023)
- [PAD: Personalized Alignment of LLMs at Decoding-Time](https://mlsanity.com/paper/2410.04070/) (2024)
- [ChatHaruhi: Reviving Anime Character in Reality via Large Language Model](https://mlsanity.com/paper/2308.09597/) (2023)
- [DeAL: Decoding-time Alignment for Large Language Models](https://mlsanity.com/paper/2402.06147/) (2024)
- [Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent](https://mlsanity.com/paper/2402.13717/) (2024)
- [GenSim: A General Social Simulation Platform with Large Language Model based Agents](https://mlsanity.com/paper/2410.04360/) (2024)
- [CharacterBox: Evaluating the Role-Playing Capabilities of LLMs in Text-Based Virtual Worlds](https://mlsanity.com/paper/2412.05631/) (2024)
- [Personality Alignment of Large Language Models](https://mlsanity.com/paper/2408.11779/) (2024)
- [ElectionSim: Massive Population Election Simulation Powered by Large Language Model Driven Agents](https://mlsanity.com/paper/2410.20746/) (2024)
- [Character is Destiny: Can Role-Playing Language Agents Make Persona-Driven Decisions?](https://mlsanity.com/paper/2404.12138/) (2024)
- [Enhancing Persona Consistency for LLMs' Role-Playing using Persona-Aware Contrastive Learning](https://mlsanity.com/paper/2503.17662/) (2025)
- [Social Science Meets LLMs: How Reliable Are Large Language Models in Social Simulations?](https://mlsanity.com/paper/2410.23426/) (2024)
- [Linear Alignment: A Closed-form Solution for Aligning Human Preferences without Tuning and Feedback](https://mlsanity.com/paper/2401.11458/) (2024)
- [Enhancing Personalized Dialogue Generation with Contrastive Latent Variables: Combining Sparse and Dense Persona](https://mlsanity.com/paper/2305.11482/) (2023)
- [TimeChara: Evaluating Point-in-Time Character Hallucination of Role-Playing Large Language Models](https://mlsanity.com/paper/2405.18027/) (2024)
- [MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents](https://mlsanity.com/paper/2408.04203/) (2024)
- [BEYOND DIALOGUE: A Profile-Dialogue Alignment Framework Towards General Role-Playing Language Model](https://mlsanity.com/paper/2408.10903/) (2024)

## Built on by (most cited first, 2 of 2)

- [PHASE-Tree: Modeling Character-State Evolution in Long-Horizon Role-Playing Dialogue](https://mlsanity.com/paper/2608.06975/) (2026)
- [Measuring and Improving Behavioral Consistency in Large Language Models through Fact-Heuristic-Emotion State Enforcement](https://mlsanity.com/paper/2607.24765/) (2026)

## Abstract (arXiv)

The utility of Role-Playing Language Agents in sociological research is growing alongside the adoption of Large Language Models. For realism in social simulation, these agents must adhere to their personas defined by character profiles, yet existing strategies-static prompt engineering or costly fine-tuning-fail to adapt personas to dynamic scenarios. Psychological theories, such as the Cognitive-Affective Personality Systems, provide a crucial explanation for this failure: a persona's influence on behavior is not static but varies with the scenarios. This context-dependence highlights the critical need for adaptive persona management. To address this gap, we propose a novel, theory-driven method that dynamically estimates context-dependent persona importance and integrates it into weighted reward-guided decoding, enabling inference-time persona following. Specifically, we introduce the Persona Dynamic Decoding (PDD) framework, which consists of two key components: (1) Persona Importance Estimation (PIE) module, which dynamically quantifies the contextual importance of persona attributes without requiring ground-truth supervision; and (2) Persona-Guided Inference-Time Alignment (PIA) paradigm, which leverages these importance scores to construct weighted multi-objective rewards and modulate generation probabilities during inference. Extensive experiments show the effectiveness of our method in utterance consistency and behavioral fidelity.