damanegi2005/captcha_vlm_evaluation

How damanegi2005/captcha_vlm_evaluation works

This repository is a set of CAPTCHA VLM evaluation scripts, not a packaged service. The workflow reads CAPTCHA examples and ground-truth answers, prompts GPT-4.1-mini, compares predictions with answers using task-specific metrics, and saves evaluation results. The README also documents OCR, object-selection, and coordinate-localization capabilities; their likely script paths appear in the tree, but those files were not sampled. The shown scripts combine prompting, scoring, and output in one runtime module, so the graph preserves those boundaries rather than inventing shared pipeline layers.

Main entry points: test_ocr.py OCR evaluation · test_object.py Object selection · click_order.py Click order · arithmetic.py Arithmetic

Components 12

CAPTCHA tasks

OCR evaluation test_ocr.py
Object selection test_object.py
Click order click_order.py
Coordinate localization click_order.py
Arithmetic arithmetic.py
Common sense Common_sense_reasoning.py
Rule filtering rule_based.py

Evaluation workflow

CAPTCHA examples
GPT-4.1-mini API
Task metrics
Evaluation results

Other components

Evaluator

Connections 29

Evaluator → CAPTCHA examples: provides examples README.md
CAPTCHA examples → Click order: supplies inputs click_order.py
CAPTCHA examples → Coordinate localization: supplies inputs click_order.py
CAPTCHA examples → Arithmetic: supplies inputs arithmetic.py
CAPTCHA examples → Common sense: supplies questions Common_sense_reasoning.py
CAPTCHA examples → Rule filtering: supplies inputs rule_based.py
CAPTCHA examples → OCR evaluation: supplies examples README.md
CAPTCHA examples → Object selection: supplies examples README.md
Click order → GPT-4.1-mini API: requests predictions click_order.py
Coordinate localization → GPT-4.1-mini API: requests predictions click_order.py
Arithmetic → GPT-4.1-mini API: requests predictions arithmetic.py
Common sense → GPT-4.1-mini API: requests predictions Common_sense_reasoning.py
Rule filtering → GPT-4.1-mini API: requests predictions rule_based.py
OCR evaluation → GPT-4.1-mini API: requests predictions README.md
Object selection → GPT-4.1-mini API: requests predictions README.md
Click order → Task metrics: scores predictions click_order.py
Coordinate localization → Task metrics: scores predictions click_order.py
Arithmetic → Task metrics: scores predictions arithmetic.py
Common sense → Task metrics: scores predictions Common_sense_reasoning.py
Rule filtering → Task metrics: scores predictions rule_based.py
OCR evaluation → Task metrics: scores predictions README.md
Object selection → Task metrics: scores predictions README.md
Click order → Evaluation results: writes results click_order.py
Coordinate localization → Evaluation results: writes results click_order.py
Arithmetic → Evaluation results: writes results arithmetic.py
Common sense → Evaluation results: writes results Common_sense_reasoning.py
Rule filtering → Evaluation results: writes results rule_based.py
OCR evaluation → Evaluation results: saves results README.md
Object selection → Evaluation results: saves results README.md

Drawn by GitDiagram on November 13, 2026 from the repository's file tree, README and source files. Read as Markdown ·