FirstFT: the day's biggest stories
This also applies to LLM-generated evaluation. Ask the same LLM to review the code it generated and it will tell you the architecture is sound, the module boundaries clean and the error handling is thorough. It will sometimes even praise the test coverage. It will not notice that every query does a full table scan if not asked for. The same RLHF reward that makes the model generate what you want to hear makes it evaluate what you want to hear. You should not rely on the tool alone to audit itself. It has the same bias as a reviewer as it has as an author.
。新收录的资料对此有专业解读
Unlike my black and white only pick, it also includes a color touchscreen, which I found helpful for both setup and any random errors or notices that came up during everyday use.
Not the day you're after? Here's the solution to yesterday's Connections.