Boyang Yang

Yanshan University, researcher in AI4SE.

[Software Engineering Papers @ComputerPapers]

HELO-APR: Enhancing Low-Resource Program Repair through Cross-Lingual Knowledge Transfer

Zhipeng Wang, Boyang Yang, Yidong Wan, Liuye Guo, You Lv, Tao Zheng, Zhuowei Wang, Tieke He
https://t.co/npvrrrtttG [𝚌𝚜.𝚂𝙴]

[Boyang Yang @buaabarty]

Codex mobile is great, but it’s only for the macOS host. Our CodexDeck is for CLI: Linux, macOS, and Windows/WSL. Browser-first UI for local Codex sessions: attach, resume, watch logs, inspect commands, and control from your phone and other laptops. https://t.co/b130VLp1ph

[Software Engineering Papers @ComputerPapers]

Beyond Localization: Recoverable Headroom and Residual Frontier in Repository-Level RAG-APR

Pengtao Zhao, Boyang Yang, Bach Le, Feng Liu, Haoye Tian
https://t.co/XplFKn0Xf1 [𝚌𝚜.𝚂𝙴]

[Software Engineering Papers @ComputerPapers]

PAFT: Preservation Aware Fine-Tuning for Minimal-Edit Program Repair

Boyang Yang, Zijian Cai, Shunfu Jin, Haoye Tain
https://t.co/pHJJAvTMmP [𝚌𝚜.𝚂𝙴]

[MSR4PS Workshop @MSR4PSworkshop]

⏰ Deadlines approaching! We are looking forward to your submissions!
The 4th edition of MSR4P&S will once more be co-located with @SANERconf 2026 in Limassol, Cyprus 🇨🇾
🌎 More info in our website: https://t.co/zYHZ2doNB6 #softwaresecurity #ai #genai #cfp

[Boyang Yang @buaabarty]

Here is the collection on Huggingface, which includes the fine-tuned CodeLlama, and all 4 evaluation benchmarks (EvalRepair-C++, EvalRepair-Java, D4J-Repair, SWE-Repair): https://t.co/H1LhIUPYit

[Boyang Yang @buaabarty]

Our work "MORepair: Teaching LLMs to Repair Code via Multi-Objective Fine-tuning" has been accepted by TOSEM, a multi-objective fine-tuning framework that teaches LLMs to understand "why" behind fixes, not just "what". All benchmarks are available on hf. https://t.co/eZv50kQrfe

[Boyang Yang @buaabarty]

Does anyone have an invitation code for https://t.co/fDe71lZ27M? HRYSPA [/doge]

[Boyang Yang @buaabarty]

We've presented our paper at ISSTA-24! Every can obtain our C++ program repair benchmark TutorCode through: https://t.co/V7g9nYsH4y
#issta #programrepair

[Boyang Yang @buaabarty]

@NVIDIAGeForce #BlackMythRTX 国产游戏之光

[Boyang Yang @buaabarty]

🎉Our paper "CREF: An LLM-based Conversational Software Repair Framework for Programming Tutors" is accepted by ISSTA 2024! We introduce a large-scale repair benchmark TutorCode, and a conversational repair framework. @HaoyeTIAN @JacquesKlein14 @BissyandeF https://t.co/3sNUZOzwsZ

[Software Engineering Papers @ComputerPapers]

CREF: An LLM-based Conversational Software Repair Framework for Programming Tutors. https://t.co/eUdb2M76Ra

[Arena.ai @arena]

Exciting Arena Leaderboard Updates!

Six new models:
- Tulu-2-DPO-70B and Yi-34B-Chat are the new SoTA open models
- Mistral-based 7B models (OpenChat, OpenHermes-2.5, Starling-7B) are stronger than ever

Big congrats to the OSS AI community!

Learn more https://t.co/4Ah31B0Fr5

[Boyang Yang @buaabarty]

@greeny1025 这家不会是要跑路吧……

[Boyang Yang @buaabarty]

@greeny1025 应该恢复了，更新一下托管就好了

[Boyang Yang @buaabarty]

https://t.co/I4m6BDM0lP
baichuan-13b 发布了，大家觉得效果如何？

[Boyang Yang @buaabarty]

https://t.co/3BsS5KqJP3

[Harsh Makadia @MakadiaHarsh]

The Essential AI tools:

Image: https://t.co/YrYhyjopsm
Copywriting: https://t.co/OMG2Ue1lK5
Coding: https://t.co/bBQmkEvcWt
Twitter: https://t.co/DmWEXuJybp
Search Engine: https://t.co/8aimJeD16b
Whatsapp: https://t.co/o9K0nsKxrB
Product visuals: https://t.co/FkySI7HnsQ

[Andrew Ng @AndrewYNg]

Lou Tiancheng--one of the world's best hackers--just joined us to work on autonomous cars!

[Boyang Yang @buaabarty]

I just registered for Google Code Jam! See https://t.co/iOfm51trW5 for more information #CodeJam2016