From Knowledge Access to Source Learning: Developing Source-Specific Competence
arXiv cs.CL · 2 Oct 2026, 03:50 AEST
Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks. Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as.
Key facts
- Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks.
- Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as.
- We study source learning: developing reusable source-specific competence over a persistent authoritative source.
- We represent this competence with a persistent source model that captures reusable understanding of the source, including how its knowledge is structured, interpreted, and applied.
- To construct and progressively refine such models, we propose SourceLearn, which combines two complementary learning mechanisms.
Read aloud
Large language model L L M agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks. Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as. Large language model L L M agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks. Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as. We study source learning: developing reusable source-specific competence over a persistent authoritative source. We represent this competence with a persistent source model that captures reusable understanding of the source, including how its knowledge is structured, interpreted, and applied. To construct and progressively refine such models, we propose SourceLearn, which combines two complementary learning mechanisms.
Sources in this cluster
- From Knowledge Access to Source Learning: Developing Source-Specific Competence
arXiv cs.CL · 2 Oct 2026, 03:50 AEST