1 | 余天予https://yutianyu.cn/zh/publication-type/1/1Wowchemy (https://wowchemy.com)zh-HansThu, 14 May 2026 00:00:00 +0000https://yutianyu.cn/media/icon_hu0ee3417c3693759e075e80aecf96b070_11607_512x512_fill_lanczos_center_3.png1https://yutianyu.cn/zh/publication-type/1/Deep Pre-Alignment for VLMsThu, 14 May 2026 00:00:00 +0000MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training RecipeTue, 16 Sep 2025 00:00:00 +0000RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V TrustworthinessMon, 27 May 2024 00:00:00 +0000RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human FeedbackFri, 01 Dec 2023 00:00:00 +0000Cross-Modal Omni Interaction Modeling for Phrase GroundingThu, 01 Oct 2020 00:00:00 +0000