About

I am a master student in Master of Intelligent Information Systems (MIIS) at Carnegie Mellon University, Language Technologies Institute, directed by Prof. Ameet Talwalkar. Before that, I received my bachelor's degree at Shanghai Jiao Tong University, majoring in Software Engineering, guided by Prof. Xiaodong Gu.

I am interested in both research and engineering. I previously interned at Ant Group (OceanBase) and StepFun AI, and co-founded Emagen AI. This summer, I will join Jump Trading as a research intern.

My research focuses on LLM Agents, spanning both application and evaluation.

LLM Agents Agent Application Agent Evaluation Multi-modal Models Trustworthy AI

Publications

* equal contribution  ·  † corresponding author

2026
A Survey of Self-Evolving Agents: On Path to Artificial Super Intelligence
Huan-ang Gao*, Jiayi Geng*, Wenyue Hua*, ..., Yixiong Fang, ..., Mengdi Wang†
TMLR 2026 arXiv · Code
Reasoning in Trees: Improving Retrieval-Augmented Generation for Multi-Hop Question Answering
Yuling Shi, Maolin Sun, Zijun Liu, Mo Yang, Yixiong Fang, Tianran Sun, Xiaodong Gu
WWW 2026 Companion arXiv
GameDevBench: Evaluating Agentic Capabilities Through Game Development
Wayne Chi, Yixiong Fang, Arnav Yayavaram, Siddharth Yayavaram, Seth Karten, Qiuhong Anna Wei, Runkun Chen, Alexander Wang, Valerie Chen, Ameet Talwalkar, Chris Donahue
arXiv · Code
Test vs Mutant: Adversarial LLM Agents for Robust Unit Test Generation
Pengyu Chang, Yixiong Fang, Silin Chen, Yuling Shi, Beijun Shen, Xiaodong Gu
arXiv
2025
From Uncertainty to Trust: Enhancing Reliability in Vision-Language Models with Uncertainty-Guided Dropout Decoding
Yixiong Fang*, Ziran Yang*, Zhaorun Chen, Zhuokai Zhao†, Jiawei Zhou†
NeurIPS 2025 arXiv · Code
LastingBench: Defend Benchmarks Against Knowledge Leakage
Yixiong Fang*, Tianran Sun*, Yuling Shi, Min Wang, Xiaodong Gu†
EMNLP 2025 Findings arXiv · Code
AttentionRAG: Attention-Guided Context Pruning in Retrieval-Augmented Generation
Yixiong Fang, Tianran Sun, Yuling Shi, Xiaodong Gu
ACL 2025 Workshop arXiv
Enhancing Speech Large Language Models for Deepfake Speech and Audio Detection through Feature Dropout and Dropin
Yupei Li*, Yixiong Fang*, Lucia Specia, Xiaodong Gu, Björn Schuller
2024
CodeCipher: Learning to Obfuscate Source Code Against LLMs
Yalan Lin, Chengcheng Wan, Yixiong Fang, Xiaodong Gu
arXiv
Efficient Computation of Eigenvalues in Diffusion Maps: A Multi-Strategy Approach
Yixiong Fang*, Weixi Yang
SPML 2024 DOI

Experience

Co-Founder Current
Emagen AI
Leading the development of Cagen, an AI-native user based organization platform.
SQL Intern Jul – Aug 2024
Ant Group · OceanBase
Optimized utf8 parsing with SIMD for up to 20x performance increase; added southeast Asia charset and external table features.
AI Research Intern Mar – Jun 2024
StepFun AI
Developed early-stage agent with tools and RAG system; built the OpenAPI platform.
Quantitative Research Intern Jan – Feb 2024
China Merchants Securities
Built quantitative trading strategies and performed market analysis using ML techniques.
Quantitative Research Intern Jul – Aug 2022
Changjiang Securities International
Developed algorithmic trading models and conducted financial data analysis for equity markets.

Teaching

11-785: Introduction to Deep Learning
Teaching Assistant · Spring 2026
Mthread GPU Open Course
Teaching Assistant · Spring 2025
AI1102: Intel's Cutting-Edge AI Algorithms and Practices
Teaching Assistant · Spring 2025
SJTU Xuesen Challenge Winter Camp (AI)
Teaching Assistant · Winter 2025