混元多模态知识图谱算法研究员（北京）
Tencent
混元多模态知识图谱算法研究员（北京）
Job Description
混元多模态知识图谱算法研究员（北京）
📋 Job Overview
The role of Multimodal Knowledge Graph Algorithm Researcher at Tencent focuses on advancing AI capabilities through innovative data handling and knowledge representation. Responsibilities include developing algorithms for cross-modal data understanding, constructing large-scale multimodal knowledge graphs, and exploring data applications to enhance large language models. This position is based in Beijing and aims to drive continuous improvements in model generation performance via iterative knowledge graph enhancements.
📍 Location: Shenzhen, China
🏢 Business Unit: TEG
📄 Full Description
1.数据理解能力：负责研究跨模态统一表征学习算法，实现跨模态/混合模态/指令遵循的高效数据检索、多模态数据分层聚类、数据多样性评估与数据剪枝均衡能力；
2.数据体系建设：参与设计、构建大规模多模态世界知识图谱，支持大模型对领域细粒度知识理解与推理能力；探索知识图谱构建的前沿技术，研究跨模态实体链指、属性/关系挖掘、泛知识挖掘算法；
3.数据应用探索：将各模态大模型训练数据挂载到知识图谱体系中，结合模型评测/实验下钻分析模型训练数据在数量、质量、配比上的问题，驱动知识图谱持续迭代，最终带来大模型生成效果的持续提升。
2.数据体系建设：参与设计、构建大规模多模态世界知识图谱，支持大模型对领域细粒度知识理解与推理能力；探索知识图谱构建的前沿技术，研究跨模态实体链指、属性/关系挖掘、泛知识挖掘算法；
3.数据应用探索：将各模态大模型训练数据挂载到知识图谱体系中，结合模型评测/实验下钻分析模型训练数据在数量、质量、配比上的问题，驱动知识图谱持续迭代，最终带来大模型生成效果的持续提升。
🎯 Key Responsibilities
- Research cross-modal unified representation learning algorithms to enable efficient data retrieval, multimodal data clustering, diversity assessment, and pruning for cross-modal/mixed-modal/instruction-following scenarios
- Participate in designing and building large-scale multimodal world knowledge graphs to support fine-grained domain knowledge understanding and reasoning in large models
- Explore cutting-edge technologies for knowledge graph construction, including cross-modal entity linking, attribute/relation mining, and general knowledge mining algorithms
- Integrate training data from various modalities into the knowledge graph system, analyze issues in data quantity, quality, and ratios through model evaluation and experiments, drive iterative improvements to the knowledge graph, and ultimately enhance large model generation performance
✅ Required Qualifications
- Strong background in AI, machine learning, or related fields
- Experience in algorithm research and development
⭐ Preferred Qualifications
- Prior work on knowledge graphs or multimodal AI systems
- Familiarity with large-scale data processing and model training
🛠️ Required Skills
- Data understanding and analysis
- Algorithm design for representation learning and knowledge mining
- Cross-modal and multimodal data processing
- Knowledge graph construction and integration
- Model evaluation and experimental analysis
Locations
- Shenzhen, China
Salary
Estimated Salary Rangemedium confidence
400,000 - 800,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Skills Required
- Data understanding and analysisintermediate
- Algorithm design for representation learning and knowledge miningintermediate
- Cross-modal and multimodal data processingintermediate
- Knowledge graph construction and integrationintermediate
- Model evaluation and experimental analysisintermediate
Required Qualifications
- Strong background in AI, machine learning, or related fields (experience)
- Experience in algorithm research and development (experience)
Preferred Qualifications
- Prior work on knowledge graphs or multimodal AI systems (experience)
- Familiarity with large-scale data processing and model training (experience)
Responsibilities
- Research cross-modal unified representation learning algorithms to enable efficient data retrieval, multimodal data clustering, diversity assessment, and pruning for cross-modal/mixed-modal/instruction-following scenarios
- Participate in designing and building large-scale multimodal world knowledge graphs to support fine-grained domain knowledge understanding and reasoning in large models
- Explore cutting-edge technologies for knowledge graph construction, including cross-modal entity linking, attribute/relation mining, and general knowledge mining algorithms
- Integrate training data from various modalities into the knowledge graph system, analyze issues in data quantity, quality, and ratios through model evaluation and experiments, drive iterative improvements to the knowledge graph, and ultimately enhance large model generation performance
Tags & Categories
Answer 10 quick questions to check your fit for 混元多模态知识图谱算法研究员（北京） @ Tencent.
10 Questions
~2 Minutes
Instant Score
Related Books and Jobs
混元多模态知识图谱算法研究员（北京）
Tencent
混元多模态知识图谱算法研究员（北京）
Job Description
混元多模态知识图谱算法研究员（北京）
📋 Job Overview
The role of Multimodal Knowledge Graph Algorithm Researcher at Tencent focuses on advancing AI capabilities through innovative data handling and knowledge representation. Responsibilities include developing algorithms for cross-modal data understanding, constructing large-scale multimodal knowledge graphs, and exploring data applications to enhance large language models. This position is based in Beijing and aims to drive continuous improvements in model generation performance via iterative knowledge graph enhancements.
📍 Location: Shenzhen, China
🏢 Business Unit: TEG
📄 Full Description
1.数据理解能力：负责研究跨模态统一表征学习算法，实现跨模态/混合模态/指令遵循的高效数据检索、多模态数据分层聚类、数据多样性评估与数据剪枝均衡能力；
2.数据体系建设：参与设计、构建大规模多模态世界知识图谱，支持大模型对领域细粒度知识理解与推理能力；探索知识图谱构建的前沿技术，研究跨模态实体链指、属性/关系挖掘、泛知识挖掘算法；
3.数据应用探索：将各模态大模型训练数据挂载到知识图谱体系中，结合模型评测/实验下钻分析模型训练数据在数量、质量、配比上的问题，驱动知识图谱持续迭代，最终带来大模型生成效果的持续提升。
2.数据体系建设：参与设计、构建大规模多模态世界知识图谱，支持大模型对领域细粒度知识理解与推理能力；探索知识图谱构建的前沿技术，研究跨模态实体链指、属性/关系挖掘、泛知识挖掘算法；
3.数据应用探索：将各模态大模型训练数据挂载到知识图谱体系中，结合模型评测/实验下钻分析模型训练数据在数量、质量、配比上的问题，驱动知识图谱持续迭代，最终带来大模型生成效果的持续提升。
🎯 Key Responsibilities
- Research cross-modal unified representation learning algorithms to enable efficient data retrieval, multimodal data clustering, diversity assessment, and pruning for cross-modal/mixed-modal/instruction-following scenarios
- Participate in designing and building large-scale multimodal world knowledge graphs to support fine-grained domain knowledge understanding and reasoning in large models
- Explore cutting-edge technologies for knowledge graph construction, including cross-modal entity linking, attribute/relation mining, and general knowledge mining algorithms
- Integrate training data from various modalities into the knowledge graph system, analyze issues in data quantity, quality, and ratios through model evaluation and experiments, drive iterative improvements to the knowledge graph, and ultimately enhance large model generation performance
✅ Required Qualifications
- Strong background in AI, machine learning, or related fields
- Experience in algorithm research and development
⭐ Preferred Qualifications
- Prior work on knowledge graphs or multimodal AI systems
- Familiarity with large-scale data processing and model training
🛠️ Required Skills
- Data understanding and analysis
- Algorithm design for representation learning and knowledge mining
- Cross-modal and multimodal data processing
- Knowledge graph construction and integration
- Model evaluation and experimental analysis
Locations
- Shenzhen, China
Salary
Estimated Salary Rangemedium confidence
400,000 - 800,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Skills Required
- Data understanding and analysisintermediate
- Algorithm design for representation learning and knowledge miningintermediate
- Cross-modal and multimodal data processingintermediate
- Knowledge graph construction and integrationintermediate
- Model evaluation and experimental analysisintermediate
Required Qualifications
- Strong background in AI, machine learning, or related fields (experience)
- Experience in algorithm research and development (experience)
Preferred Qualifications
- Prior work on knowledge graphs or multimodal AI systems (experience)
- Familiarity with large-scale data processing and model training (experience)
Responsibilities
- Research cross-modal unified representation learning algorithms to enable efficient data retrieval, multimodal data clustering, diversity assessment, and pruning for cross-modal/mixed-modal/instruction-following scenarios
- Participate in designing and building large-scale multimodal world knowledge graphs to support fine-grained domain knowledge understanding and reasoning in large models
- Explore cutting-edge technologies for knowledge graph construction, including cross-modal entity linking, attribute/relation mining, and general knowledge mining algorithms
- Integrate training data from various modalities into the knowledge graph system, analyze issues in data quantity, quality, and ratios through model evaluation and experiments, drive iterative improvements to the knowledge graph, and ultimately enhance large model generation performance
Tags & Categories
Answer 10 quick questions to check your fit for 混元多模态知识图谱算法研究员（北京） @ Tencent.
10 Questions
~2 Minutes
Instant Score
Related Books and Jobs