Lulu Tang

I am a researcher at Beijing Academy of Artificial Intelligence (BAAI). I completed my postdoctoral research at BAAI and Tsinghua University, under the supervision of Prof. Tiejun Huang and Prof. Jiwen Lu. Prior to that, I obtained my Ph.D. degree from University of Macau, advised by Prof. Zhi-xin Yang and frequently collaborating with Prof. Kui Jia.

My research interests lie in the area of 3D computer vision, 3D generation models, and Vision-Language foundation models.

Email  |  GitHub  |  Google Scholar

(*: Equal Contribution, ♣: corresponding author)

See3D is a scalable visual-conditional MVD model for open-world 3D creation, which can be trained on web-scale video collections without camera pose annotations.

PiCo is a jailbreaking framework that bypasses advanced MLLM defenses using token-level typographic attacks to evade input filters and embeds malicious intent in programming instructions to avoid runtime monitoring.

TAP is a unified and promptable model capable of simultaneously segmenting, recognizing, and captioning arbitrary regions, with flexible visual prompts (point, box and sketch).

Point-BERT is a new paradigm for learning Transformers in an unsupervised manner by generalizing the concept of BERT onto 3D point cloud data.

Spike-T is a Transformer-based network for learning spike data and estimating monocular depth from continuous spike streams.