The ReLE framework introduces a dynamic, variance‑aware evaluation system that diagnoses capability anisotropy across 304 Chinese large language models, exposing ranking instability, commercial‑vs‑open‑source gaps, and format barriers while cutting evaluation cost by 70%.
Huolala shares its LaLaEval framework, detailing how large‑model applications are evaluated through defined stages—background analysis, metric design, dataset generation, standards setting, and statistical analysis—while illustrating real‑world use cases in freight and driver invitation scenarios, and outlining future automation prospects.
The LiveVideoStackCon 2023 session revealed how a human‑perception‑driven end‑cloud super‑resolution framework, AI‑based no‑reference video quality assessment, and rigorous AB‑testing methods can dramatically reduce video bandwidth while enhancing visual quality, illustrating the broader challenges and opportunities in modern audio‑video systems.
