University of Chinese Academy of Sciences

A paper in the Springer journal Annals of Data Science from the University of Chinese Academy of Sciences introduces an “Agent Capability Periodic Table” that maps agents into 243 configurations. The framework treats an agent as an open information-processing system and distills five basic capabilities—control, generation, memory, output, input—each varying along absent, finite, or infinite levels, yielding 243 combinations. The authors place humans, bacteria, thermostats, Turing machines, and large-model agents in the same coordinate system. They also argue that classical mechanics, relativity, and quantum mechanics may each presuppose observers occupying different positions in the table.

Peking University DCAI Team, with University of Chinese Academy of Sciences, Shanghai Institute of Algorithm Innovation and Zhongguancun Academy, has open-sourced DataFlex-RL, a data-strategy framework for RL post-training built as a verl plugin. It turns data selection, sample reweighting and domain ratio adjustment into configurable components, so developers can test and compare strategies inside a single RLVR/GRPO pipeline.

The paper reports 591 experiments across Qwen2.5-7B-base, Llama-3.1-8B-base and math, logic and science tasks, and ranked second on Hugging Face's daily paper chart. Licence was not stated.