Zhiqi Huang

I currently work at AGI startup Moonshot AI (月之暗面).
My research interests are LLM & MLLM Scaling, Vision Reasoning, Efficient Deep Learning Methods, etc.

We are hiring! If you are passionate about developing next generation of AI, or if you're keen on crafting the next super app, or if you see yourself fitting into a role at Moonshot AI, don't hesitate to get in touch with me!

Intern hiring! We are hiring multimodal interns with strong engineering and research experience to work on vision reasoning and post training! Please send your CV to "huangzhiqi AT moonshot.cn" or apply through our official referral link.

[Email] [Google Scholar]

Experiences

• Multimodal Researcher @ Moonshot AI
  Feb. 2024 ~ Now
  
  
  We built K1.5 and K1, a multimodal reasoning model

• Scaling Researcher @ Moonshot AI
  Jul. 2023 ~ Feb. 2024
  
  
  We built Kimi, a long context AI assistant

• Senior Researcher @ Tencent
  Dec. 2021 ~ Jul. 2023
  
  
  Developed AI solutions for game optimization in LoL and PUBG

Publications

• Kimi-VL Technical Report
  Kimi Team

• Kimi k1.5: Scaling reinforcement learning with llms
  Kimi Team

• MoBA: Mixture of Block Attention for Long-Context LLMs
  Kimi Team

• CountLLM: Towards Generalizable Repetitive Action Counting via Large Language Model
  [CVPR'25] Z. Yao, X. Cheng, Z. Huang, L. Li

• Towards Zero-shot Cross-lingual SLU with Syntax-aware Multi-view Contrastive Learning
  [ICASSP'25] Y. Xie, Z. Xiong, T. Zhang, M. Cui, Y. Li, Z. Huang, Z. Zhu

• FD2Talk: Towards Generalized Talking Head Generation with Facial Decoupled Diffusion Model
  [ACMMM'24] Z. Yao, X. Cheng, Z. Huang

• Mixture of Bidirectional Adapter for Multi-modal Sarcasm Detection
  [ACMMM'24] Y. Xie, Z. Zhu, X. Chen, Z. Chen, Z. Huang

• Dual-oriented Disentangled Network with Counterfactual Intervention for Multimodal Intent Detection
  [EMNLP'24] Z. Chen, Z. Zhu, X. Zhuang, Z. Huang, Y. Zou

• Advancing End-to-End Spoken Language Understanding with the Power of Large Language Models
  [EMNLP'24] X. Cheng, Z. Zhu, Z. Chen, X. Zhuang, Z. Huang, Y. Zou

• Divide and Conquer: Scenario-aware Label Graph Interaction for Multi-intent Spoken Language Understanding
  [CIKM'24] Z. Zhu, X. Cheng, Z. Chen, Z. Wang, Z. Huang, Y. Zou

• Multivariate Cooperative Game for Image-Report Pairs: Hierarchical Semantic Alignment for Medical Report Generation
  [MICCAI'24] Z. Zhu, X. Cheng, Y. Zhang, Z. Chen, Q. Long, H. Li, Z. Huang, X. Wu, Y. Zheng

• Code-Switching Can be Better Aligners: Advancing Cross-Lingual SLU through Representation-Level and Prediction-Level Alignment
  [ACL'24] Z. Zhu, X. Cheng, Z. Chen, X. Zhuang, Z. Huang, Y. Zou

• Towards ASR-Robust Spoken Language Understanding via Mixture-of-Experts
  [ACL'24] X. Cheng, Z. Zhu, X. Zhuang, Z. Chen, Z. Huang, Y. Zou

• KC-Prompt: End-to-end Knowledge-Complementary Prompting for Rehearsal-free Continual Learning
  [ICASSP'24] Y. Li, Y. Liu, X. Cheng, Z. Zhu, H. Li, B. Yang, Z. Huang

• Towards Multimodal Sentiment Analysis via Two-Stage Bottleneck Filtering and Optimal Transport
  [COLING'24] Y. Xie, Z. Zhu, X. Lu, Z. Huang, H. Xiong

• Alignment before Awareness: Towards Visual Question Localized-Answering in Robotic Surgery via Optimal Transport and Answer Semantics
  [COLING'24] Z. Zhu, Y. Zhang, X. Cheng, Z. Huang, D, Xu, X. Wu, Y. Zheng

• Zero-Shot Natural Language Understanding via Large Language Models
  [COLING'24] Z. Zhu, X. Cheng, H. An, Z. Wang, D. Chen, Z. Huang

• Towards Multi-modal Sarcasm Detection via Disentangled Multi-grained Multi-modal Distilling
  [COLING'24] Z. Zhu, X. Cheng, G. Hu, Y. Li, Z. Huang, Y. Zou

• Knowledge-enhanced Prompt Tuning for Dialogue-based Relation Extraction with Trigger and Label Semantic
  [COLING'24] H. An, Z. Zhu, X. Cheng, Z. Huang, Y. Zou

• Syntax Matters: Towards Spoken Language Understanding via Syntax-Aware Attention
  [EMNLP'23 Finding] Y. Xie, Z. Zhu, X. Cheng, Z. Huang, D. Chen

• Enhancing Code-Switching for Cross-lingual SLU: A Unified View of Semantic and Grammatical Coherence
  [EMNLP'23] Z. Zhu, X. Cheng, Z. Huang, D. Chen, Y. Zou

• Towards Unified SLU Decoding via Label-aware Compact Linguistics Representations
  [ACL'23 Finding] Z. Zhu, X. Cheng, Z. Huang, D. Chen, Y. Zou

• Mix before Align: Towards Zero-shot Cross-lingual Sentiment Analysis via Soft-Mix and Multi-View Learning
  [INTERSPEECH'23] Z. Zhu, X. Cheng, D. Chen, Z. Huang, H. Li, Y. Zou

• Towards Joint Intent Detection and Slot Filling via Higher-order Attention
  [IJCAI'22 Oral] D. Chen, Z. Huang, X. Wu, S. Ge, Y. Zou

• A Multi-grained Contrastive Learning Framework for ASR-robust Language Understanding
  [EMNLP'23] Z. Huang, D. Chen, Z. Zhu, X. Cheng

• Federated Learning for Spoken Language Understanding
  [COLING'20 Oral] Z. Huang, F. Liu, Y. Zou
  

• Audio-Oriented Multimodal Machine Comprehension via Dynamic Inter- & Intra-modality Attention
  [AAAI'21] Z. Huang, F. Liu, X. Wu, S. Ge, H. Wang, W. Fan, Y. Zou

• Sentiment Injected Iteratively Co-Interactive Network for Spoken Language Understanding
  [ICASSP'21] Z. Huang, F. Liu, P. Zhou, Y. Zou
  

• GhostBERT: Generate More Features with Cheap Operations for BERT
  [ACL'21 Oral] Z. Huang, L. Hou, L. Shang, X. Chen, X. Jiang, Q. Liu
  

• DynaBERT: Dynamic BERT with Adaptive Width and Depth
  [NeurIPS'20 Spotlight] L. Hou, Z. Huang, L. Shang, X. Jiang, X. Chen, Q. Liu