Upload final_competition.csv with huggingface_hub (#4)

Update final_competition.csv: added year column and formatted paper_ids (#3)

Upload final_competition.csv

Upload coalescence-bigbang-competition-interactions.json with huggingface_hub

Add final_competition.csv: 30 sampled papers (ICLR 2025 + 2026)

Upload iclr_2025_oral_200.csv with huggingface_hub

Restore frontend_paper_id UUIDs in iclr_2025_oral_200.csv

Upload flaws_iclr2026_leaderboard.csv with huggingface_hub

Fix iclr_2025_oral_200.csv: clear frontend_paper_id (not assigned yet)

Move iclr_2025_oral_200.csv to root alongside other CSV files

Add iclr_2025_oral_200.csv (200 sampled oral papers with citations normalized by months)

Add hidden_leaderboard.csv for ICLR 2026 sample

Upload molbook_leaderboad.csv (#2)

Upload molbook_leaderboad.csv (#1)

Parallelize pipeline with asyncio (--workers 10 default)

Add patched FLAWS pipeline source (env-var config, Claude CLI support)

Add Claude paper manifest (50 papers)

Add Gemini paper manifest (50 papers)

Add FLAWS perturbation pipeline script

Add FLAWS pipeline README

Add molbook benchmark data (972 papers: metadata + PDFs)

Re-add parquet files (iclr25v2, iclr_2025_raw, iclr_2026_raw)

Update iclr_2025_full.json with citation counts (arxiv_id, citations_serper)

Remove parquet files

Update iclr_2026_full.json with citation counts (arxiv_id, citations_serper)

Update iclr_2025_small.json with citation counts (arxiv_id, citations_serper)

Add iclr-dataset: threads batch 1/7

Add iclr-dataset: raw parquets

Add iclr-dataset: dataset JSONs

Add iclr-dataset: small files

initial commit