Search papers, labs, and topics across Lattice.
Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia, Beijing University of Posts and Telecommunications
5
0
11
Achieving a score of $2.68 \times 10^{-3}$ in a depth estimation challenge reveals the untapped potential of zero-shot learning in complex visual tasks.
Paraphrased prompts may seem equivalent, but they can drastically alter learning dynamics, with some prompts leading to better generalization and less forgetting.
GAIS enables the synthesis of complex tasks that not only outperform existing methods but also do so with dramatically improved data efficiency.
Reward models, despite excelling at general response quality, stumble when it comes to capturing individual user preferences, achieving only 76% accuracy on a new personalized benchmark.
Forget months of manual experimentation: ResearchEVO autonomously discovers and documents novel scientific algorithms, even generating the LaTeX paper.
