arXiv 机器学习· Shengjun Kris Dong, Dima Nikiforov, Sahil Modi, Sana Damani, Edward C Lin, Siva Kumar Sastry Hari, Christos Kozyrakis·· 19 hours agoAI score40
ModelBlaster：通过学习增强智能体跨模型、算子与硬件后端弥合性能差距
ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends
Not yet translated; shown in Chinese.
AI guide
ModelBlaster 提出内存增强上下文强化学习（MAIC-RL）框架，通过推理时更新知识库实现跨算子、模型与异构硬件的优化及经验复用。
Source:arXiv 机器学习 · arxiv.org