FINDING #6203 · UNIT ID 1204330026
NVlabs/GR00T-VisualSim2Real
GR00T-VisualSim2Real: Open-source sim-to-real framework for humanoid visual loco-manipulation. Train in simulation, deploy zero-shot on real robots with RGB + proprioception for tasks like pick-and-place and door opening.
SURPRISE SCORE
0.00
Score Breakdown
SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE
SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)
99% OF STARS IN ARCHIVE
[BOT] SUSPECTED STAR BOT — SCORE PENALIZED. SIGNATURES:
S2 · NO EXTERNAL ISSUE/PR AUTHORS DESPITE 100+ STARS
S5 · PREDATES WINDOW, YET HALF+ OF ALL ITS STARS LANDED IN IT
Growth Telemetry
VELOCITY /D
3.87
ACCEL
-0.10
RETENTION
8.1%
PEAK 2026-04-24 · FORK-RETENTION 0.0% · 348 STARS / WINDOW
Author Audience
AUDIENCE
60,946
FOLLOWERS
10,906
OWNER ★
195,670
Engagement Signals
FORKS
33
ISSUE AUTH
0
PR AUTH
0
UNIQUE STARGAZERS 348 / 348 (DIVERSITY 1.00)
Why This Is A Finding
NVlabs/GR00T-VisualSim2Real собрал 348 звёзд за окно, тогда как у автора всего 10,906 подписчиков — эффективная аудитория ≈ 60,946. Это даёт surprise-индекс 0.0000634 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 0.0% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.
Related Findings
RANKS ABOVE 3% OF 6380 FINDINGS
METRICS IN CONTEXT
MEDIAN ACROSS ALL 6380 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW
METRICVALUEMEDIANΔ MEDPERCENTILE
SCORE0.000.00-0.00ABOVE 3%
VELOCITY3.873.99-0.13ABOVE 49%
RETENTION8.1%17.1%-9.0 PPABOVE 23%
FORKS3390-56ABOVE 21%
SURPRISE0.000.01-0.01ABOVE 3%