Zihan Dong

Reliable Efficient AI Multi Modality

[Zihan Dong @CrD104]

CADWorld: a benchmark for long-horizon computer-use agents in mechanical CAD. 200 tasks, 11 FreeCAD workflows, scored by executable checks on real artifacts. Best agent: 17.5% vs 87% expert. @OpenAI @AnthropicAI @Alibaba_Qwen — run your agents on it. https://t.co/a4Km2kzWNe

[CAMEL-AI.org @CamelAIOrg]

📢 Join our CAMEL-AI Live Talk next Friday with Zihan Dong @CrD104 , AI Solutions Engineer at Lenovo, presenting CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design

He'll introduce CADWorld, a benchmark that evaluates how effectively AI agents can perform realistic mechanical CAD workflows by operating professional software directly through its GUI. It includes 200 tasks across 11 workflow categories, covering the end-to-end design process from modeling and assembly to manufacturing and simulation.

📅 October 16, 2026
🕐 14:00 BST | 9:00 EDT | 21:00 CST
🔗 Register: https://t.co/2Hm78axGEm

🐫 If you're interested in computer-use agents, CAD, or AI for engineering, don't miss this talk!

[Zekun Wang @zekunwang28]

As more context arrives, SCoL learns which layers to update to absorb new information while limiting forgetting.

Check out our latest work: Self-Consolidating Language Models
https://t.co/CMqfIzygrb
1/7

[Zihan Dong @CrD104]

In practice, that means more trustworthy stance/opinion detection out of everyday tools like review analysis, social listening, or survey coding, with no retraining needed.

[Zihan Dong @CrD104]

Long posts bury the one or two sentences that actually reveal someone's stance under filler and hedging — we built TEF to have the model weigh only the confident, decisive sentences instead of treating every line equally.

https://t.co/xOj9y9B85U

[Zihan Dong @CrD104]

There is no benchmark tells you if your agent could resolve your mechanical CAD design in accurate, editable, reliable, efficiently.
We designed the benchmark CADWORLD from professional prospective that really focused on the engineers workflow. https://t.co/a4Km2kzoXG