In autonomous driving, perceiving surrounding objects is not enough: failing to understand how they move and interact over time can lead to dangerous decisions. Evaluating vision-language models (VLMs) therefore requires testing whether they connect past spatial observations to future consequences. Yet existing benchma...
Lie-Qi Liu, Rui Gao, Jia-Chen Gu et al.· 0 citations
Recent studies investigate the weight geometry through its spectral structure, which characterizes how the weights transform different input directions. Within this view, supervised fine-tuning (SFT) is shown to induce dense updates that perturb the spectral geometry of pre-SFT weights along their principal directions,...
Jian-Zhu Bao, Jia-Chen Gu, Qing-Yuan Liu et al.· 0 citations
It is argued that effective knowledge editing must account for the intricate nature of knowledge representation, and three promising research directions are proposed that respect the complexity of knowledge representation in a real-world setting.