Metraipolis › archive › dc_620e0d035b15

How Correct Is Your Answer? A Semantic Correctness Framework for Open QA Evaluation — arXiv:2609.01369

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01369
License (as archived)
CC0-1.0
Archived
2026-10-17T06:41:37.866Z
SHA-256 of the archived text
2d41868d6e69faa7df064f7ea56ed8d4664aa1af61c13acab34be43a03419853
Stable URL of this page
https://metraipolis.com/documento/dc_620e0d035b15

How to cite

How Correct Is Your Answer? A Semantic Correctness Framework for Open QA Evaluation — arXiv:2609.01369 — arXiv (en), https://arxiv.org/abs/2609.01369 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_620e0d035b15 (Metraipolis archive dc_620e0d035b15, sha256 2d41868d6e69faa7df064f7ea56ed8d4664aa1af61c13acab34be43a03419853)
BibTeX
@misc{dc620e0d035b15,
  title = {How Correct Is Your Answer? A Semantic Correctness Framework for Open QA Evaluation — arXiv:2609.01369},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01369},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_620e0d035b15; sha256 2d41868d6e69faa7df064f7ea56ed8d4664aa1af61c13acab34be43a03419853},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01369. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reliable evaluation of open-ended question answering remains a bottleneck for measuring answer correctness of modern LLMs. Unlike multiple-choice tasks, free-form answers may be correct in many surface forms and may fail in qualitatively different ways, including incompleteness, contradiction, overgeneration, and endorsement of false premises. Existing judgment-based and similarity-based metrics often collapse these distinctions. We address this gap with three reusable contributions. First, we introduce a semantic correctness taxonomy that assigns open-ended answers to eight ordered classes, separating verbose-but-correct answers from those contaminated by hallucinated content. Second, we release CAP-Correctness, an 8.8k-example benchmark spanning widely used QA datasets, and CAP-Statements, an 11k-example dataset for converting question-answer pairs into declarative statements for natural language inference (NLI) training and statement-based evaluation. Third, we introduce CAP (Context-Aware Precision), a reference-based metric that scores question-conditioned statements using bidirectional NLI. Under a monotonicity protocol testing whether metrics respect the taxonomy's intended ordering, CAP outperforms established baselines.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.01369 — 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 2026-10-17T06:33:38.861Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).