3月12日消息，根据今年政府工作报告，今年拟发行超长期特别国债1.3万亿元，持续支持“两重”建设、“两新”工作等。其中安排8000亿元超长期特别国债用于“两重”建设，安排4500亿元支持“两新”工作。这意味着还有剩余的500亿元超长期特别国债资金用于其他领域。近日，财政部公开《关于修订2026年政府收支分类科目的通知》，或从侧面给出答案，即超长期特别国债资金还可用于金融方面支出。
圖像來源，AFP via Getty Images
。51吃瓜对此有专业解读
longer-lived certs that have the risk of missing
Credit: Ecovacs
。手游是该领域的重要参考
The BrokenMath benchmark (NeurIPS 2025 Math-AI Workshop) tested this in formal reasoning across 504 samples. Even GPT-5 produced sycophantic “proofs” of false theorems 29% of the time when the user implied the statement was true. The model generates a convincing but false proof because the user signaled that the conclusion should be positive. GPT-5 is not an early model. It’s also the least sycophantic in the BrokenMath table. The problem is structural to RLHF: preference data contains an agreement bias. Reward models learn to score agreeable outputs higher, and optimization widens the gap. Base models before RLHF were reported in one analysis to show no measurable sycophancy across tested sizes. Only after fine-tuning did sycophancy enter the chat. (literally)
First FT: the day’s biggest stories。关于这个话题，超级权重提供了深入分析