List your company

Agentic Task on ResearchQA

73.7ScoreDR-Rubric-8B (GPT-5)
62.67665.53868.471.262May 31, 2026

Evaluation results

MethodLinks
DR-Rubric-8B (GPT-5)Training=SFT+RL, 1K2026.0573.7
DR-Rubric-8B (BS-3)Training=SFT+RL, 3K2026.0572.4
DR-Rubric-8B (Gemini)Training=SFT+RL, 1K2026.0571.7
Qwen3-8B-SFTTraining=SFT, 1K2026.0569.9
DR-Tulu-RL-8BTraining=SFT+RL, 25K2026.0567.1
DR-Tulu-SFT-8BTraining=SFT, 16K2026.0566.7
Qwen3-8B2026.0566.6
WebExplorer-8BTraining=SFT+RL, 25K2026.0566.1
Qwen2.5-7B2026.0565.5
Search-R1-7BTraining=RL, 90K2026.0563.1