7 papers
Beyond Myopic World Models: Long-Horizon End-to-End Training for Direct Future Prediction
Xinyi Li, Zaishuo Xia, Chenjie Hao +1
World models are expected to support imagination over extended temporal horizons, yet most are still trained through local few-step prediction objectives and deployed by recursivel…
Improving Weak World Models Behind Strong Agents in Atari Pong
Yukuan Lu, Zaishuo Xia, Weyl Lu +1
Strong world-model agents frequently contain weak world models. We study this agent-world-model gap by reproducing five visual world-model agents in Atari Pong: DreamerV3, DIAMOND,…
MotionPyramid: Hierarchical Motion Representation and Residual Interfaces
Gao Zhu, Zaishuo Xia, Yubei Chen
We ask whether the representational hierarchy seen in perception, from local primitives such as edges to higher level structures such as parts and objects, can be established for m…
How Do Large Language Models Learn Concepts During Continual Pre-Training?
Barry Menglong Yao, Sha Li, Yunzhi Yao +4
Human beings primarily understand the world through concepts (e.g., dog), abstract mental representations that structure perception, reasoning, and learning. However, how large lan…
SmallWorlds: Assessing Dynamics Understanding of World Models in Isolated Environments
Xinyi Li, Zaishuo Xia, Weyl Lu +2
Current world models lack a unified and controlled setting for systematic evaluation, making it difficult to assess whether they truly capture the underlying rules that govern envi…
Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models
Zaishuo Xia, Yukuan Lu, Xinyi Li +2
A world model is an internal model that simulates how the world evolves. Given past observations and actions, it predicts the future physical state of both the embodied agent and i…
Self-Supervised Learning for Building Damage Assessment from Large-scale xBD Satellite Imagery Benchmark Datasets
Zaishuo Xia, Zelin Li, Yanbing Bai +2
In the field of post-disaster assessment, for timely and accurate rescue and localization after a disaster, people need to know the location of damaged buildings. In deep learning,…