Character Beyond Speech: Leveraging Role-Playing Evaluation in Large Audio Language Models via Reinforcement Learning
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 2044-2053
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (U25B2064)
- \"Pioneer\" and \"Leading Goose\" R&D Program of Zhejiang (2025C02110)
- Public Welfare Research Program of Ningbo (2024S062)
- Yongjiang Talent Project of Ningbo (2024A-161-G)
This publication has 9 references indexed in Scilit:
- Emphasizing Domain Differences Through Interactive-Augmented Prompts in Continual Audio-Visual Speech RecognitionIEEE Transactions on Image Processing, 2026
- VITA-Audio: Fast Interleaved Audio-Text Token Generation for Efficient Large Speech-Language Model2025
- PACHAT: Persona-Aware Speech Assistant for Multi-party Dialogue2025
- OmniCharacter: Towards Immersive Role-Playing Agents with Seamless Speech-Language Personality Interaction2025
- RolePlot: A Systematic Framework for Evaluating and Enhancing the Plot-Progression Capabilities of Role-Playing Agents2025
- EmoCharacter: Evaluating the Emotional Fidelity of Role-Playing Agents in Dialogues2025
- Boosting Speech Recognition Robustness to Modality-Distortion with Contrast-Augmented Prompts2024
- Character-LLM: A Trainable Agent for Role-Playing2023
- Contrastive Disentangled Meta-Learning for Signer-Independent Sign Language Translation2021