# Tasks in GAIA subdataset

- [ ] Extract from Wikipedia (2022)
- [ ] Video from Youtube
- [ ] Inverted text
- [ ] Chess move from image
- [ ] Extract from Wikipedia (2016)
- [ ] Algebra
- [ ] Video from Youtube
- [ ] Retrieval from LibreText pdf
- [ ] Structured output
- [ ] ASR
- [ ] Extract from Wikipedia with polish wording
- [ ] Python code output
- [ ] Wikipedia baseball info
- [ ] ASR
- [ ] Web browsing
- [ ] Web browsing
- [ ] Web browsing, wikipedia sport info
- [ ] Wikipedia baseball info from table
- [ ] Excel file
- [ ] Web browsing, wikipedia

# Priorities

1. Web browsing
2. Wikipedia API
3. Video processing
4. PDF Extraction
5. ASR

# Extensions

- Sympy/Mathematica : Algebra
- OpenCV : CV
- Stockfish : Chess moves
- NLTK/spaCy : NLP
- docling : PDF extraction