# When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translations

arXiv: https://arxiv.org/abs/2609.33691 · HTML: https://mlsanity.com/paper/2609.33691/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 38 of its 63 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 30 of 38)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [Language-agnostic BERT Sentence Embedding](https://mlsanity.com/paper/2007.01852/) (2020)
- [Large Language Models Are State-of-the-Art Evaluators of Translation Quality](https://mlsanity.com/paper/2302.14520/) (2023)
- [NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmark](https://mlsanity.com/paper/2310.18018/) (2023)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs](https://mlsanity.com/paper/2406.01297/) (2024)
- [AI Agents That Matter](https://mlsanity.com/paper/2407.01502/) (2024)
- [LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs](https://mlsanity.com/paper/2408.07055/) (2024)
- [The 'Problem' of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation](https://mlsanity.com/paper/2211.02570/) (2022)
- [MMTEB: Massive Multilingual Text Embedding Benchmark](https://mlsanity.com/paper/2502.13595/) (2025)
- [Let Me Speak Freely? A Study on the Impact of Format Restrictions on Performance of Large Language Models](https://mlsanity.com/paper/2408.02442/) (2024)
- [Are Large Language Model-based Evaluators the Solution to Scaling Up Multilingual Evaluation?](https://mlsanity.com/paper/2309.07462/) (2023)
- [DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression](https://mlsanity.com/paper/2609.19969/) (2026)
- [Error Span Annotation: A Balanced Approach for Human Evaluation of Machine Translation](https://mlsanity.com/paper/2406.11580/) (2024)
- [(Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Texts](https://mlsanity.com/paper/2405.11804/) (2024)
- [Itihasa: A large-scale corpus for Sanskrit to English translation](https://mlsanity.com/paper/2106.03269/) (2021)
- [F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual World](https://mlsanity.com/paper/2603.19223/) (2026)
- [Investigating Length Issues in Document-level Machine Translation](https://mlsanity.com/paper/2412.17592/) (2024)
- [SentAlign: Accurate and Scalable Sentence Alignment](https://mlsanity.com/paper/2311.08982/) (2023)
- [Sentence Embedding Models for Ancient Greek Using Multilingual Knowledge Distillation](https://mlsanity.com/paper/2308.13116/) (2023)
- [Extending Automatic Machine Translation Evaluation to Book-Length Documents](https://mlsanity.com/paper/2509.17249/) (2025)
- [MITRA: A Large-Scale Parallel Corpus and Multilingual Pretrained Language Model for Machine Translation and Semantic Retrieval for P=ali, Sanskrit, Buddhist Chinese, and Tibetan](https://mlsanity.com/paper/2601.06400/) (2026)
- [Align-then-Slide: A complete evaluation framework for Ultra-Long Document-Level Machine Translation](https://mlsanity.com/paper/2509.03809/) (2025)
- [Building from Scratch: A Multi-Agent Framework with Human-in-the-Loop for Multilingual Legal Terminology Mapping](https://mlsanity.com/paper/2512.12950/) (2025)
- [PaliBench: A Multi-Reference Blueprint for Classical Language Translation Benchmarks](https://mlsanity.com/paper/2605.16881/) (2026)
- [Automatic Translation Alignment Pipeline for Multilingual Digital Editions of Literary Works](https://mlsanity.com/paper/2410.13255/) (2024)

## Abstract (arXiv)

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.