CNNVD-202509-3908 Information
Sep 25, 2025
cve
CNNVD ID
CNNVD-202509-3908
Related CVE
- CNNVD Published: 2025-09-25
Description (Chinese)
VLA-RL是lg(x)个人开发者的一个视觉语言动作模型。 VLA-RL存在代码问题漏洞，该漏洞源于对文件experiments/robot/bridge/reasoning_server.py中参数Message的错误操作，可能导致反序列化攻击。
Description (English)
VLA-RL is a visual language action model for lg (x) personal developers. VLA-RL has a code problem loophole, which stems from the wrong operation of Message, the parameter in file files/robot/bridge/reasing server.py, which could lead to a back-serialization attack.
Hazard Level
High
Vulnerability Type
代码问题
Affected Vendor
个人开发者
Published
2025-09-25
Last Modified
2026-02-24
References
https://github.com/GuanxingLu/vlarl/issues/18 https://github.com/GuanxingLu/vlarl/issues/18#issue-3408978610 https://vuldb.com/?ctiid.325846 https://vuldb.com/?id.325846 https://vuldb.com/?submit.653279
Share on: