List your company

Long-form Writing on Creative-W.B.

83.17ScoreQwen2.5-14B-Writing-RL
15.850833.327950.80568.2821Jun 6, 2025

Evaluation results

MethodLinks
Qwen2.5-14B-Writing-RLSFT=✓, RL=PPO2025.0683.17
Llama3.1-8B-Writing-RLSFT=✓, RL=PPO2025.0682.67
GPT-4oSFT=–, RL=–2025.0680.45
Qwen2.5-14B-WritingBench-SFTSFT=✓, RL=✗2025.0679.27
Llama3.1-8B-WritingBench-SFTSFT=✓, RL=✗2025.0677.7
Qwen-PlusSFT=–, RL=–2025.0676.78
Llama3.1-8B-Reference-SFTSFT=✓, RL=✗2025.0676.7
Qwen2.5-7B-Writing-RLSFT=✓, RL=GRPO2025.0675.41
Qwen2.5-7B-Writing-RLSFT=✓, RL=PPO2025.0673.19
Qwen2.5-7B-WritingBench-SFT (12k)SFT=✓, RL=✗2025.0670.24
Qwen2.5-7B-WritingBench-SFT (24k)SFT=✓, RL=✗2025.0669.6
Qwen2.5-7B-Reference-SFTSFT=✓, RL=✗2025.0668.89
Longwriter-Zero-32BSFT=✓, RL=GRPO2025.0661.14
Qwen2.5-14B-InstructSFT=✗, RL=✗2025.0656.89
Qwen2.5-7B-InstructSFT=✗, RL=✗2025.0649.29
Llama3.1-8B-InstructSFT=✗, RL=✗2025.0648.83
Longwriter-9BSFT=✓, RL=DPO2025.0644.15
Suri-7BSFT=✓, RL=✗2025.0618.44