微信安全-大模型安全算法工程师-多模态方向
Tencent
微信安全-大模型安全算法工程师-多模态方向
Job Description
微信安全-大模型安全算法工程师-多模态方向
📋 Job Overview
The role focuses on developing multimodal safety algorithms for large models in WeChat's security team. Responsibilities include building specialized safety models to enhance machine review accuracy for videos, live streams, search, and AIGC products; creating safety guardrails for generative AI features like text-to-image and image-to-text; and developing safe response mechanisms for multimodal Q&A scenarios to mitigate user risks.
📍 Location: Beijing, China
🏢 Business Unit: WXG
📄 Full Description
2.建设多模态大模型安全护栏能力：针对微信生态的文生图、文生视频、图生文等功能做内容安全检测，包括但不限于：deepfake的识别、AIGC检测、侵权问题、底线问题的识别等；
3.建设多模态大模型安全代答能力：针对多模态大模型问答场景涉及到的用户风险问题，进行安全可信代答。
🎯 Key Responsibilities
- Build multimodal safety vertical large models to improve machine review accuracy and coverage, including but not limited to short videos, live streams, search, and AIGC products.
- Construct multimodal large model safety guardrail capabilities for WeChat ecosystem features such as text-to-image, text-to-video, image-to-text, etc., including content safety detection for deepfake recognition, AIGC detection, infringement issues, and bottom-line problem identification.
- Develop multimodal large model safety proxy response capabilities to address user risk issues in multimodal large model Q&A scenarios with safe and trustworthy responses.
Locations
- Beijing, China
Salary
300,000 - 800,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Responsibilities
- Build multimodal safety vertical large models to improve machine review accuracy and coverage, including but not limited to short videos, live streams, search, and AIGC products.
- Construct multimodal large model safety guardrail capabilities for WeChat ecosystem features such as text-to-image, text-to-video, image-to-text, etc., including content safety detection for deepfake recognition, AIGC detection, infringement issues, and bottom-line problem identification.
- Develop multimodal large model safety proxy response capabilities to address user risk issues in multimodal large model Q&A scenarios with safe and trustworthy responses.
Tags & Categories
Answer 10 quick questions to check your fit for 微信安全-大模型安全算法工程师-多模态方向 @ Tencent.
Related Books and Jobs
微信安全-大模型安全算法工程师-多模态方向
Tencent
微信安全-大模型安全算法工程师-多模态方向
Job Description
微信安全-大模型安全算法工程师-多模态方向
📋 Job Overview
The role focuses on developing multimodal safety algorithms for large models in WeChat's security team. Responsibilities include building specialized safety models to enhance machine review accuracy for videos, live streams, search, and AIGC products; creating safety guardrails for generative AI features like text-to-image and image-to-text; and developing safe response mechanisms for multimodal Q&A scenarios to mitigate user risks.
📍 Location: Beijing, China
🏢 Business Unit: WXG
📄 Full Description
2.建设多模态大模型安全护栏能力：针对微信生态的文生图、文生视频、图生文等功能做内容安全检测，包括但不限于：deepfake的识别、AIGC检测、侵权问题、底线问题的识别等；
3.建设多模态大模型安全代答能力：针对多模态大模型问答场景涉及到的用户风险问题，进行安全可信代答。
🎯 Key Responsibilities
- Build multimodal safety vertical large models to improve machine review accuracy and coverage, including but not limited to short videos, live streams, search, and AIGC products.
- Construct multimodal large model safety guardrail capabilities for WeChat ecosystem features such as text-to-image, text-to-video, image-to-text, etc., including content safety detection for deepfake recognition, AIGC detection, infringement issues, and bottom-line problem identification.
- Develop multimodal large model safety proxy response capabilities to address user risk issues in multimodal large model Q&A scenarios with safe and trustworthy responses.
Locations
- Beijing, China
Salary
300,000 - 800,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Responsibilities
- Build multimodal safety vertical large models to improve machine review accuracy and coverage, including but not limited to short videos, live streams, search, and AIGC products.
- Construct multimodal large model safety guardrail capabilities for WeChat ecosystem features such as text-to-image, text-to-video, image-to-text, etc., including content safety detection for deepfake recognition, AIGC detection, infringement issues, and bottom-line problem identification.
- Develop multimodal large model safety proxy response capabilities to address user risk issues in multimodal large model Q&A scenarios with safe and trustworthy responses.
Tags & Categories
Answer 10 quick questions to check your fit for 微信安全-大模型安全算法工程师-多模态方向 @ Tencent.
Related Books and Jobs