Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent

2024/02/21 by Xiaoyan Yu, Yu, Xiaoyan, Tongxu Luo +11 · 12 citations
Computer Science · Engineering · #Human Pose and Action Recognition #Human Motion and Animation #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2402.13717

Abstract

Large Language Models (LLMs) have revolutionized open-domain dialogue agents but encounter challenges in multi-character role-playing (MCRP) scenarios. To address the issue, we present Neeko, an innovative framework designed for efficient multiple characters imitation. Unlike existing methods, Neeko employs a dynamic low-rank adapter (LoRA) strategy, enabling it to adapt seamlessly to diverse characters. Our framework breaks down the role-playing process into agent pre-training, multiple characters playing, and character incremental learning, effectively handling both seen and unseen roles. This dynamic approach, coupled with distinct LoRA blocks for each character, enhances Neeko's adaptability to unique attributes, personalities, and speaking patterns. As a result, Neeko demonstrates superior performance in MCRP over most existing methods, offering more engaging and versatile user interaction experiences. Code and data are available at https://github.com/weiyifan1023/Neeko.

Cited by

• MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-Roles
• Causal-Guided Detoxify Backdoor Attack of Open-Weight LoRA Models
• Persistent Personas? Role-Playing, Instruction Following, and Safety in Extended Interactions
• Efficient Multi-Adapter LLM Serving via Cross-Model KV-Cache Reuse with Activated LoRA
• HyCoRA: Hyper-Contrastive Role-Adaptive Learning for Role-Playing
• BluePrint: A Social Media User Dataset for LLM Persona Evaluation and Training
• Unlocking the Effectiveness of LoRA-FP for Seamless Transfer Implantation of Fingerprints in Downstream Models
• Scaling Personality Control in LLMs with Big Five Scaler Prompts
• PersonaEval: Are LLM Evaluators Human Enough to Judge Role-Play?
• From Individuals to Crowds: Dual-Level Public Response Prediction in Social Media
• Test-Time-Matching: Decouple Personality, Memory, and Linguistic Style in LLM-based Role-Playing Language Agent
• MEraser: An Effective Fingerprint Erasure Approach for Large Language Models

Related

• A Multi-Task Role-Playing Agent Capable of Imitating Character Linguistic Styles
• AdaMARP: An Adaptive Multi-Agent Interaction Framework for General Immersive Role-Playing
• Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment
• OpenCharacter: Training Customizable Role-Playing LLMs with Large-Scale Synthetic Personas
• Test-Time-Matching: Decouple Personality, Memory, and Linguistic Style in LLM-based Role-Playing Language Agent