Modern vision–language models (VLMs) excel at many multimodal tasks, yet their grasp of temporal information in video remains weak and has not been adequately evaluated. We probe this gap with a deceptively simple but revealing challenge: judging the arrow of time (AoT)—whether a short clip is played forward or backward. We introduce AoT-PsyPhyBENCH, a psychophysically validated benchmark that tests whether VLMs can infer temporal direction in natural videos using the same stimuli and behavioral baselines established for humans. Our comprehensive evaluation of open-weight and proprietary, reasoning and non-reasoning VLMs reveals that most models perform near chance, and even the best model lags far behind human accuracy on physically irreversible processes (e.g., free fall, diffusion/explosion) and causal manual actions (division/addition) that humans recognize almost instantly. These results highlight a fundamental gap in current multimodal systems: while they capture rich visual–semantic correlations, they lack the inductive biases required for temporal continuity and causal understanding. We release the code and data for AoT-PsyPhyBENCH to encourage further progress in the physical and temporal reasoning capabilities of VLMs.

Existing in-context learning (ICL) methods for relation extraction (RE) often prioritize language similarity over structural similarity, which may result in overlooking entity relationships. We propose an AMR-enhanced retrieval-based ICL method for RE to address this issue. Our model retrieves in-context examples based on semantic structure similarity between task inputs and training samples. We conducted experiments in the supervised setting on four standard English RE datasets. The results show that our method achieves state-of-the-art performance on three datasets and competitive results on the fourth. Furthermore, our method outperforms baselines by a large margin across all datasets in the more demanding unsupervised setting.