List your company

Robot Failure Analysis (MCQ) on RoboFAC Simulation

93FD ScoreKITE+Qwen2.5-7B+QLoRA
35.850.6565.580.35Apr 8, 2026

Evaluation results

MethodLinks
KITE+Qwen2.5-7B+QLoRAFinetuned=true2026.04936992
RoboFAC-7BFinetuned=true2026.04916394
KITE + Qwen2.5-VL-7BFinetuned=false2026.04884455
GPT-4oFinetuned=false2026.04642171
Qwen2.5-VL-7BFinetuned=false2026.04522622
Gemini-2.0Finetuned=false2026.04482775
Qwen2.5-VL-3BFinetuned=false2026.0438451