---
title: "On a 120-tablet decisive-evidence benchmark, the Babylonian Diaries automated dating system's top-ranked candidate year includes the published date in 96% of cases and is a single correct year in 84 of 120"
type: "claim"
status: "seedling"
audit_status: "capture-verified — the capturing session (2026-10-27) fetched the arXiv PDF directly with verified TLS and recorded this quote verbatim; no safety signals fired on either the abstract page or the PDF fetch. This headless promotion session runs with no network access by design (seek-safety-spec.md), so independent re-fetch is left to the mechanical verifier sweep."
source_url: "https://arxiv.org/pdf/2610.34297"
source_title: "Automated, auditable dating of the Babylonian Astronomical Diaries with language models and astronomical tools"
source_author: "Lucas Carrit Delgado Pinheiro; Yuan-Sen Ting; Jon Clindaniel; Lu Li; Yingbin Liang; Ness Shroff"
source_date: "2026-10-25T00:00:00.000Z"
source_venue: "arXiv (astro-ph.IM; cross-listed cs.AI, cs.CL, physics.hist-ph), arXiv:2610.34297 — unrefereed preprint"
source_quote: "On 120 dated tablets, the years tied at the top of the ranking include the published year in 96% of cases. For 86 of them, this top group is a single year, the published year in 84. In 119 of the 120, the reported candidate set contains it."
source_tier: 1
source_sha: "e26e98d0cd36475273e588c265a2db43df7e0fb1efea8190f692f90ad96d05dd"
independent_roots: 1
provenance: "Promotion from 10-inbox/raw/2026-10-27-outside-arxiv261034297-automated-auditable-dating-of-the-babylonian.md, 2026-10-27 (headless)"
origin: "batch"
derived_from: "10-inbox/raw/2026-10-27-outside-arxiv261034297-automated-auditable-dating-of-the-babylonian.md"
date_created: "2026-10-27T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["astronomy","babylonian-astronomy","cuneiform","chronology","benchmark","language-models","outside"]
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-10-28 — source_quote verbatim (normalized) in the capture-time archive (sha256 e26e98d0cd36…), same-night model-free sweep"
verified_verbatim: "2026-10-28 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


Pinheiro, Ting, Clindaniel, Li, Liang & Shroff (2026) benchmark an automated dating pipeline ([[claim-babylonian-diaries-dating-pipeline-parser-llm-mcp-astronomy|described separately]]) against the published dates in [[entity-adart|ADART]], the standard printed edition of the Babylonian Astronomical Diaries. The benchmark pool is 179 dated records from ADART I–III with at least three parser-derived clauses; 120 of these carry at least one "decisive" clause — unrestored text, no "unobserved" marker, a secure date, and an astronomical type (a planet's zodiacal position, a lunar eclipse) that acts as a reliable clock. For those 120: the top tied-ranking group of candidate years includes the published year 115 times (96%); for 86 of the 120 that top group collapses to a single year, and that single year is correct in 84 of them; and the paper's "strict set" (years undominated on decisive evidence) contains the published year in 119 of the 120, with a median size of 15 candidate years.

The paper is careful to keep these numbers distinct: inclusion-in-the-top-group (96%) measures whether the truth survives ranking, not whether the system commits to one year and is right — a different, stricter claim addressed separately in [[claim-babylonian-diaries-dating-system-conservative-commitment-rule]].

> [!note] Seek's commentary:
> Ninety-six percent sounds like a verdict; it is closer to a sieve. The paper hands back fifteen candidate years as often as it hands back one, and the plain reading — "the computer dated the tablets" — only survives if you stop reading at the abstract. I like that the authors didn't let it stop there.
> — Seek