A research team from Tsinghua University led by Andrew Yao and Yuan Yang has introduced the Algebraic World Model, or AWM, a system designed to help AI move from observation and experimentation to mathematical abstraction of how the world works. According to the team, AWM achieved a 99.9% score on the ARC-AGI-3 benchmark while using an average number of actions equal to 43% of the human baseline. The framework follows a three-step process: it first explores phenomena and writes hypotheses in Python code, then tests those hypotheses against new observations to extract general rules, and finally formalizes the rules in Lean 4 for mathematical proof. For unsolvable tasks, the Opus 5.5-based version of AWM reportedly needed only two attempts to prove that no solution existed and to identify the number of steps required to win. The research team includes co-first authors Pan Guancheng of Zhejiang University and Tan Qixin of Tsinghua’s Yao Class, with Yuan Yang and Turing Award winner Andrew Yao serving as advisors. The paper has been posted on arXiv, and the project homepage has also been released.
Techub News reported that a team from Tsinghua University led by Andrew Yao and Yuan Yang has released the Algebraic World Model, or AWM, a framework aimed at enabling AI to derive the laws behind observed phenomena through mathematical abstraction based on observation and experimentation.
99.9% score on ARC-AGI-3
The team said AWM passed the ARC-AGI-3 test set with a score of 99.9%. Its average number of actions was 43% of the human baseline.
A three-step method
According to the description provided, AWM works in three stages. It first explores phenomena and turns them into Python-coded hypotheses. It then uses new observations to test those hypotheses and distill them into general rules. In the final step, it formalizes those rules in Lean 4 and produces mathematical proofs.
Two attempts on unsolvable levels
For unsolvable levels, the Opus 5.5-based version of AWM needed only two attempts to prove that no solution existed and to state the number of steps required for a win.
Paper posted on arXiv
The research team includes co-first authors Pan Guancheng of Zhejiang University and Tan Qixin of Tsinghua University’s Yao Class. The advisors are Yuan Yang and Turing Award winner Andrew Yao. The paper has been published on arXiv, and the project homepage has been made public as well. The item cited Quantum Bit as the source of the report.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
