Story

hackernews_ai ยท Nov 14, 2026 ยท news

Source brief

Ask HN: Is post-training make AI models dumber?

news.ycombinator.comNov 14, 2026
original source linked

In brief

Post-training is making models better at benchmarks. They seem more consistent at coding, but dumber to me and I want to know what y'all are noticing. Here are the big three I'm noticing (mostly in claude): 1) Agents...

Feed lens
agent

Continue reading

Read the original at news.ycombinator.com โ†’Open in live feedRead that dayโ€™s brief

Earlier in this thread 4 items