Stronger models defer maze turns only when humans hold the advantage
Reliable solo solvers almost never deferred without a human route advantage, but deferred on about 80% of initial decisions when the human path was shorter.
36
- Mentions · 7d0
- Mentions · 30d1
- All mentions2
- First seenSep 11, 2026
- Last mentioned23d ago
Timeline
2 articles · newest firstinternlm/Atria-Dawn-Preview — new model trending #30 on Hugging Face
Shanghai AI Laboratory released Atria-Dawn-Preview, a 744B-parameter MoE agentic model with 256K context built on GLM-5.2, targeting research, coding, and cybersecurity automation.
70
Appears with
- Qwen 3.8 27Bai model · 1
- Command A+ai model · 1
- Hugging Faceproduct · 1
- Atria-Dawn-Previewproduct · 1
- Claude Opus 5ai model · 1
- Atria-Dawn-Previewai model · 1
- ModelScopeproduct · 1
- GLM-5.2ai model · 1
- Command Rai model · 1
- KIMI K3ai model · 1
- Qwen 3.8 Maxai model · 1
- DeepSeek V4 Pro 0813ai model · 1
- Shanghai AI Laboratoryorganization · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.