Eval scores per skill
Each bar is one agent model with or without the plugin, sorted by mean score.
With pluginWithout plugin
solve-inverse-problem (pylops plugin)
Mean score over 6 cases, 0–1 scale.
debug-inversion (pylops plugin)
Mean score over 9 cases, 0–1 scale.
tune-regularization (pylops plugin)
Mean score over 7 cases, 0–1 scale.
solve-nonsmooth-problem (pyproximal plugin)
Mean score over 6 cases, 0–1 scale.