Publications (37)
Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigm
Yichen Xie, Han Lu, Junchi Yan +3
Given the large-scale data and the high annotation cost, pretraining-finetuning becomes a popular paradigm in multiple computer vision tasks. Previous research has covered both the…
Rethinking k-means from manifold learning perspective
Quanxue Gao, Qianqian Wang, Han Lu +2
Although numerous clustering algorithms have been developed, many existing methods still leverage k-means technique to detect clusters of data points. However, the performance of k…
Gemini: A Family of Highly Capable Multimodal Models
Gemini Team, Rohan Anil, Sebastian Borgeaud +1340
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consist…
Continual Predictive Learning from Videos
Geng Chen, Wendong Zhang, Han Lu +4
Predictive learning ideally builds the world model of physical processes in one or more given environments. Typical setups assume that we can collect data from all environments at…
Part II: ROLL Flash -- Accelerating RLVR and Agentic Training with Asynchrony
Han Lu, Zichen Liu, Shaopan Xiong +19
Synchronous Reinforcement Learning (RL) post-training has emerged as a crucial step for enhancing Large Language Models (LLMs) with diverse capabilities. However, many systems desi…
Augmenting Transformer-Transducer Based Speaker Change Detection With Token-Level Training Loss
Guanlong Zhao, Quan Wang, Han Lu +2
In this work we propose a novel token-based training strategy that improves Transformer-Transducer (T-T) based speaker change detection (SCD) performance. The conventional T-T base…