Learning to Ask: Autonomous Knowledge Acquisition Using a Frozen Language Model as Environment

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autore principale: Aggarwal, Mohit
Natura: Recurso digital
Lingua:inglese
Pubblicazione: Zenodo 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866901686259286016
author Aggarwal, Mohit
author_facet Aggarwal, Mohit
contents <p>This paper presents an architecture where a lightweight outer learning layer uses a frozen large language model (LLM) as an environment for autonomous knowledge acquisition. The outer layer observes the LLM's outputs, extracts concepts and features, builds a knowledge graph, and generates questions driven by the graph's own structural gaps. Crucially, the outer layer forms its own abstractions and then queries the LLM about these self-generated abstractions, creating a cycle where learned structure drives new inquiry.</p> <p>A four-way ablation study over 200 steps shows that (1) feedback-driven questioning produces 4.9x denser knowledge graphs than structure-driven questioning without feedback, (2) every question strategy improves by 34-97% when feedback is enabled, and (3) structure alone performs worse than random exploration. An extended 2000-step run demonstrates that information gain accelerates rather than plateaus as the knowledge graph densifies. All code and data are publicly available.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_20113783
institution Zenodo
language eng
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Learning to Ask: Autonomous Knowledge Acquisition Using a Frozen Language Model as Environment
Aggarwal, Mohit
autonomous learning
knowledge graphs
language models
information gain
curiosity-driven exploration
concept formation
ablation study
<p>This paper presents an architecture where a lightweight outer learning layer uses a frozen large language model (LLM) as an environment for autonomous knowledge acquisition. The outer layer observes the LLM's outputs, extracts concepts and features, builds a knowledge graph, and generates questions driven by the graph's own structural gaps. Crucially, the outer layer forms its own abstractions and then queries the LLM about these self-generated abstractions, creating a cycle where learned structure drives new inquiry.</p> <p>A four-way ablation study over 200 steps shows that (1) feedback-driven questioning produces 4.9x denser knowledge graphs than structure-driven questioning without feedback, (2) every question strategy improves by 34-97% when feedback is enabled, and (3) structure alone performs worse than random exploration. An extended 2000-step run demonstrates that information gain accelerates rather than plateaus as the knowledge graph densifies. All code and data are publicly available.</p>
title Learning to Ask: Autonomous Knowledge Acquisition Using a Frozen Language Model as Environment
topic autonomous learning
knowledge graphs
language models
information gain
curiosity-driven exploration
concept formation
ablation study
url https://doi.org/10.5281/zenodo.20113783