# Visual Information Extraction on POIE Source: [Visual Information Extraction on POIE]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:29.795Z Evaluation of Visual Information Extraction performance on the POIE dataset, typically measured using Accuracy and ANLS metrics. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Visual Information Extraction | | Dataset (including any stated split) | POIE | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. - [ANLS](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Qwen2-VL-7B + RIDGE | 94.45 | Not recorded | 68fa78131e209131cea76a9c | | Qwen2-VL-7B + RIDGE + DocVQA + FUNSD | 94.11 | Not recorded | 68fa78131e209131cea76a9c | | Qwen2-VL-7B + RIDGE + DocVQA | 94.08 | Not recorded | 68fa78131e209131cea76a9c | | Qwen2-VL-7B+ | 94.04 | Not recorded | 68fa78131e209131cea76a9c | | Qwen2-VL-7B | 93.51 | Not recorded | 68fa78131e209131cea76a9c | | Qwen2-VL-7B | 93.51 | Not recorded | 68fa78131e209131cea76a9c | | DocOwl-1.5-Chat | 51.57 | Not recorded | 68fa78131e209131cea76a9c | | LLaVA-NeXT-Mistral-7B+ | 51.46 | Not recorded | 68fa78131e209131cea76a9c | | LLaVA-NeXT-Mistral-7B | 49.88 | Not recorded | 68fa78131e209131cea76a9c | | Monkey | 30.64 | Not recorded | 68fa78131e209131cea76a9c | ### Qwen2-VL-7B + RIDGE Accuracy: 94.45. Result ID: 69957bfb3badfaffe6733567. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Dataset Combinations":"+RIDGE"}. This record does not assert independent reproduction. ### Qwen2-VL-7B + RIDGE + DocVQA + FUNSD Accuracy: 94.11. Result ID: 69957bfb3badfaffe6733569. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Dataset Combinations":"+RIDGE+DocVQA+FUNSD"}. This record does not assert independent reproduction. ### Qwen2-VL-7B + RIDGE + DocVQA Accuracy: 94.08. Result ID: 69957bfb3badfaffe6733568. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Dataset Combinations":"+RIDGE+DocVQA"}. This record does not assert independent reproduction. ### Qwen2-VL-7B+ Accuracy: 94.04. Result ID: 69957bfb3badfaffe6733596. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":true}. This record does not assert independent reproduction. ### Qwen2-VL-7B Accuracy: 93.51. Result ID: 69957bfb3badfaffe6733566. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Dataset Combinations":"Base Model"}. This record does not assert independent reproduction. ### Qwen2-VL-7B Accuracy: 93.51. Result ID: 69957bfb3badfaffe6733595. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":false}. This record does not assert independent reproduction. ### DocOwl-1.5-Chat Accuracy: 51.57. Result ID: 69957bfb3badfaffe6733592. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":false}. This record does not assert independent reproduction. ### LLaVA-NeXT-Mistral-7B+ Accuracy: 51.46. Result ID: 69957bfb3badfaffe6733594. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":true}. This record does not assert independent reproduction. ### LLaVA-NeXT-Mistral-7B Accuracy: 49.88. Result ID: 69957bfb3badfaffe6733593. [Reporting paper]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":false}. This record does not assert independent reproduction. ### Monkey Accuracy: 30.64. Result ID: 69957bfb3badfaffe6733591. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"zero-shot":true,"RIDGE fine-tuned":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.