Metraipolis › archive › dc_4645ada36726
Externalized CPDAG Summaries Improve LLM Causal Deduction — arXiv:2609.31071
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.31071
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T01:38:31.908Z
- SHA-256 of the archived text
37e54eaa57c9293e050e924fd8a088cce2ba0744e7afb6679aa8a6ce6318c5da- Stable URL of this page
https://metraipolis.com/documento/dc_4645ada36726
How to cite
Externalized CPDAG Summaries Improve LLM Causal Deduction — arXiv:2609.31071 — arXiv (en), https://arxiv.org/abs/2609.31071 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4645ada36726 (Metraipolis archive dc_4645ada36726, sha256 37e54eaa57c9293e050e924fd8a088cce2ba0744e7afb6679aa8a6ce6318c5da)
BibTeX
@misc{dc4645ada36726,
title = {Externalized CPDAG Summaries Improve LLM Causal Deduction — arXiv:2609.31071},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.31071},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4645ada36726; sha256 37e54eaa57c9293e050e924fd8a088cce2ba0744e7afb6679aa8a6ce6318c5da},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31071. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Corr2Cause asks whether a causal claim holds in every DAG compatible with observed correlations and conditional independencies. We frame this as latent-object reasoning: the label is defined by a CPDAG query, but free-form chain-of-thought often collapses the Markov-equivalence-class problem into local pattern matching. We propose Structured Thinking, a two-turn pipeline that first externalizes a typed, schema-constrained CPDAG summary and then answers against that graph state. On the Corr2Cause full test, Structured Thinking raises Qwen3.5-27B from $73.0$ to $86.4$ $F_1$(Yes) over a strong PC-instruction baseline in the primary paired run ($+13.4$ pp; McNemar $p=2.4\times 10^{-6}$; bootstrap $95\%$ CI [$+8.4$, $+18.6$]); across three full-ID seeds, the mean gain is $+8.1 \pm 5.3$ pp. A PC-scaffolded two-turn prose control reaches only $67.6$ $F_1$, indicating that a detailed PC scaffold plus a schema-free prose intermediate is not sufficient. The same pattern holds on Qwen3.6-27B, Paraphrase-OOD, and GPT-5.4-mini. Scrambling the emitted CPDAG costs $12.0$ pp $F_1$, and a full-split audit shows close agreement with the reference CPDAG (ID skeleton $F_1$ $0.960$; exact match $75.9\%$). These results support a bounded design principle: externalize the latent object that defines the label, constrain its form, and test whether downstream answers use it.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.31071 — 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 2026-10-08T01:12:14.333Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).