Xi Zhang @ EMNLP 2025
@_xizhang
Followers
90
Following
619
Media
17
Statuses
89
CS PhD @UofGlasgow & https://t.co/XsGsWe11Gz | Alum @Sydney_Uni | Interests: AI Agents, Web3 | Formulator of the Three Laws of Self-Evolving AI Agents
Glasgow
Joined May 2022
It’s COMING! Stay tuned. 🤖 https://t.co/dAgHdvwnmN
x-izhang.github.io
A New Paradigm for AI Agents
0
0
4
Looking forward to meeting everyone in Suzhou! 🦀
Heading to #EMNLP2025 next Tuesday — Suzhou, here we come! 🌟 Excited to share that the AI4BioMed Lab (led by @mengzaiqiao& @jakelever0) has four papers accepted this year!🎉 Come chat with us!👋 @JinyuanF, @_xizhang, @Zhaohan_Meng, @XinhaoYi, @TedSiwei and @mengzaiqiao
0
0
2
HowOne AI Alpha Test Begins NOW! You don’t need a team to build your AI startup anymore. With HowOne AI, you can turn your idea into a fully functional AI app—in minutes. Build → Launch → Earn—all in one platform. 🎥 Here’s how it works 👇 #Solopreneur #AIapps #HowOneAI
8
8
17
✨ This work was led by me and advised by @mengzaiqiao, @jakelever0, and @EdmondSLHo. 📨 Please reach out if you want to chat more!
0
1
2
Details & Resources 📄 Paper: https://t.co/HK0yJ8TXW4 🌐 Project: https://t.co/TiDLiMBWeZ 🎮 Demo: https://t.co/YqxizxhFK9 🤗 Dataset:
huggingface.co
1
0
1
💡 Our solution: We introduce CCD, a plug-and-play inference framework that reduces medical hallucinations in Radiology MLLMs - no retraining, no retrieval. By injecting structured clinical supervision from expert models (e.g., DenseNet, MedSigLIP) during decoding, CCD
1
0
1
🚨 The gap: Current Radiology MLLMs have already gained strong general capabilities through large-scale instruction tuning. Further fine-tuning is often used to boost performance on specific benchmarks (e.g., RRG, VQA). However, subtle medical hallucinations still occur - often
1
0
1
📸 New Preprint Out! If you already have a SOTA Radiology MLLM, why bother training it again? Especially when new data are scarce and GPUs are even scarcer - that's a frustrating dilemma. 🚀 We introduce a training-free and retrieval-free inference framework that boosts
1
1
3
Our survey paper’s GitHub repo just hit ⭐️1K stars! 🚀🎉 ---> https://t.co/6xZchrT0EY
Self-Evolving AI Agents are about to blow up 🚀🔥 New Survey Alert 🚨 From static LLMs ➡️ lifelong, adaptive agents 🌱🤖 A Comprehensive Survey of Self-Evolving AI Agents... — your roadmap to the future of AI that learns & evolves after deployment! 💡 https://t.co/TBYfLLdaD1
0
3
9
🎉 Truly honoured to collaborate with such an outstanding team. 🚀 RadEval marks a significant step forward in Radiology AI. 🤗 Many thanks to @IAMJBDEL for the invitation!
💥 Today we unveil RadEval, an unified framework for evaluating AI-generated radiology text. RadEval will be presented as an Oral at #EMNLP25 Joint effort by HOPPR, University of Oxford, University of Glasgow RadEval integrates 11+ state-of-the-art metrics, ranging from lexical
0
0
1
🚀 Excited to share that EvoAgentX, the world’s 1st open-source self-evolving framework for AI Agents, has been accepted at EMNLP 2025 (Demo Track)! It now received ⭐ 1.4k stars on GitHub: https://t.co/XrvxxjVwp4 🎉 We’ll be presenting it in Suzhou. @EvoAgentX
github.com
🚀 EvoAgentX: Building a Self-Evolving Ecosystem of AI Agents - EvoAgentX/EvoAgentX
0
6
15
Thanks for spreading the word! @omarsar0
Overview of Self-Evolving Agents There is a huge interest in moving from hand-crafted agentic systems to lifelong, adaptive agentic ecosystems. What's the progress, and where are things headed? Let's find out:
0
0
9
Huge thanks to everyone—our survey paper has surpassed 600⭐ in just 20 days! We’re incredibly proud to contribute to the rapidly growing field of self‑evolving AI agents. The repo and paper are actively maintained, and we’d love to hear your suggestions for related works.
3
2
4
Thanks @rohanpaul_ai for sharing!
Absolutely Golden resource: A Comprehensive Survey of Self-Evolving AI Agents Self‑evolving agents are built to adapt themselves safely, not just run fixed scripts, guided by 3 laws, endure, excel, evolve. The survey maps a 4‑stage shift, MOP (Model Offline Pretraining) to
0
0
5
A Comprehensive Survey of Self-Evolving AI Agents A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
5
77
250
Self-Evolving AI Agents are about to blow up 🚀🔥 New Survey Alert 🚨 From static LLMs ➡️ lifelong, adaptive agents 🌱🤖 A Comprehensive Survey of Self-Evolving AI Agents... — your roadmap to the future of AI that learns & evolves after deployment! 💡 https://t.co/TBYfLLdaD1
2
14
36
