SearcharxivSearch

arXiv subjects

Zi-Long Wang

Publications and source records attributed to Zi-Long Wang.

2 recordsLinked to original sources

String Model Predicts Relaxation in Sheared Glass-Forming Liquids

Understanding how structural relaxation evolves from equilibrium to nonequilibrium conditions remains a central problem in glass physics. Using simulations of model glass formers under steady shear, we show that the String Model, which links stringlike cooperative rearrangements to relaxation in equilibrium glass-forming liquids, predicts the structural relaxation time across the full range of temperatures and shear rates investigated without additional nonequilibrium fitting parameters, when combined with a shear-dependent effective temperature that is captured by a generalized fluctuation-dissipation relation. These results indicate that equilibrium and nonequilibrium relaxation are governed by the same underlying cooperative mechanism, but they occur under different effective thermodynamic conditions under steady shear. More broadly, our study provides a unified microscopic description of thermal and mechanically driven dynamics in glass-forming liquids.

cond-mat.soft

GI-Bench: A Panoramic Benchmark Revealing the Knowledge-Experience Dissociation of Multimodal Large Language Models in Gastrointestinal Endoscopy Against Clinical Standards

Multimodal Large Language Models (MLLMs) show promise in gastroenterology, yet their performance against comprehensive clinical workflows and human benchmarks remains unverified. To systematically evaluate state-of-the-art MLLMs across a panoramic gastrointestinal endoscopy workflow and determine their clinical utility compared with human endoscopists. We constructed GI-Bench, a benchmark encompassing 20 fine-grained lesion categories. Twelve MLLMs were evaluated across a five-stage clinical workflow: anatomical localization, lesion identification, diagnosis, findings description, and management. Model performance was benchmarked against three junior endoscopists and three residency trainees using Macro-F1, mean Intersection-over-Union (mIoU), and multi-dimensional Likert scale. Gemini-3-Pro achieved state-of-the-art performance. In diagnostic reasoning, top-tier models (Macro-F1 0.641) outperformed trainees (0.492) and rivaled junior endoscopists (0.727; p>0.05). However, a critical "spatial grounding bottleneck" persisted; human lesion localization (mIoU >0.506) significantly outperformed the best model (0.345; p<0.05). Furthermore, qualitative analysis revealed a "fluency-accuracy paradox": models generated reports with superior linguistic readability compared with humans (p<0.05) but exhibited significantly lower factual correctness (p<0.05) due to "over-interpretation" and hallucination of visual features. GI-Bench maintains a dynamic leaderboard that tracks the evolving performance of MLLMs in clinical endoscopy. The current rankings and benchmark results are available at https://roterdl.github.io/GIBench/.

cs.CV