Arena 研究：前沿图像模型后训练需复合奖励，FLUX.2-dev 提升 69 Elo

How to design rewards for post-training frontier image models? Our research suggests human preference reward is necessary, but insufficient: A preference model may still reward outputs that look appealing but miss details, introduce unrequested content, or exhibit other forms of reward-hacking. We therefore optimize towards a composite reward: - Bradley-Terry reward model trained on ~5.6M pairwise human votes - Faithfulness reward from auto-generated prompt checklists evaluated by a vision-language model - Constraint reward covering explicit and implicit user intent - Anti-reward-hacking rubric rewards targeting failures such as garbled text and photorealism drift This post-training recipe improves two already-strong open image models: - Post-trained FLUX.2-dev gains 69 Elo points on our live T2I leaderboard, scoring 1202 - Post-trained Ideogram 4 gains 20 Elo points reaching a score of 1224 and surpassing all publicly listed open models (as of Sep 04, 2026). Offline ablations with Gem

来源：X：Arena (@arena) · x.lingyaoai.com