6 papers
MiniWorld: Democratizing the Training of Video World Models from Scratch
Yian Zhao, Ruochong Zheng, Hongcan Guo +3
Video world models predict future observations conditioned on historical observations and control signals, enabling long-horizon generation through autoregressive state transitions…
PosIR: Position-Aware Heterogeneous Information Retrieval Benchmark
Ziyang Zeng, Dun Zhang, Yu Yan +4
In real-world documents, the information relevant to a user query may reside anywhere from the beginning to the end. This makes position bias -- a systematic tendency of retrieval…
OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
Feilong Tang, Xiang An, Yunyao Yan +16
Hypothesis. Artificial general intelligence is, at its core, a compression problem. Effective compression demands resonance: deep learning scales best when its architecture aligns…
Watch Closely: Mitigating Object Hallucinations in Large Vision-Language Models with Disentangled Decoding
Ruiqi Ma, Yu Yan, Chunhong Zhang +4
Large Vision-Language Models (LVLMs) bridge the gap between visual and linguistic modalities, demonstrating strong potential across a variety of domains. However, despite significa…
RT-DETRv4: Painlessly Furthering Real-Time Object Detection with Vision Foundation Models
Zijun Liao, Yian Zhao, Xin Shan +5
Real-time object detection has achieved substantial progress through meticulously designed architectures and optimization strategies. However, the pursuit of high-speed inference v…
Sigma: Differential Rescaling of Query, Key and Value for Efficient Language Models
Zhenghao Lin, Zihao Tang, Xiao Liu +31
We introduce Sigma, an efficient large language model specialized for the system domain, empowered by a novel architecture including DiffQKV attention, and pre-trained on our metic…