# Cua — NeurIPS 2025: Computer-Use Agent Papers You Should Know About

- Company: Cua (cua.ai)
- Announced: 2025-11-28
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://cua.ai/blog/neurips-2025-cua-papers
- Record: https://forck.live/items/16825-neurips-2025-computer-use-agent-papers-you-should-know-about
- Subject: Cua Driver / Lume / Cua Bench / Cua Fleets

Cua compiled a curated list of 45 computer-use agent papers presented at NeurIPS 2025, covering benchmarks, safety, grounding, visual reasoning, and agent architectures. The post highlights that the benchmark landscape is maturing with evaluations across macOS, professional tools, and real-world websites, and that safety is becoming a first-class concern with multiple papers documenting agent failures under adversarial inputs, privacy requirements, or misuse scenarios.

## Evidence

Verbatim from https://cua.ai/blog/neurips-2025-cua-papers:

> We did that work for you. We're excited to announce that Cua will be at NeurIPS 2025 , and we've compiled a curated list of 45 papers focused specifically on Computer-Use Agents—covering benchmarks, safety, grounding, visual reasoning, and agent architectures.

---

Record: https://forck.live/items/16825-neurips-2025-computer-use-agent-papers-you-should-know-about
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md