Metraipolis › archive › dc_bc8ab8b5fefc

FinEvolveKG: Controlled and Attributable Prompt Evolution for Financial Knowledge Graph Construction — arXiv:2611.08643

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.08643
License (as archived)
CC0-1.0
Archived
2026-11-09T22:34:26.722Z
SHA-256 of the archived text
630f6027a1715db955462d391539a4039fe1d5788b1738e79155d58341fa9ce5
Stable URL of this page
https://metraipolis.com/documento/dc_bc8ab8b5fefc

How to cite

FinEvolveKG: Controlled and Attributable Prompt Evolution for Financial Knowledge Graph Construction — arXiv:2611.08643 — arXiv (en), https://arxiv.org/abs/2611.08643 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc8ab8b5fefc (Metraipolis archive dc_bc8ab8b5fefc, sha256 630f6027a1715db955462d391539a4039fe1d5788b1738e79155d58341fa9ce5)
BibTeX
@misc{dcbc8ab8b5fefc,
  title = {FinEvolveKG: Controlled and Attributable Prompt Evolution for Financial Knowledge Graph Construction — arXiv:2611.08643},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.08643},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc8ab8b5fefc; sha256 630f6027a1715db955462d391539a4039fe1d5788b1738e79155d58341fa9ce5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08643. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Financial institutions construct knowledge graphs (KGs) from shared SEC 10-Ks to serve risk analysis, portfolio management, and investment research, whose analytical needs differ. Extraction must adapt without fragmenting shared grounding, typing, and temporal rules. Hand-authored persona prompts and fine-tuning add cost. Weight-free prompt optimization offers an alternative, but unconstrained evolution of a monolithic prompt can rewrite shared rules alongside persona guidance, obscuring changes and complicating review. We introduce FinEvolveKG, a KG-construction harness separating reusable general and persona-policy blocks and mechanically confining GEPA, a reflective prompt optimizer, to the designated block. During persona adaptation, the approved general block remains byte-stable and every optimizer edit is localized to a separately versioned persona policy. A dual objective balances generic temporal-KG quality with persona-relevant facts, motifs, and multi-hop connectivity. We evaluate five mutation scopes on ticker-disjoint 10-K splits, with three seeds for the persona-conditioned arms (v3-v5) across three personas. On the three-seed mean, the complete two-stage block-constrained workflow (v4) improves over the frozen prompt baseline (v1) by 19.4-30.9% and over the tested unconstrained full-prompt workflow (v5) by 2.2-6.4%, leading v5 on both objective components and winning all nine paired-seed comparisons with the lowest seed variance in every persona. Without further optimization, v4 also exceeds v5 across all evaluated out-of-distribution (OOD) shifts. An SME audit supports using the generated triplets as ground truth; a blinded two-SME audit favors v4 over v5 in 70.1% of contrast-enriched judgments. These results show that mechanically attributable evolution can coexist with strong empirical performance, without attributing the gains causally to the boundary.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.08643 — 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 2026-11-09T22:30:47.301Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).