List your company

Behavior Prediction on RecRM-Bench

77.78AccuracyOurs
29.149641.774854.467.0252May 12, 2026

Evaluation results

MethodLinks
Ours2026.0577.7881.46
Deepseek-V3.2Thinking=false2026.0547.852.1
LongCat-Flash-ChatThinking=false2026.0546.4952.22
Qwen3-MaxThinking=true2026.0544.4748.5
Qwen3-MaxThinking=false2026.0542.7952.08
GPT-4.12026.0539.2849.31
Deepseek-V3.2Thinking=true2026.0535.1750.76
LongCat-Flash-ThinkingThinking=true2026.0531.0245.41