Loop Engineering with Adaptive Parsing in Action: Parsing Flat Tables with Azure and Figures with a Vision LLM
The article describes an adaptive parsing strategy for enterprise document intelligence that starts with a cheap parser and escalates to more powerful parsers only when needed. It adds a large language model as a final check to catch errors that deterministic checks miss, using self‑evaluation and groundedness checks. Two real‑world examples—escalating a flat table to Azure Layout and a figure to a vision LLM—demonstrate the cost‑effective, loop‑engineered approach.
- ▪Adaptive parsing begins with a fast, inexpensive parser and escalates to deeper parsers based on a cascade of checks.
- ▪Deterministic checks flag obvious parsing failures, while the LLM’s self‑evaluation and groundedness checks catch subtler errors at generation time.
- ▪The article walks through two end‑to‑end escalations: a flat table parsed with Azure Layout and a diagram parsed with a vision LLM.
- ▪By escalating only when necessary, the system reduces computational cost compared to running the heaviest parser on every page.
Towards Data Science files mainly under ai. We currently carry 104 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Towards Data Science |
| Canonical URL | https://towardsdatascience.com/loop-engineering-with-adaptive-parsing-in-action-parsing-flat-tables-with-azure-and-figures-with-a-vision-llm/ |
| Publication time | Mon, 20 Jul 2026 15:00:00 +0000 |
| Retrieval time | 2026-07-20T15:26:58.871Z |
| Last seen | 2026-07-20T15:53:38.998Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | __Qax_84xHLs |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Large Language Models Loop Engineering with Adaptive Parsing in Action: Parsing Flat Tables with Azure and Figures with a Vision LLM Enterprise Document Intelligence [Vol.1 #10B] – The LLM as last line of defence, then two real escalations walked end to end: a flat table to Azure, a figure to a vision model angela shi Jul 20, 2026 25 min read Share Photo by Larry Hyler, via Pexels. Some bad parses do not look bad until the answer comes out (classic OCR is the textbook case: EasyOCR recovers the words and quietly drops the table structure around them, and the answer reads fine until you check it against the page).
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Towards Data Science.
Discussion
0 commentsMore from Towards Data Science
