Toward Training Superintelligent Software Agents Through Self-Play SWE-RLarxiv.org1 point·pama··0 commentsOpen articleSaveView on HN