This video details the transition from Generative AI (which we cannot trust, probabilistic system) to Agentic/Verifiable Reasoning Systems (which we hopefully can trust, after three rounds of AI verifiers).

All rights w/ authors:
"Learning to Reason with Insight for Informal Theorem Proving"
Yunhe Li∗1 Hao Shi∗2 Bowen Deng∗1 Wei Wang∗3 Mengzhe Ruan1 Hanxu Hou4, Zhongxiang Dai5 Siyang Gao1 Chao Wang13 Shuang Qiu†1 Linqi Song†1
from
1 City University of Hong Kong
2 Tsinghua University
3 Ke Holdings Inc.
4 Shenzhen University of Advanced Technology
5 Chinese University of Hong Kong, Shenzhen.
arXiv:2604.16278

AgentV-RL: Scaling Reward Modeling with Agentic Verifier
Jiazheng Zhang1* , Ziche Fu1∗, Zhiheng Xi1∗, Wenqing Jing1, Mingxu Chai1, Wei He1,
Guoqiang Zhang1, Chenghao Fan2, Chenxin An3, Wenxiang Chen1, Zhicheng Liu4,
Haojie Pan4, Dingwei Zhu1, Tao Gui5,6†, Qi Zhang5,6, Xuanjing Huang5,6
from
1 College of Computer Science and Artificial Intelligence, Fudan University
2 Huazhong University of Science and Technology
3 The University of Hong Kong
4 ByteDance Seed
5 Institute of Trustworthy Embodied AI, Fudan University
6 Shanghai Key Laboratory of Multimodal Embodied AI.
arXiv:2604.16004

#newtech
#mythos
#scienceexplained #newtechnologyideas