A model trained only to pass tests learned to add unrequested code and let errors pass silently. 🧪 Xiaomi fixed it by multiplying each test result by quality checklist scores. Result: MiMo-V2.6-Pro RL leads open weights models on Artificial Analysis’ Intelligence Index (46).

DeepLearning.AI @DeepLearningAI ·
A model trained only to pass tests learned to add unrequested code and let errors pass silently. 🧪…

更多来自 @DeepLearningAI

  1. #1

    This assistant remembers where you left your keys. No network connection. No remote server. Build one yourself in a course built in partnership with Qdrant and taught by Dylan Cou…

    DeepLearning.AI @DeepLearningAI ·
    This assistant remembers where you left your keys. No network connection. No re…
  2. #2

    An open model nearly matches Claude Mythos on cybersecurity capabilities: GLM-5.3 solves 12% and Claude Mythos 14% of ExploitBench exploit tasks, per Anthropic. Just $20.40 in tok…

    DeepLearning.AI @DeepLearningAI ·
    An open model nearly matches Claude Mythos on cybersecurity capabilities: GLM-5…
  3. #3

    This week in The Batch, Andrew's Letter covers how open weights model GLM-5.3 nearly matched Claude Mythos at exploiting vulnerabilities, 12% vs. 14%. Also inside: 🧪 Xiaomi's new…

    DeepLearning.AI @DeepLearningAI ·
    This week in The Batch, Andrew's Letter covers how open weights model GLM-5.3 n…
  4. #4

    Andrew Ng's AI Engineering Skills Map shows what to learn. AI Dev is where you hear from the people who built it in the real world. Keynotes from Andrew Ng and Yann LeCun. Enginee…

    DeepLearning.AI @DeepLearningAI ·
    Andrew Ng's AI Engineering Skills Map shows what to learn. AI Dev is where you…
  5. #5

    The Data Engineering Professional Certificate is now available on DeepLearningAI. Across four courses, design and build the systems that generate, ingest, store, transform, and se…

    DeepLearning.AI @DeepLearningAI ·
    视频封面
  6. #6

    Everyone knows that applying rigid testing requirements to early stage AI projects causes them to stall… but some companies do it anyway. Andrew Ng explains why AI engineering tac…

    DeepLearning.AI @DeepLearningAI ·
    Everyone knows that applying rigid testing requirements to early stage AI proje…

查看 @DeepLearningAI 的全部帖子