AdaPRA adds adaptive process rubrics for data science agents
Five attempts per task raised an 80B Qwen model’s leaderboard percentile by 24.5 on DSPredict-Easy and 14.8 on DSPredict-Hard.
1 stories, newest first · one card per story
Five attempts per task raised an 80B Qwen model’s leaderboard percentile by 24.5 on DSPredict-Easy and 14.8 on DSPredict-Hard.
Today 1.6K readers · 2.5K views · 778 reads · all time 189K views