panel | column | model | effect [95% CI] (* Holm p < 0.05); two-mark rows: circle first, diamond second Contract review | Hidden instruction | Haiku 5.5 | -0.15 [ -0.55, +0.25] Contract review | Hidden instruction | Opus 5.5 | -0.05 [ -0.30, +0.20] Contract review | Hidden instruction | Fable 5.1 | +0.15 [ -0.15, +0.40] Contract review | Hidden instruction | GPT-6 Luna | +0.60 [ +0.30, +0.90]* Contract review | Hidden instruction | GPT-6.1 Sol | -0.40 [ -0.80, +0.00] Contract review | Hidden instruction | dots.3 note | +20.75 [ +16.75, +24.45]* Contract review | Hidden instruction | Gemini 3.6 Flash | -3.40 [ -5.65, -1.10]* Contract review | Hidden instruction | Nemotron 3 Ultra | -0.75 [ -2.45, +0.90] Contract review | Hidden instruction | Xing 4.0 | -1.30 [ -7.40, +4.65] Contract review | Combined framing | Haiku 5.5 | -0.40 [ -0.80, +0.00] Contract review | Combined framing | Opus 5.5 | +0.25 [ -0.05, +0.55] Contract review | Combined framing | Fable 5.1 | +0.05 [ -0.30, +0.35] Contract review | Combined framing | GPT-6 Luna | +0.75 [ +0.40, +1.10]* Contract review | Combined framing | GPT-6.1 Sol | -0.10 [ -0.55, +0.30] Contract review | Combined framing | dots.3 note | +7.20 [ +3.20, +11.90]* Contract review | Combined framing | Gemini 3.6 Flash | +2.35 [ -1.20, +5.95] Contract review | Combined framing | Nemotron 3 Ultra | +2.65 [ +0.25, +5.10] Contract review | Combined framing | Xing 4.0 | -2.00 [ -8.80, +4.75] Interpretive drift | Favoured items first | Haiku 5.5 | -6.2 [ -9.1, -3.5]* -10.8 [ -14.1, -7.4]* Interpretive drift | Favoured items first | Opus 5.5 | -4.5 [ -5.8, -3.3]* -8.1 [ -9.5, -6.7]* Interpretive drift | Favoured items first | Fable 5.1 | -0.7 [ -2.6, +1.2] -5.0 [ -7.5, -2.5]* Interpretive drift | Favoured items first | GPT-6 Luna | -9.3 [ -12.6, -5.9]* -0.4 [ -3.6, +2.8] Interpretive drift | Favoured items first | GPT-6.1 Sol | -4.5 [ -5.8, -3.3]* -3.2 [ -4.2, -2.2]* Interpretive drift | Favoured items last | Haiku 5.5 | +8.2 [ +2.5, +13.5]* +1.1 [ -2.8, +5.1] Interpretive drift | Favoured items last | Opus 5.5 | +3.1 [ +1.8, +4.4]* +3.7 [ +2.1, +5.4]* Interpretive drift | Favoured items last | Fable 5.1 | +1.8 [ +0.1, +3.5] +0.1 [ -1.6, +1.8] Interpretive drift | Favoured items last | GPT-6 Luna | +5.6 [ +2.6, +8.7]* +0.6 [ -3.0, +4.2] Interpretive drift | Favoured items last | GPT-6.1 Sol | +2.5 [ +1.5, +3.4]* +4.1 [ +2.9, +5.3]* Interpretive drift | Slanted glosses | Haiku 5.5 | +7.8 [ +4.0, +11.7]* +14.3 [ +11.6, +17.0]* Interpretive drift | Slanted glosses | Opus 5.5 | -7.8 [ -8.9, -6.6]* -0.4 [ -1.8, +1.1] Interpretive drift | Slanted glosses | Fable 5.1 | -0.8 [ -2.8, +1.1] +3.0 [ +1.3, +4.7] Interpretive drift | Slanted glosses | GPT-6 Luna | +5.5 [ +2.0, +9.0]* +7.4 [ +4.5, +10.3]* Interpretive drift | Slanted glosses | GPT-6.1 Sol | -1.9 [ -3.1, -0.8]* +4.4 [ +3.3, +5.4]* IP assignment | Assertion only | Haiku 5.5 | -0.2 [ -0.4, +0.1] +0.1 [ +0.0, +0.2] IP assignment | Assertion only | Opus 5.5 | +0.0 [ +0.0, +0.1] -0.5 [ -0.8, -0.3]* IP assignment | Assertion only | Fable 5.1 | -0.2 [ -0.6, +0.2] -0.7 [ -1.0, -0.4]* IP assignment | Assertion only | GPT-6 Luna | -0.7 [ -1.5, +0.2] +0.1 [ -0.1, +0.4] IP assignment | Assertion only | GPT-6.1 Sol | -0.2 [ -0.4, -0.1] +0.0 [ +0.0, +0.0] IP assignment | Adversarial bundle | Haiku 5.5 | +4.7 [ +3.9, +5.7]* +13.7 [ +12.1, +15.4]* IP assignment | Adversarial bundle | Opus 5.5 | +9.4 [ +8.9, +9.9]* +25.0 [ +23.6, +26.4]* IP assignment | Adversarial bundle | Fable 5.1 | +9.5 [ +8.5, +10.5]* +23.6 [ +21.5, +26.2]* IP assignment | Adversarial bundle | GPT-6 Luna | +7.9 [ +6.0, +9.8]* +15.9 [ +12.4, +19.5]* IP assignment | Adversarial bundle | GPT-6.1 Sol | +11.1 [ +10.2, +12.1]* +25.0 [ +24.0, +26.2]*