From Knowledge Access to Source Learning: Developing Source-Specific Competence

Oct 1, 2026·
Lucheng Fu
,
Kejing Xia
,
Yiyang Wang
Yiqiao Jin
Yiqiao Jin
,
Jinjin He
,
Xiyuan Yang
,
Haoxin Liu
,
Ye Yu
,
Haibo Jin
,
Yijia Xiao
,
Wenke Lee
,
B. Aditya Prakash
,
Haohan Wang
· 1 min read
Abstract
SourceLearn studies how an LLM agent can progressively learn from a persistent authoritative source. It maintains a source model that represents how knowledge is structured, interpreted, and applied. Self-directed learning revisits incompletely understood material, while task-guided learning identifies recurring representational needs. Both mechanisms reconstruct persistent updates from the original source.
Type
Publication
arXiv preprint arXiv:2610.02150

Overview

SourceLearn studies how an LLM agent can progressively learn from a persistent authoritative source. It maintains a source model that represents how knowledge is structured, interpreted, and applied. Self-directed learning revisits incompletely understood material, while task-guided learning identifies recurring representational needs. Both mechanisms reconstruct persistent updates from the original source.

LLM Agents Agent Memory Retrieval-Augmented Generation
Yiqiao Jin
Authors
Yiqiao Jin
Final-year Ph.D. Candidate in Computer Science
My research sits at the intersection of multimodal foundation models, intelligent agents, and spatial/physical intelligence. I build general-purpose models that perceive, reason, learn, and act in interactive virtual and physical environments.