Metraipolis › archive › dc_b1108fa2c2fb
Rethinking Structured Generation: Can Graph-Based Reasoning Resolve Ambiguity? — arXiv:2610.31681
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.31681
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-26T19:50:51.175Z
- SHA-256 of the archived text
084877342ba98c2cb4af8a85dcbc3447c9d31015484111a403ff9e208c110a28- Stable URL of this page
https://metraipolis.com/documento/dc_b1108fa2c2fb
How to cite
Rethinking Structured Generation: Can Graph-Based Reasoning Resolve Ambiguity? — arXiv:2610.31681 — arXiv (en), https://arxiv.org/abs/2610.31681 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b1108fa2c2fb (Metraipolis archive dc_b1108fa2c2fb, sha256 084877342ba98c2cb4af8a85dcbc3447c9d31015484111a403ff9e208c110a28)
BibTeX
@misc{dcb1108fa2c2fb,
title = {Rethinking Structured Generation: Can Graph-Based Reasoning Resolve Ambiguity? — arXiv:2610.31681},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.31681},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b1108fa2c2fb; sha256 084877342ba98c2cb4af8a85dcbc3447c9d31015484111a403ff9e208c110a28},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31681. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Structured generation is typically formulated as a deterministic mapping from input to a single structured output. This assumption is fundamentally misaligned with natural language, which often admits multiple structurally distinct yet semantically valid interpretations. As a result, existing methods rely on single-reference supervision that collapses ambiguity into one prediction, leading to unstable and inconsistent outputs under underspecified inputs. We reformulate structured generation as an \textit{ambiguity-aware inference problem}, where the objective is to reason over a set of plausible structured outputs rather than predict a single solution. We show that standard training inherently suppresses valid alternatives, and instead adopt a multi-valid learning framework that preserves ambiguity during optimization. To operationalize this formulation, we construct a candidate space of structured outputs, represent each candidate as a graph capturing entities and their relationships, and perform selection via graph-based reasoning over semantic alignment and relational consistency. This enables principled comparison across competing interpretations beyond token-level likelihoods. Experiments on structured generation benchmarks demonstrate consistent improvements in accuracy, stability, and robustness under varying levels of ambiguity. These results highlight the importance of modeling ambiguity explicitly and suggest a shift from single-output prediction to set-level reasoning in structured generation.
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/2610.31681 — 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 2026-10-26T19:45:33.343Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).