He Ye

Assistant professor at @ucl, co-founder of @Euni_AI, working on code agents.

[He Ye @ye_he_ye]

Excited to introduce our new work ContextBench, designed to evaluate coding agents’ ability to retrieve the right context from large codebases during SE task resolution process!!

[EuniAI @Euni_AI]

We just released ContextBench 🎉
A benchmark built to answer a question many repo-level evaluations still miss:
Do coding agents truly retrieve and use the right context, or do they just get lucky?👀✨

📊 Highlights
🧩 1,136 real-world issues across 66 repos and 8 languages
🧠 Expert verified gold contexts at file, block, and line levels
👣 Full trajectory tracking of what the agent actually reads and explores
📈 Metrics covering Recall, Precision, F1, Efficiency, and Usage Drop

🔍 Key Findings
1️⃣ Complex agentic scaffolds do not improve context retrieval quality 😅
In many cases, they introduce over-engineering, echoing "The Bitter Lesson" in AI research
2️⃣ Many SOTA LLMs favor high recall over precision 📉
They retrieve more context, but also much more noise
3️⃣ Retrieved does not mean utilized ❗
Agents often inspect the right code but fail to incorporate it into the final patch
4️⃣ Retrieval strategies that are more balanced tend to achieve stronger Pass@1 while keeping compute cost reasonable ⚖️✨

🌐 Homepage 👉 https://t.co/4upNH9XDYi
📄 Paper 👉 https://t.co/xXPziAFins
💻 Code 👉 https://t.co/belNHEFfyi
🗂️ Dataset 👉 https://t.co/XaljqOdg4r

[He Ye @ye_he_ye]

@Chasen86341870 @NUSComputing Big congratulations to Chengpeng🥳🥳🥳

[He Ye @ye_he_ye]

Join Chengpeng's group at NUS to do awesome research🥳🥳🥳

[Chengpeng Wang @Chasen86341870]

Excited to share that I will join NUS (@NUSComputing) as an Assistant Professor this Aug! 🎉

I’m recruiting Ph.D./RAs/interns interested in Program Analysis, Code LLMs, and Agents. 🔥

Self-motivated students with strong backgrounds are especially welcome.

[Chengpeng Wang @Chasen86341870]

Excited to share that I will join NUS (@NUSComputing) as an Assistant Professor this Aug! 🎉

I’m recruiting Ph.D./RAs/interns interested in Program Analysis, Code LLMs, and Agents. 🔥

Self-motivated students with strong backgrounds are especially welcome.

[Zhenpeng Chen @ZhenpengChen18]

Excited to share that I’ve recently joined the School of Software at Tsinghua University as an Assistant Professor. Looking forward to the journey ahead! 💻🚀

[He Ye @ye_he_ye]

@ZhenpengChen18 Huge congratulations🥳🥳🥳

[Delysium - $AGI 🟨 @The_Delysium]

Our AI coding model Prometheus, co-developed with Professor He Ye @ye_he_ye, @ucl SSE Team and @Euni_AI, has surpassed @Google's newly released Gemini 3 Pro Preview @geminiapp and achieved Top 5 globally among independent models on SWE-Bench.

- Prometheus GitHub: https://t.co/NqPgz5cKBg
- SWE-Bench results: https://t.co/8SNgZ78PUK (select "Open Scaffold + All Tags")

Stay tuned.

[EuniAI @Euni_AI]

We’re excited to announce our partnership with @The_Delysium — together, we’re setting open, autonomous, multilingual, and cost-efficient standards for AI coding.

By bridging AI systems engineering with Web3 best practices — @LucyOSAI and the YKILY Network — we aim to make AI development more transparent, verifiable, and accessible for everyone.

[Yiling Lou @yiling__LOU]

Thrilled to announce that I'll be joining UIUC CS @siebelschool as an Assistant Professor in Spring 2026!

📢 I’m looking for Fall '26 PhD students who are interested in the intersection of Software Engineering and AI, especially in LLM4Code and Code Agents. Please drop me an email if you are interested in working with me.

[He Ye @ye_he_ye]

@yiling__LOU @siebelschool Congratulations🥳🥳🥳

[Zhaoyang Chu @zhaoyang_c68411]

Great vibes at #GraphSummit London 🎉
Learned a ton + made great memories by Big Ben ✨

@Euni_AI, we share a vision with the graph community:
👉 LLMs need a better way to understand codebases — knowledge graphs stand between them.

Excited for what’s ahead! 🚀

#AI #Neo4j

[EuniAI @Euni_AI]

🚀 New milestone: Hermes hits 50% resolve rate on Terminal-Bench!

📊 Performance:
✅ 91.67% easy
⚡ 54.55% medium
🔥 20.83% hard

Now ranked #6 on the leaderboard, Hermes proves the power of iterative AI agents in complex terminal environments.

#LLM #Agents #AIResearch

[He Ye @ye_he_ye]

Attending OOPSLA/ICFP 2025 in Singapore? 🌏✈️

JOIN us on October 15th, we warmly welcome you to join the Workshop on Language Models for Programming Languages (LMPL) co-hosted at SPLASH 2025 !
👉https://t.co/yTUWj1AglA

🔥This year we are excited to have accepted 22 papers! 🥳

[EuniAI @Euni_AI]

🚀 New milestone!
Prometheus + GPT-5 → 70% resolve rate on SWE-bench Verified.

State-of-the-art performance for autonomous code agents.
The future of self-managing software is here.

#LLM #Agents #SoftwareEngineering #AutomatedProgramRepair #AIResearch #OpenSource #AgenticAI

[Sergey Mechtaev @sergey_mechtaev]

Prompting is the new programming, but NL suffers from ambiguity. SpecFix (ASE'25) is the first to automatically repair ambiguous programming problem descriptions, resulting in ↑ Pass@k, and its repairs generalizing across models.

https://t.co/ZPgBIqBr4a
https://t.co/1dqNqD1WRj

[EuniAI @Euni_AI]

🚀 Benchmark update 🚀

Prometheus + DeepSeek-Chat reached 35.33% resolve rate on SWE-bench Lite — achieving state-of-the-art performance in autonomous code agents.

This is just the beginning 👀

[EuniAI @Euni_AI]

🧠 Athena Memory System Launched! 🧠

An innovative memory system purpose-built for autonomous code agents:
⚡️ Long-term learning & memory
⚡️ Deeper context retention
⚡️ Greater adaptability

Athena marks the next leap toward truly self-managing, intelligent code agents.

[EuniAI @Euni_AI]

🧠 Prometheus keeps evolving.
With new integrations & features, we’re pushing the boundaries of what code agents can do:
•Context-rich problem solving
•Flexible workflows
•Real-world performance breakthroughs

[EuniAI @Euni_AI]

🚀 Prometheus Major Update 🚀

Big leap for autonomous code agents:
🔗 MCP Integration for stronger context & multimodal interaction
🔧 Web Search + Custom Build Commands
📊 44% solve rate on SWE-bench Verified

The future of self-managing code starts now.

[EuniAI @Euni_AI]

🚀 Introducing EuniAI 🚀

We’re building autonomous code agents to:
⚡ Resolve issues automatically
⚡ Understand codebases seamlessly
⚡ Cut bug costs & boost productivity

👀 Demo invites coming soon!
🌐 https://t.co/TU5K543sjI
💼 Follow us on LinkedIn for deeper updates.