G-UBS: Towards Robust Understanding of Implicit Feedback via Group-Aware User Behavior Simulation

2025-08-07

AI Summary

To address the challenges of high noise and ambiguous user intent in implicit feedback—limiting recommendation performance—this paper proposes Group-aware User Behavior Simulation (GUBS), a novel paradigm. GUBS introduces the first LLM-based “summarize–cluster–reflect” user grouping pipeline, integrating multimodal large language models (MLLMs) with group-aware reinforcement learning to achieve deep denoising of implicit behaviors (e.g., skips) and robust preference intent inference. It constructs group-level user profiles via multimodal data fusion and leverages these aggregated representations to enhance individual-level modeling. Evaluated on our newly established IF-VR benchmark—a multimodal video recommendation dataset grounded in implicit feedback—GUBS achieves a 4.0% improvement in video playback rate (>30 seconds) and a 14.9% gain in implicit feedback intent inference accuracy, significantly outperforming state-of-the-art LLM- and MLLM-based baselines.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Video Understanding & Activity Analysis
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

User feedback is critical for refining recommendation systems, yet explicit feedback (e.g., likes or dislikes) remains scarce in practice. As a more feasible alternative, inferring user preferences from massive implicit feedback has shown great potential (e.g., a user quickly skipping a recommended video usually indicates disinterest). Unfortunately, implicit feedback is often noisy: a user might skip a video due to accidental clicks or other reasons, rather than disliking it. Such noise can easily misjudge user interests, thereby undermining recommendation performance. To address this issue, we propose a novel Group-aware User Behavior Simulation (G-UBS) paradigm, which leverages contextual guidance from relevant user groups, enabling robust and in-depth interpretation of implicit feedback for individual users. Specifically, G-UBS operates via two key agents. First, the User Group Manager (UGM) effectively clusters users to generate group profiles utilizing a ``summarize-cluster-reflect" workflow based on LLMs. Second, the User Feedback Modeler (UFM) employs an innovative group-aware reinforcement learning approach, where each user is guided by the associated group profiles during the reinforcement learning process, allowing UFM to robustly and deeply examine the reasons behind implicit feedback. To assess our G-UBS paradigm, we have constructed a Video Recommendation benchmark with Implicit Feedback (IF-VR). To the best of our knowledge, this is the first multi-modal benchmark for implicit feedback evaluation in video recommendation, encompassing 15k users, 25k videos, and 933k interaction records with implicit feedback. Extensive experiments on IF-VR demonstrate that G-UBS significantly outperforms mainstream LLMs and MLLMs, with a 4.0% higher proportion of videos achieving a play rate > 30% and 14.9% higher reasoning accuracy on IF-VR.

Problem

• Addressing noise in implicit feedback for accurate user preference inference
• Enhancing recommendation systems via group-aware user behavior simulation
• Improving robustness in interpreting implicit feedback using contextual group guidance

Innovation

• Group-aware User Behavior Simulation (G-UBS) paradigm
• User Group Manager clusters users via LLM workflow
• Group-aware reinforcement learning for feedback modeling