EUVIMEDEUVIMED entdecken
Crossref

VRAgent-R1: Boosting Video Recommendation with MLLM-based Agents via Reinforcement Learning

Siran Chen, Boyu Chen, Yuxiao Luo, Chenyun Yu, Yi Ouyang, Lei Cheng, Chengxiang Zhuo, Zang Li, Yali Wang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (3) · 1739-1747

DOI: 10.1609/aaai.v40i3.37152 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Siran Chen und 8 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large language model (LLM) agents have emerged as a promising solution for enhancing recommendation systems via user simulation. However, existing studies predominantly resort to prompt-based simulation using frozen LLMs, which frequently results in suboptimal item modeling and user preference learning, thereby ultimately constraining recommendation performance. To address these challenges, we introduce VRAgent-R1, a novel agent-based paradigm that incorporates human-like intelligence in user simulation. Specifically, VRAgent-R1 comprises two distinct agents: the Item Perception (IP) Agent and the User Simulation (US) Agent, designed for interactive user-item modeling. Firstly, the IP Agent emulates human-like progressive thinking based on MLLMs, effectively capturing hidden recommendation semantics in videos. With a more comprehensive multimodal content understanding provided by the IP Agent, the video recommendation system is equipped to provide higher-quality candidate items. Subsequently, the US Agent refines the recommended video sets based on in-depth chain-of-thought (CoT) reasoning and achieves better alignment with real user preferences through reinforcement learning. Experimental results on a large-scale video recommendation benchmark MicroLens-100k have demonstrated the effectiveness of our proposed VRAgent-R1 method, e.g., the IP Agent achieves a 6.0% improvement in NDCG@10, while the US Agent shows approximately 45.0% higher accuracy in user decision simulation compared to state-of-the-art baselines.

Zitieren

APA

Chen, S., Chen, B., Luo, Y., Yu, C., Ouyang, Y., Cheng, L., Zhuo, C., Li, Z., & Wang, Y. (2026). VRAgent-R1: Boosting Video Recommendation with MLLM-based Agents via Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(3), 1739-1747. https://doi.org/10.1609/aaai.v40i3.37152

Vancouver

Chen S, Chen B, Luo Y, Yu C, Ouyang Y, Cheng L, et al. VRAgent-R1: Boosting Video Recommendation with MLLM-based Agents via Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(3):1739-1747. doi:10.1609/aaai.v40i3.37152

BibTeX
@article{Chen2026,
  author = {Chen, Siran and Chen, Boyu and Luo, Yuxiao and Yu, Chenyun and Ouyang, Yi and Cheng, Lei and Zhuo, Chengxiang and Li, Zang and Wang, Yali},
  title = {VRAgent-R1: Boosting Video Recommendation with MLLM-based Agents via Reinforcement Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {3},
  pages = {1739--1747},
  doi = {10.1609/aaai.v40i3.37152}
}
Kostenlos starten →