
平安科技多模态大模型工程师
社招全职3年以上计算机网络技术类地点:深圳状态:招聘
任职要求
1、硕士及以上学历,计算机、人工智能、模式识别、计算机视觉、NLP相关专业,博士优先; 2、3年及以上多模态大模型研发实战经验,完整深度参与过至少一款多模态理解模型从预训练、微调至业务落地全流程项目;有顶会论文、模型开源产出优先; 3、精通多模态大模型底层原理,熟练掌握LLM预训练、增量训练、模态编码器设计、跨模态对齐、SFT微调、RLHF/DPO对齐全套技术;熟悉ViT、CLIP、Qwen-VL、LLaVA等主流多模…
登录查看完整任职要求
微信扫码,1秒登录
工作职责
1、负责多模态大模型(MLLM)全流程算法研发,完成模型持续预训练、增量后训练、指令微调SFT、偏好对齐Alignment全链路工作;开展模型结构设计、模态融合方案设计、损失函数设计优化、训练策略迭代制定,提升模型综合能力。 2、攻坚跨模态语义对齐、多模态特征融合、异构图文/图表/文档数据解析理解等核心技术难点,优化模型在跨模态检索、视觉问答VQA、Chart/DocVQA、图文理解、长文档逻辑推理、图文生成等任务效果。 3、跟踪全球多模态大模型前沿技术演进,结合业务场景输出技术路线规划,主导模型迭代优化、技术难点攻关,沉淀可复用技术方案。 4、深耕业务场景,探索多模态模型与智能体Agent融合落地,围绕文档智能解析、grounding,图表推理、专业内容问答、复杂逻辑推理搭建垂域应用方案,实现算法技术和业务深度融合。 5、牵头多模态模型项目完整研发、迭代与落地交付,针对业务痛点定制模型优化方案,持续验证、复盘提升模型线上表现。
包括英文材料
学历+
模式识别+
https://www.mathworks.com/discovery/pattern-recognition.html
Pattern recognition is the process of classifying input data into objects, classes, or categories using computer algorithms based on key features or regularities.
https://www.microsoft.com/en-us/research/wp-content/uploads/2006/01/Bishop-Pattern-Recognition-and-Machine-Learning-2006.pdf
Pattern recognition has its origins in engineering, whereas machine learning grew out of computer science.
OpenCV+
https://learnopencv.com/getting-started-with-opencv/
At LearnOpenCV we are on a mission to educate the global workforce in computer vision and AI.
https://opencv.org/university/free-opencv-course/
This free OpenCV course will teach you how to manipulate images and videos, and detect objects and faces, among other exciting topics in just about 3 hours.
NLP+
https://www.youtube.com/watch?v=fNxaJsNG3-s&list=PLQY2H8rRoyvzDbLUZkbudP-MFQZwNmU4S
Welcome to Zero to Hero for Natural Language Processing using TensorFlow!
https://www.youtube.com/watch?v=R-AG4-qZs1A&list=PLeo1K3hjS3uuvuAXhYjV2lMEShq2UYSwX
Natural Language Processing tutorial for beginners series in Python.
https://www.youtube.com/watch?v=rmVRLeJRkl4&list=PLoROMvodv4rMFqRtEuo6SGjY4XbRIVRd4
The foundations of the effective modern methods for deep learning applied to NLP.
大模型+
https://www.youtube.com/watch?v=xZDB1naRUlk
You will build projects with LLMs that will enable you to create dynamic interfaces, interact with vast amounts of text data, and even empower LLMs with the capability to browse the internet for research papers.
https://www.youtube.com/watch?v=zjkBMFhNj_g
SFT+
https://cameronrwolfe.substack.com/p/understanding-and-using-supervised
Understanding how SFT works from the idea to a working implementation...
RLHF+
[英文] What is RLHF?
https://aws.amazon.com/what-is/reinforcement-learning-from-human-feedback/
Reinforcement learning from human feedback (RLHF) is a machine learning (ML) technique that uses human feedback to optimize ML models to self-learn more efficiently.
https://www.ibm.com/think/topics/rlhf
Reinforcement learning from human feedback (RLHF) is a machine learning technique in which a “reward model” is trained with direct human feedback, then used to optimize the performance of an artificial intelligence agent through reinforcement learning.
还有更多 •••