Metraipolis › archive › dc_37bdfddec40a

Co-Evolving Structured Knowledge and Reasoning in Language Models — arXiv:2608.26386

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.26386
License (as archived): CC0-1.0
Archived: 2026-10-17T14:25:03.999Z
SHA-256 of the archived text: 1c81e20b69ec4d6a2520a75eb739f5863e79a08cd531b5d99e9aa808ff39b77c
Stable URL of this page: https://metraipolis.com/documento/dc_37bdfddec40a

How to cite

Co-Evolving Structured Knowledge and Reasoning in Language Models — arXiv:2608.26386 — arXiv (en), https://arxiv.org/abs/2608.26386 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_37bdfddec40a (Metraipolis archive dc_37bdfddec40a, sha256 1c81e20b69ec4d6a2520a75eb739f5863e79a08cd531b5d99e9aa808ff39b77c)

BibTeX

@misc{dc37bdfddec40a,
  title = {Co-Evolving Structured Knowledge and Reasoning in Language Models — arXiv:2608.26386},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26386},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_37bdfddec40a; sha256 1c81e20b69ec4d6a2520a75eb739f5863e79a08cd531b5d99e9aa808ff39b77c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26386. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-augmented methods improve factual accuracy by grounding language models in external knowledge, but retrieving over unstructured text often introduces irrelevant context and offers limited control over the retrieved information. Structured knowledge bases offer a more controllable alternative, yet they are expensive to construct and often brittle to reason over. To address these limitations, we propose KBevo: a co-evolving framework that jointly learns to construct a structured knowledge base and reason over it for knowledge-intensive question answering. By optimizing both components end-to-end with QA outcome rewards, our method enables reasoning success to directly improve the quality of the constructed knowledge base. This leads to larger, better-connected knowledge structures with higher answer reachability, while also improving compositional factual reasoning and controllability compared to standard retrieval baselines.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.26386 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T13:16:53.635Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).