Search
On May 26, Sequoia China introduced xbench, a groundbreaking AI benchmarking tool, accompanied by a scholarly paper titled "xbench: Tracking Agents' Productivity, Scaling with Profession-Aligned Real-World Evaluations." This marks the first AI benchmark spearheaded by an investment firm, in partnership with over a dozen universities and research institutions from both China and abroad, uniting the expertise of numerous doctoral students. The tool employs a dual-track evaluation system and a dynamic, evergreen assessment mechanism. xbench is designed to evaluate and foster the enhancement of AI systems' capabilities and the expansion of their technological boundaries. Furthermore, it prioritizes quantifying the practical value of AI in real-world applications and maintains a sustained focus on pivotal breakthroughs in Agent products.
-
C114 Communication Network -
Communication Home
