Unverified50% confidenceFactExact time
Both Qwen models trained solely in WildAWM synthetic environments achieved significant improvements on all 8 OOD benchmarks, indicating that the acquired capabilities generalize to unseen computer-use tasks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
Related Entities
Related Claims
UnverifiedThe Qwen2.5-Max-0902 update conducted targeted post-training optimization focused on two areas: programming and collaborative office work.75% similarUnverifiedQwen 2.5系列训练数据超过18万亿Token，采用Transformer解码器架构并引入分组查询注意力（GQA）机制66% similarUnverifiedAchieving a significant improvement in programming capabilities through post-training on the same base model provides a new model iteration approach for teams with limited computing resources.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073290API
curl https://kongchang.com/api/v1/knowledge/claims/1073290MCP
get_claim(id=1073290)