跳到正文
𝕏 Arena.ai(x.com)·· 2 天前品質評分44
重點導讀

LMSYS Arena 更新了 Arena Alignment Index,調整偵測 AI 助理出錯的方式,希望涵蓋更多情況、判斷更可靠。

  • 這份指數會看 AI 助理在真實使用情境中,是否擅自做事、謊稱工作已完成,或把使用者沒說過、沒做過的事算到使用者頭上。新版排名中,OpenAI 包辦前七名。
  • Claude Fable 5.2 首次上榜排第 8,Kimi K3.1 Pro 是進步幅度最大的開放模型,也排在第 9。這提醒我們,模型排名不只看能不能回答問題,也要看它會不會越權或誤報工作狀態。

AI 伴隨問答

自備 Key 深度分析

針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:

Arena.ai
Arena.ai@arena
·2 天前
𝕏

The Arena Alignment Index has been updated!

We've refined how our signals detect model failures, providing broader coverage and more reliable judgements. Based on the updated methodology:
- @OpenAI continues to dominate, holding all seven top positions.
- @Anthropic's latest release, Claude Fable 5.2, debuts at #8 with a score of 78.2. Claude Opus 5.5, which ranked #6 on the previous index, is now #17.
- Kimi K3.1 Pro by @Kimi_Moonshot is the biggest upgrade over its predecessor (more info in thread).
- Most open source models are trailing closed releases. The best open model - Kimi K3.1 Pro - occupies #9, and the second best, Mimo-V2.6-Flash, is #26 on the Index, followed by Hy4, Beam, Inkling, DeepSeek and GLM.

The Top-6 lab:
- #1 @OpenAI GPT-6.2 Sol
- #2 @AnthropicAI Claude Fable 5.2
- #3 @Kimi_Moonshot Kimi K3.1 Pro
- #4 @GoogleDeepmind Gemini 4 Hydrogen
- #5 @SpaceXSI Grok 4.7
- #6 @AIatMeta Muse Ember

Alignment Index measures how agents behave in real user sessions across three dimensions:
- Unauthorized Action: Doing what the user didn’t ask for or approve
- Deceptive Completion: Saying work is done, checked or working when it isn't
- False Attribution: Crediting the user with something they never said or did

More insights in the thread🧵

推文來源 · X (Twitter)在 X 參與討論
來源: Arena.ai (x.com)
在 X 查看推文

讀者討論

理性交流 · 即時討論