Search AI Products and News

Researchers from DeepSeek and Tsinghua University recently published a new paper exploring scaling methods for reward model inference, seemingly advancing DeepSeek R2. Reinforcement learning is widely used in the large-scale post-training phase of large language models, but faces the challenge of obtaining accurate reward signals for these models. The researchers found that using pointwise generative reward modeling (GRM) improves model adaptability and scalability during inference. To this end, they propose Self-Principle Calibration Tuning (SPCT) learning.

In today's development of artificial intelligence (AI), the speed of data transmission has become a significant bottleneck hindering its progress. To break through this barrier, a research team led by the University of Michigan (U-M) is developing a brand new chip connection system that uses optical waves instead of traditional cables for data transmission. This innovation is expected to address the 'memory wall' problem that limits computational speed and promote further growth of AI models. The project has received a $2 million grant from the National Science Foundation's Future Semiconductor program, with participation from the University of Washington and Penn State.

GPT-4 has undergone an epic version update, adding 128K ultra-long context and a new Assistants API. The advent of the GPT Store means anyone can earn money by creating their own specialized GPTs. Currently, there are barriers and pressing needs for applications in the GPT Store. The two main technological logics driving the development of LLMs are information fusion and autonomous decision-making.