truthrl-ternary-reward-hallucination-reduction

【效率与协作】Reduce LLM hallucinations by training with a ternary reward signal that distinguishes correct answers, hallucinations, and abstentions. This technique incentivizes truthfulness over accuracy-only metrics, enabling safer, more calibrated language models through GRPO-based optimization.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准