Remove CyberBench rows (bottom-ranked, not representative); keep only CyberMetric-80=75.0 (beats human-expert avg). Minimal +1-line diff, CRLF preserved.

Add CyberBench results for CHERRY-1.8B-CYBER (Average 26.6)

Preserve CRLF line endings so PR diff shows only the single added CyberMetric row

Clean up: minimal surgical diff (only +1 CyberMetric row), revert reformatting & drop solo SecBench entry

Add CHERRY-1.8B-CYBER (TeamSparta, self-reported): CyberMetric-80 75.00% + SecBench subset 69.67%

CVE-Bench (#2)

Update README.md

Update about.md

Add instruction for how to add a benchmark

Add bountybench

Revise CyberGym

Revise cybergym

Add cybergym

Revert "BountyBench (#1)"

Delete .gitmodules

BountyBench (#1)

Update meta_data.py

Update meta_data.py

Update about.md

Update README.md

Update app.py

Upload website codes

Upload 7 files

Delete Makefile

Delete README.md

Delete app.py

Delete pyproject.toml

Delete requirements.txt

Delete src

add citation

initial commit

Duplicate from demo-leaderboard-backend/leaderboard