tracked author

Jubayer Ibn Hamid

Equal-contribution first author of SPIRAL. Public homepage lists Stanford CS PhD status, SAIL, Google student researcher role, and advisors Dorsa Sadigh and Chelsea Finn.

1 archived notes X: high HomepageGitHubX

Related Notes

按论文归档时间排序,展示该作者在本站已经出现的材料。

Jul 03, 2026

SPIRAL: Learning to Search and Aggregate

SPIRAL 从共享策略生成的 8 条 search traces 中随机抽取 4 个互异的四元集合,再为每个集合生成 4 条 aggregation traces;它用集合聚合成功率的 participation averaged advantage 更新搜索轨迹,并用同集合内中心化 reward 更新聚合轨迹,从而把 parallel search 与 model based aggregation 放进同一个最终答案 rewar...

2606.23595-spiral-learning-search-aggregate Test-Time ScalingCredit AssignmentRL Algorithm