List your company

Context-aware generation on TriState-Bench Resistance corrupted context, prior correct (Sres)

33.25EMARR
-1.077.8416.7525.66Jun 9, 2026

Evaluation results

MethodLinks
ARRModel=Llama3-8B2026.0633.2538.59
ARRModel=Mistral-7B2026.0622.530.66
ARRModel=Llama2-13B2026.061723.21
ARRModel=Qwen2.5-7B2026.0615.7522.92
GreedyModel=Mistral-7B2026.065.2512.99
GreedyModel=Llama3-8B2026.064.59.78
AdaCADModel=Llama3-8B2026.063.259.67
AdaCADModel=Mistral-7B2026.062.2510.06
COIECDModel=Llama3-8B2026.061.756.29
GreedyModel=Llama2-13B2026.061.759.58
AdaCADModel=Llama2-13B2026.061.759.65
CoCoAModel=Llama2-13B2026.061.59.13
CoCoAModel=Llama3-8B2026.0617.68
COIECDModel=Mistral-7B2026.0618.61
GreedyModel=Qwen2.5-7B2026.06110.04
COIECDModel=Llama2-13B2026.0618.62
CADModel=Llama3-8B2026.060.757.11
AdaCADModel=Qwen2.5-7B2026.060.758.94
CoCoAModel=Mistral-7B2026.060.56.98
COIECDModel=Qwen2.5-7B2026.060.57.19
CADModel=Llama2-13B2026.060.57.75
CADModel=Mistral-7B2026.060.255.47
CADModel=Qwen2.5-7B2026.060.256.3
CoCoAModel=Qwen2.5-7B2026.060.256.19