AI Cosmos improves learning with scenario based agent interactions
AI Cosmos: An Agentic Infra for Situated Learning
Computation and Language
Summary
Learning with big language models can help people do tasks better during training, but that doesn’t always mean they can do those tasks well later on their own. The authors created AI Cosmos, a system that sets up learning like real-world scenarios with agents that have specific roles, plus tools to practice tasks directly. This system keeps track of what was learned and how, so learners can build on previous work. The authors tested AI Cosmos on legal reasoning and programming skills, finding it helped learners improve more than just using a language model as a teacher.
What this means in practice
- •For legal tech developers: Create interactive training tools that improve legal reasoning skills through scenario based agent collaboration.
- •For software engineering trainers: Build programming practice platforms that support mastering unseen coding problems with integrated agent assistance.
Authors
AI Cosmos Team, Jinfeng Zhou, Yuxuan Chen, Fangming Chang, Mingyu Su, Yifan Zhu, Yuan Wang, Bo Zhang, Bosi Wen, Chenhao Yuan, Chenwen Liao, Dazhi Jiang, Deyao Hong, Erle Zhu, Genrui Zhang, Guchan Li, Hongwen Gan, Huichao Bi, Jiahao Yan, Jincheng Fan, Kehan Zheng, Qinren Luo, Qiran Liu, Ruoqi Cao, Shizhi Zhang, Sihan Chen, Siyan Deng, Xuehan Zhou, Xujun Li, Yang Dang, Yanzhuo Zhu, Yu Zhang, Zheyu Chen, Hongning Wang, Manli Li
Abstract
Large language models (LLMs) can improve task performance during learning, yet these gains do not necessarily reflect learners' ability to thereafter apply the knowledge and skills independently. We present AI Cosmos, an agentic infrastructure for situated learning that organizes learning around two complementary principles: scenario-based participation and cumulative assets of practice. AI Cosmos organizes learners and role-specialized agents within learning scenarios that integrate task-relevant knowledge and, when needed, executable workbenches. An asset-memory loop preserves the processes and outcomes of practice as reusable assets with explicit provenance and dependencies. We evaluate scenario-based interaction through two preregistered randomized controlled studies examining the acquisition of legal reasoning and programming competence, holding the same underlying LLM across conditions. AI Cosmos achieves greater learning gains than a conventional LLM in legal reasoning and higher mastery of unseen programming problems than an LLM-as-Teacher baseline.