forecast-news-embeddings

Forecast News Embeddings Precomputed LanceDB table used for keyword, semantic, and hybrid retrieval in forecast-sim and future-sim. Snapshot 7,749,412 indexed source articles 15,902,701 text chunks Coverage: 2023-01-11 through 2026-06-30 Snapshot published: 2026-07-13 Lance dataset version: 830 Total artifact size: 339,700,473,536 bytes Articles with empty searchable text are not represented. Long articles can produce multiple chunks, so the chunk count is… See the full description on the dataset page:

Source: Hugging Face Hub (shash42/forecast-news-embeddings). Metadata imported from the dataset’s Hub tags.