Tianfu Wang | Human-Centered Agentic AI

I am a 2nd-year Ph.D. student at the AI Trust of HKUST(GZ) supervised by Prof. Hui Xiong and Dr. Nicholas Jing Yuan. Previously, I received my M.S. degree from USTC in 2025 under the supervision of Prof. Hui Xiong, and my B.E. degree from CQU in 2022. I am currently a research intern at Z.ai, with previous experience at Kimi, Tencent Codebuddy/ Workbuddy, Microsoft AI/ MSRA and JD Explore Academy.

My research focuses on Human-centered Agentic AI, particularly in:

Foundations (Model & Harness): MindOS, Agentic Consensus, Kimi K3 (contributor)
Applications (Education & Design): GenMentor, SocialCoach, EvoDesign, Agentic Artifact Creation Survey

Previously, I worked on reinforcement learning for network resource allocation and data mining.

💡 I am passionate about practical research that explores innovative applications and production-ready solutions.

🌱 I seek to answer: How can AI enrich people’s experiences while empowering them to thrive?

💌 Please feel free to contact me if you would like to explore potential discussions or collaborations.

Email | Github | Google Scholar | DBLP

What’s New

2026.09: Four papers are accepted to NeurIPS 2026!
- EvoDesign: Agentic Editable Diagram Creation via Design Expertise Evolution (first author)
- Population-Aligned Persona Generation for LLM-based Social Simulation (co-author)
- Machine Unlearning in Diffusion LLMs (co-author)
- Two Drifts, One Principle: Conflict-Aware Spectral Consolidation for Multimodal Continual Learning (co-author)
2026.08: Our new survey on Agentic Artifact Creation is available!
- Agentic Artifact Creation: Systems, Evaluation, Principles, and Opportunities (first author)
2026.08: One paper is accepted to EMNLP-2026 Findings!
- Unsupervised Post-Training of Foundation Models: A Survey (co-author)
2026.07: The Kimi K3 Tech Report is available!
- Contributed to visual coding benchmarks & data, such as 3D, game, and SVG tasks.
2026.04: One paper is accepted by ICML-2026!
- A Computational Framework for Evaluating Human-likeness in LLMs’ Open-ended Human Behaviors (co-author)

2026.04: One paper is accepted by ACL-2026!
- Anti-Length Shift Dynamic Outlier Truncation for Training Efficient Reasoning Models (co-author)
2026.02: Two papers are accepted by CVPR-2026 Findings!
- BVE Beyond Voxel 3D Editing: Learning from 3D Masks and Self-Constructed Data (co-author)
- AlignFL Adaptive Learning and Intelligent Generation of Networks for Federated Learning (co-author)
2026.01: One paper is accepted by ICLR-2026!
- Virne A Comprehensive Simulator and Benchmark for RL-based NFV Resource Allocation (first author)

2025.11: One paper is accepted by KDD-2026!
- HumanLLM Personalized Understanding and Simulation of Human Nature (co-author)
2025.10: One paper is accepted by ICDE-2026!
- TopFGL Topology-Aware and Distribution-Agnostic Federated Graph Learning Framework (co-author)
2025.09: One paper is accepted by NeurIPS-2025!
- A Cognitive Framework for Unveiling the Learning Mind of Language Models(co-author)
2025.08: Two papers are accepted by EMNLP-2025 (One Main and One Findings)!
- TokenSelect Token-Level KV Cache Selection for LLM Long-Context Inference (co-author)
- Explaining Length Bias in LLM-Based Preference Evaluations (co-author)
2025.08: One paper is accepted by JSA!
- RL-based Computation Task Scheduling in Autonomous Multi-Robot Systems (co-author)
2025.08: One paper is accepted by TGCN!
- IKENGA Infeasibility Knowledge and RL-Enhanced Genetic Algorithm for Network Resource Allocation (co-author)
2025.07: One paper is accepted by ICML-2025 Workshop on ML4Wireless with Best Paper Award!
- CONAL Constraint-aware Learning for Robust Network Resource Allocation (first author, Award)
2025.04: One paper is accepted by IJCAI-2025!
- CodeAgent LLM-based Student Behavior Simulation for Programming Learning (co-author)
2025.01: One paper is accepted by WWW-2025 as Oral Representation!
- GenMentor Agentic LLM Framework for Goal-oriented Tutoring System (first-author, Oral)
2024.11: One paper is accepted by ICDE-2025!
- Similar Subtrajectory Search Considering Constraints and Simplification (co-author)
2024.11: One paper is accepted by VLDB-2025!
- Data-Aware Distance Estimation for Nearest Neighbor Search (co-author)
2024.10: One paper is accepted by VLDB-2025!
- MILLION Direct Optimization for Risk-controllable Portfolio Management (co-first-author)
2024.05: One paper is accepted by KDD-2024!
- COMET Wallet Profiling with Temporal GNN for NFT Price Prediction (first-author)
2024.04: Two papers are accepted by IJCAI-2024!
- FlagVNE Meta RL for Generalizable Network Resource Allocation (first-author)
- Over-smoothing of GNN in Recommendation System (co-author)
2023.12: One paper is accepted by 电子学报!
- Imitation Learning for Network Resource Allocation (co-author)
2023.10: One paper is accepted by TSC!
- HRL-ACRA Hierarchal RL for Online Network Resource Allocation (first-author)
2023.07: One paper is accepted by MM-2023!
- RL-based Diffusion Model for Profit-oriented NFT Generation (co-author)

Selected Projects

View all

MindOS

Knowledge infrastructure

Open source

A local-first knowledge base that keeps notes, decisions, and context connected across AI workflows.

SocialCoach

Agentic learning system

An agentic learning system for personalized social skill development through tutoring and practice.

Virne

Research benchmark

ICLR 2026

A simulator and benchmark for learning to allocate network resources with deep reinforcement learning.

Selected Publications

View all

Agent Foundations

MindOS: A Shared Mind for Human-agent Coevolution

Tianfu Wang, Nicholas Jing Yuan, and Hui Xiong

Website Code
Scaling Human-AI Coding Collaboration Requires a Governable Consensus Layer

Tianfu Wang, Zhezheng Hao, Yin Wu, Wei Wu, Qiang Lin, Hande Dong, Nicholas Jing Yuan, and Hui Xiong

Website PDF Code

TR

Kimi K3: Open Frontier Intelligence

Kimi Team (Tianfu Wang is a co-author)

Website PDF Code
Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems

Zhezheng Hao*, Tianfu Wang*, Huanshuo Dong, Ziyan Liu, Hong Wang, Xiankun Lin,

2 more authorsQiang Lin, Can Wang

, Hande Dong, and Jiawei Chen

PDF Code
KDD

HumanLLM: Towards Personalized Understanding and Simulation of Human Nature

Yuxuan Lei, Tianfu Wang, Jianxun Lian, Zhengyu Hu, Defu Lian, and Xing Xie

In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2026

PDF Code

Agentic Applications

Agentic Artifact Creation: Systems, Evaluation, Principles, and Opportunities

Tianfu Wang, Zhezheng Hao, Xilin Xia, Lixin Liu, Mengkang Hu, Hongzhang Liu,

4 more authorsXi Chen, Ziyan Liu, Xiankun Lin, Weijia Zhang

, Nicholas Jing Yuan, and Hui Xiong

Website PDF Code
NeurIPS

EvoDesign: Agentic Editable Diagram Creation via Design Expertise Evolution

Tianfu Wang, Leilei Ding, Ziyang Tao, Yi Zhan, Zhiyuan Ma, Wei Wu,

7 more authorsYuxuan Lei, Junyang Wang, Yizhao Xu, Yin Wu, Zhengyu Hu, Hongyuan Zhu, Nicholas Jing Yuan

, Yanyong Zhang, and Hui Xiong

In Advances in Neural Information Processing Systems (NeurIPS), 2026

Website PDF Code
SocialCoach: Personalized Social Skill Learning with Agentic Tutoring and Practice

Tianfu Wang, Max Xiong, Yuxuan Lei, Jianxun Lian, Hongyuan Zhu, Zhengyu Hu,

4 more authorsLinxiao Gong, Dapeng Hu, Xiaofang Li, Peiting Tsai

, Nicholas Jing Yuan, and Qi Zhang

Website Demo PDF Code
WWW

LLM-powered Multi-agent Framework for Goal-oriented Learning in Intelligent Tutoring System

Tianfu Wang, Yi Zhan, Jianxun Lian, Zhengyu Hu, Nicholas Jing Yuan, Qi Zhang, Xing Xie, and Hui Xiong

In ACM Web Conference (WWW), 2025

Oral Presentation

Website Demo PDF Code Poster Slides

Emerging Tech

VLDB

MILLION: A General Multi-Objective Framework with Controllable Risk for Portfolio Management

Liwei Deng*, Tianfu Wang*, Yan Zhao, and Kai Zheng

In International Conference on Very Large Data Bases (VLDB), 2025

PDF Code
KDD

COMET: NFT Price Prediction with Wallet Profiling

Tianfu Wang, Liwei Deng, Chao Wang, Jianxun Lian, Yue Yan, Nicholas Jing Yuan, Qi Zhang, and Hui Xiong

In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2024

PDF Code Poster Slides

RL for Networking

ICLR

Virne: A Comprehensive Benchmark for RL-based Network Resource Allocation in NFV

Tianfu Wang, Liwei Deng, Xi Chen, Junyang Wang, Huiguo He, Zhengyu Hu,

2 more authorsWei Wu, Leilei Ding

, Qilin Fan, and Hui Xiong

In International Conference on Learning Representations (ICLR), 2026

Website PDF Code
ICMLW

Towards Constraint-aware Learning for Resource Allocation in NFV Networks

Tianfu Wang, Long Yang, Chao Wang, Chuan Qin, Liwei Deng, Li Shen, and Hui Xiong

In ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless), 2025

Best Paper Award

PDF Code
IJCAI

FlagVNE: A Flexible and Generalizable Reinforcement Learning Framework for Network Resource Allocation

Tianfu Wang, Qilin Fan, Chao Wang, Long Yang, Leilei Ding, Nicholas Jing Yuan, and Hui Xiong

In International Joint Conference on Artificial Intelligence (IJCAI), 2024

PDF Code Poster Slides
TSC

Joint Admission Control and Resource Allocation of Virtual Network Embedding via Hierarchical Deep Reinforcement Learning

Tianfu Wang, Shen Li, Qilin Fan, Tong Xu, Tongliang Liu, and Hui Xiong

IEEE Transaction on Services Computing (TSC), 2023

PDF Code
ICC

DRL-SFCP: Adaptive Service Function Chains Placement with Deep Reinforcement Learning

Tianfu Wang, Qilin Fan, Xiuhua Li, Xu Zhang, Qingyu Xiong, Shu Fu, and Min Gao

In IEEE International Conference on Communications (ICC), 2021

PDF Code