Static baseline report

split	method	condition	Hit@3 (95% CI over seeds)	MRR@10	nDCG@10	Δ Hit@3 vs none [95% CI]	Δ Hit@3 vs shuffled [95% CI]
within	none	none	0.382 [0.353, 0.411]	0.334	0.384
within	ranker	genuine	0.410 [0.380, 0.439]	0.356	0.404	+0.027 [+0.012, +0.042]	+0.025 [+0.010, +0.040]
within	ranker	shuffled	0.384 [0.354, 0.415]	0.333	0.384	+0.002 [-0.002, +0.006]
within	ranker	noisy	0.382 [0.357, 0.407]	0.334	0.385	+0.000 [-0.007, +0.007]	-0.002 [-0.009, +0.005]
within	static-tuned	none	0.426 [0.399, 0.452]	0.366	0.413	+0.043 [+0.026, +0.060]
test	none	none	0.386 [0.361, 0.411]	0.334	0.378
test	ranker	genuine	0.407 [0.386, 0.428]	0.353	0.398	+0.021 [+0.011, +0.031]	+0.021 [+0.011, +0.031]
test	ranker	shuffled	0.386 [0.361, 0.411]	0.334	0.378	+0.000 [+0.000, +0.000]
test	ranker	noisy	0.389 [0.362, 0.416]	0.341	0.384	+0.003 [-0.001, +0.008]	+0.003 [-0.001, +0.008]
test	static-tuned	none	0.420 [0.393, 0.448]	0.365	0.406	+0.034 [+0.022, +0.046]

method	feedback phase (s, mean)	eval search p50 (ms)
none	0	2.2
ranker	3.9	2.2