From RLVR to RLSVR
Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement
AI 摘要
从 RLVR 到 RLSVR 任务转换可为开放式大语言模型自我改进引入自验证奖励 论文：https://huggingface.co/papers/2607.23802
From RLVR to RLSVR
Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement