Stars
Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs
We introduce 'Thinking with Video', a new paradigm leveraging video generation for multimodal reasoning. Our VideoThinkBench shows that Sora-2 surpasses GPT5 by 10% on eyeballing puzzles and reache…
TransMLA: Multi-Head Latent Attention Is All You Need (NeurIPS 2025 Spotlight)
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
雀魂牌谱屋 (See also: https://github.com/SAPikachu/amae-koromo-scripts )
expectqwq / DanXi
Forked from DanXi-Dev/DanXi[Windows / Mac / Linux / Android / iOS] Maybe the best all-rounded service app for Fudan University students. 可能是复旦学生最好的第三方校园服务APP。
[Windows / Mac / Linux / Android / iOS] Maybe the best all-rounded service app for Fudan University students. 可能是复旦学生最好的第三方校园服务APP。
Principle of Programming Language 2025 final project