Rosinality(@rosinality):https://t.co/NxJ1EM1qhY Improving exploration for rubric RL using OPSD to generate guidance for underexplored or penalized criteria. As combining criteria to construct rewards is common maybe it could be useful beyond rubric RL.

2026.07.21 10:56

Improving exploration for rubric RL using OPSD to generate guidance for underexplored or penalized criteria. As combining criteria to construct rewards is common maybe it could be useful beyond rubric RL.

106

커뮤니티로 전달

인기 있는 사용자

2M 팬

4.3M 팬

260.1K 팬

101.7K 팬

28.3K 팬

26.4M 팬

578K 팬

3.4M 팬

39.5K 팬

2.1M 팬

10.5M 팬

6.6M 팬

549.7K 팬

8.9M 팬

68.7K 팬