Xinyu Mao
3 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The paper demonstrates that positional encodings are not necessary for transformers to achieve universal computation, showing that the inherent mechanism of sliding context windows already provides sufficient positional information.
The paper demonstrates that positional encodings are not necessary for transformers to achieve universal computation, showing that the inherent mechanism of sliding context windows already provides sufficient positional information.
This paper introduces CineCap, a framework for cinematographic captioning using structured reasoning, spatio-temporal anchors, and reinforcement learning.
Papers
CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning
Xinyu Mao, Yuhui Zeng, Xiaokun Liu, Wenyu Qin +5 more
This paper introduces CineCap, a framework for cinematographic captioning using structured reasoning, spatio-temporal anchors, and reinforcement learning.