List your company

Language Understanding on MMLU (Baseline/Final Accuracy)

71.3Baseline AccFlipLLM
28.86839.88450.961.916Dec 10, 2025

Evaluation results

MethodLinks
FlipLLMType=LLM, Target Model=DeepSeek-V2, Parameters=7B (MoE)2025.1271.30.19
FlipLLMType=LLM, Target Model=LLaMA 3.1 8B, Parameters=8B2025.1269.90.21
FlipLLMType=LLM, Target Model=GPT-2 Large, Parameters=774M2025.1230.50.35