#data-curation
Corpora, filtering, dedup, synthetic data.
2 chapters, oldest era first: one idea traced across the books.
- From Noise to Image Cross-era Ch. 21: Evaluation, Provenance, and Safety FID's math and failure modes, human preference data, memorization and extraction, watermarking and C2PA.
- The Language Machines 2020–2025 Ch. 11: The Data Question Corpora, dedup, filtering, synthetic data, and contamination — data as the real moat.