List your company

Personalized Interaction on AI2 ARC Synthetic

7.38Personalization ScorePPOpt
4.50965.254866.7452Feb 12, 2026

Evaluation results

MethodLinks
PPOptBackbone=Llama-3-8b-instruct2026.027.389.5
PPOptBackbone=GPT-oss-20b2026.027.349.48
PPOptBackbone=Qwen3-8b2026.026.69.3
VanillaBackbone=Llama-3-8b-instruct2026.024.629.56
VanillaBackbone=Qwen3-8b2026.024.629.56
VanillaBackbone=GPT-oss-20b2026.024.629.56