Metraipolis › archive › dc_c3e8ac4e3402
When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translations — arXiv:2609.33691
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33691
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:53:07.826Z
- SHA-256 of the archived text
25f6c659e371b57275631a31c280c74b2bf9cce2ec005ea8bf3c121839616077- Stable URL of this page
https://metraipolis.com/documento/dc_c3e8ac4e3402
How to cite
When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translations — arXiv:2609.33691 — arXiv (en), https://arxiv.org/abs/2609.33691 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c3e8ac4e3402 (Metraipolis archive dc_c3e8ac4e3402, sha256 25f6c659e371b57275631a31c280c74b2bf9cce2ec005ea8bf3c121839616077)
BibTeX
@misc{dcc3e8ac4e3402,
title = {When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translations — arXiv:2609.33691},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33691},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c3e8ac4e3402; sha256 25f6c659e371b57275631a31c280c74b2bf9cce2ec005ea8bf3c121839616077},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33691. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Classical texts aligned with their translations support machine translation, retrieval and computational research, but evidence comparing alignment workflows is scattered. This study compares seven systems on 452 texts in Pali, Sanskrit, Mishnaic Hebrew and Tibetan, comprising 9,833 human-aligned units: four embedding pipelines, a direct LLM call, an autonomous agent, and the agent revised by an independent auditor. Generative workflows recover 93-94% of reference correspondences, against at most 77% for embeddings. A ceiling analysis shows that sentence boundaries make some references unrepresentable by the embedding pipelines. Reference recovery is similar across generative workflows: the agent's advantage is 0.5 percentage points (95% CI -0.02 to 1.17), and auditing adds no established benefit. Agents nevertheless produce structurally valid output for all 452 texts, against 437 for direct calls. A blinded three-LLM panel assesses every generative mismatch against the source and human reference. Most mismatches are labelled defensible editorial variation; consensus major-error labels cover only 0.06-0.14% of units. The panel labels significantly fewer residual defects for agents than direct calls (0.7% versus 1.4%), suggesting that reference recovery alone understates alignment quality. On ten long Pali discourses taken as published online, agents and audited agents raise recovery from the direct call's 71% to 84% and 92%. Identical reference-located chunks bring all three to 93%. Agents thus improve structural reliability and reduce judged defects on short passages, while their large recovery advantage on long documents disappears after chunking. In this setting, independent auditing offers little measurable additional benefit on prepared passages. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.33691 — 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 2026-10-07T21:27:08.543Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).