Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal understanding, but their practical deployment is still limited by the substantial computation introduced by excessive visual tokens. Visual token pruning offers an effective solution by removing redundant visual tokens during inference...
Peng-Cheng Zheng, Chao-Ning Zhang, Ya Wen et al.· 0 citations
Large language models (LLMs) are increasingly used to discover scientific equations from experimental data. Such a discovery system has two parts that must work together: a library of executable computational tools and the model's tool-use policy. The tools supply the solving capability, from diagnosing structure in th...
Peng-Ze Li, Jia-Quan Zhang, En-Cheng Su et al.· 0 citations
This work proposes MoR-MLLM, a computation-sparse MLLM based on the recent Mixture-of-Recursions (MoR) framework, which introduces adaptive per-token recursion, allowing the model to dynamically adjust its recursive depth and allocate more computation to visually or linguistically challenging tokens while skipping redu...
Peng-Cheng Zheng, Chao-Ning Zhang, Jia-Xing Yan et al.· IEEE transactions on multime...· 1 citation
These findings establish correct-solution retention as a complementary objective of recursive self-improvement (RSI) and show that generation-time intervention can improve the solution repertoire retained by subsequent students.
Yun-Bo Long, Wen-Jie Chen, Jia-Quan Zhang et al.· 0 citations
A training-free decoding method that scores each masked position by a multiplicative composite of softmax confidence and neighbor proximity, promoting contextually ready tokens above isolated candidates while suppressing low-confidence positional noise, operating block-free to keep high-readiness anchors globally acces...
Yi-Kai Zhao, Qiyan Zhao, Jia-Quan Zhang et al.· 0 citations
A spatially adaptive neural operator (SANO), which replaces this spatially shared parameterization with a spatially continuous field of location-dependent operator parameters, which consistently outperforms competitive neural-operator, hypernetwork-based, and physics-informed baselines.
Jia-Quan Zhang, Chao-Ning Zhang, Shu-Xu Chen et al.· 0 citations
A geometry-aware incremental neural operator (GeoIncNO) for stable long-horizon PDE prediction and a mean--fluctuation decoupled reconstruction mechanism, where stable mean structures and dynamic fluctuations are fused separately, and phase correction is applied only to the zero-mean fluctuation component.
Jia-Quan Zhang, Shuxu Chen, Haifan Meng et al.· 0 citations
Experiments show that HERO consistently improves long-horizon accuracy, stable rollout length, and out-of-distribution robustness at no inference-time cost, indicating that history-enriched relative supervision is effective for stabilizing long-horizon autoregressive prediction.