added human 100 gold

fix to readme

v1.2: add human_gold_50 split (database-scoped gold standard, evaluation only) + explicit viewer configs

Merge origin/main: keep Hub README edits (data files identical on both sides)

Update README.md

Update README.md

Change entity_to_token to list-of-objects schema; update README accordingly

Change entity_to_token to list-of-objects schema; update README accordingly

Normalize entity_to_token schema (0..19) and document split groups; update human_train, human_test, synthetic_train

Normalize entity_to_token schema (0..19) and document split groups; update human_train, human_test, synthetic_train

Normalize entity_to_token schema (0..19 keys) and document split groups; replace all three splits (human_train, human_test, synthetic_train)

dataset release v1.0

Update README.md

Update README.md

initial commit