#ai4code search results
New models, same struggle! DeepSeek-R1, GPT-4.5 Preview, and Claude 3.7 Sonnet still show low performance on realistic coding tasks (RepoCod-Lite), with Pass@1 of 4.5%, 4.0%, and 3.5%. Top models just reach 10% Pass@1. We need better approaches for code generation! #AI4Code #LLM
Finally, check out the appendix! There's a few cool little bonus results in there that we couldn't fit in the main paper. This project was a labor of love and we really tried to be as comprehensive as possible - if you work in #ai4code, I hope you enjoy this knowledge dump!
Slow Friday in the SF Bay Area? Join us at the berkeleyhaasaisummit.com 👇 I’ll give a talk on “AI meets Source Code: status quo and outlooks”! #AI4Code #LLM
Microsoft Research India is hiring for full-time positions in Bengaluru! Please reach out if you are working in #AI4Code. microsoft.com/en-us/research…
Very excited to be hearing Michele Catasta (@pirroh) talk about #AI4Code at the Synapse #AI Symposium. @bendingspoons
After 36 hours of travel, finally at Porto de Galinhas, Brazil to attend @AIwareConf @FSEconf . I'll be here throughout the conference and would love to discuss #AI4Code & #AI4SE in general. Also, share our learnings from building GitHub Copilot inside VS. DMs are open.
The #LLMs for Code Seminar is launching this week! 🧠 Join us online: bit.ly/LLMs-for-Code-… #AI4Code #LLMs4Code
Introducing SemCoder, a semantic-aware Code LLM excelling in code generation and execution reasoning. Trained with high-quality data and novel way of aligning execution, only 6.7B model is outperforming GPT3.5 and CodeLlama 34B. link: arxiv.org/pdf/2406.01006 #LLMs, #AI4Code
Please review the flyer for the job description and requirements, and apply online if you feel qualified and interested. Please reach out to me directly (after applying online) only if you have an established track record in the research area of #AI4Code.
My summary on progresses in the area of AI4Code - applying AI/ML techniques to source code centric usecases - appears in the March 2021 issue of @acm_sigai's AI Matters magazine, out now. 🔗 Paper: sigai.acm.org/static/aimatte… Full issue: sigai.acm.org/aimatters/7-1.… #AI #ML #AI4Code
New Code Mentor launch! 🚀 Fetch and manipulate your data easily using natural language outsystems.com/product-update… #AI4Code
OctoPack: Instruction Tuning Code Large Language Models Presents HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). arxiv.org/abs/2308.07124
Tool Augmented Language Models. The #AI4Code reading group is also back in July with Aaron Parisi, #Google researcher and lead author of the #TALM paper👇 arxiv.org/pdf/2205.12255…
#ai4code peeps at #ICLR2024: come chat self-repair with me at 10:45 AM on Friday, May 10! I'll be presenting the camera-ready version of this fan favorite from last year... now with even more data! Keep reading for a sneak peek of the results 👇
Demystifying GPT Self-Repair for Code Generation paper page: huggingface.co/papers/2306.09… Large Language Models (LLMs) have shown remarkable aptitude in code generation but still struggle on challenging programming tasks. Self-repair -- in which the model debugs and fixes mistakes…
🔑📝Toward Code-Aware AI Models for Code @msrconf by @baishakhir. She shared results of models that beat state-of-the-art models already with zero-shot. Promising results for good results with lower costs. 🙌🌱 #MSR2023 #AI4code
The #LLMs for Code Seminar is about to launch! bit.ly/LLMs-for-Code-… Our first speaker is Jacob Andreas from MIT (abstract below). Register for his talk at the link in the comments. @jacobandreas #AI4Code #AI4SE
Updating real-world large legacy projects like binutils? Meet (arxiv.org/abs/2501.14257) C2SaferRust: leveraging program analysis & LLMs to create idiomatic, safer Rust with (↓38%) raw pointers & (↓28%) unsafe code while preserving functionality 🚀 #rustlang #AI4code #AIAgent
The #LLMs for Code Seminar is launching this week! 🧠 Join us online: bit.ly/LLMs-for-Code-… #AI4Code #LLMs4Code
Finally, check out the appendix! There's a few cool little bonus results in there that we couldn't fit in the main paper. This project was a labor of love and we really tried to be as comprehensive as possible - if you work in #ai4code, I hope you enjoy this knowledge dump!
New models, same struggle! DeepSeek-R1, GPT-4.5 Preview, and Claude 3.7 Sonnet still show low performance on realistic coding tasks (RepoCod-Lite), with Pass@1 of 4.5%, 4.0%, and 3.5%. Top models just reach 10% Pass@1. We need better approaches for code generation! #AI4Code #LLM
OctoPack: Instruction Tuning Code Large Language Models Presents HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). arxiv.org/abs/2308.07124
After 36 hours of travel, finally at Porto de Galinhas, Brazil to attend @AIwareConf @FSEconf . I'll be here throughout the conference and would love to discuss #AI4Code & #AI4SE in general. Also, share our learnings from building GitHub Copilot inside VS. DMs are open.
Slow Friday in the SF Bay Area? Join us at the berkeleyhaasaisummit.com 👇 I’ll give a talk on “AI meets Source Code: status quo and outlooks”! #AI4Code #LLM
Yo! Check out my new "incomplete" repo for the #ai4code kaggle competition by @google and @theteamatx! "Understand the relationship between code and comments in Python notebooks" View the repo at: github.com/Pranshu-Bahadu… @kaggle Competition at: kaggle.com/competitions/A…
Very excited to be hearing Michele Catasta (@pirroh) talk about #AI4Code at the Synapse #AI Symposium. @bendingspoons
Please review the flyer for the job description and requirements, and apply online if you feel qualified and interested. Please reach out to me directly (after applying online) only if you have an established track record in the research area of #AI4Code.
🔑📝Toward Code-Aware AI Models for Code @msrconf by @baishakhir. She shared results of models that beat state-of-the-art models already with zero-shot. Promising results for good results with lower costs. 🙌🌱 #MSR2023 #AI4code
The #LLMs for Code Seminar is about to launch! bit.ly/LLMs-for-Code-… Our first speaker is Jacob Andreas from MIT (abstract below). Register for his talk at the link in the comments. @jacobandreas #AI4Code #AI4SE
Tool Augmented Language Models. The #AI4Code reading group is also back in July with Aaron Parisi, #Google researcher and lead author of the #TALM paper👇 arxiv.org/pdf/2205.12255…
OutSystems.AI researchers @miguelventura and J. Nadkarni go over the State-of-the-Art in #MLonCode #AI4Code #OutSystemsAI bit.ly/2DY5Car
My summary on progresses in the area of AI4Code - applying AI/ML techniques to source code centric usecases - appears in the March 2021 issue of @acm_sigai's AI Matters magazine, out now. 🔗 Paper: sigai.acm.org/static/aimatte… Full issue: sigai.acm.org/aimatters/7-1.… #AI #ML #AI4Code
New Code Mentor launch! 🚀 Fetch and manipulate your data easily using natural language outsystems.com/product-update… #AI4Code
#ai4code peeps at #ICLR2024: come chat self-repair with me at 10:45 AM on Friday, May 10! I'll be presenting the camera-ready version of this fan favorite from last year... now with even more data! Keep reading for a sneak peek of the results 👇
Demystifying GPT Self-Repair for Code Generation paper page: huggingface.co/papers/2306.09… Large Language Models (LLMs) have shown remarkable aptitude in code generation but still struggle on challenging programming tasks. Self-repair -- in which the model debugs and fixes mistakes…
(6/6) In summary, INDICT's cooperative critic system demonstrates potential for responsible AI for code, achieving state-of-the-art security without compromising functionality. Explore the future of safe and helpful AI-generated code today! #AI #Innovation #AI4Code
Coding time! @NeurIPSConf from @IBMResearch and @RedHat on coding assistance for data science, recommendation of software stacks for AI applications, app modernization, and vulnerability analysis. Starting now! #NeurIPS2020 #AI4Code 👩💻👨💻
A great way to kick off #NeurIPS2020 - check out the @IBMResearch EXPO talk on the AI4Code efforts at @IBM and @RedHat - 🔗 bit.ly/3ghRKIv Sun, Dec 6 2pm EST / 11am PST 👆 We will be around for questions after the 45min talk. See you there! #neurips #AI #ML #nlproc
How do you turn a text-to-code model into a smarter AI developer? It starts with supervised fine-tuning, curated code examples, and human feedback. Read how it’s done: hubs.ly/Q03sq_xG0 #GenAI #AI4Code #LLMTraining #DataAnnotation
Something went wrong.
Something went wrong.
United States Trends
- 1. Good Wednesday 22.1K posts
- 2. Hump Day 9,682 posts
- 3. #MAYATVAWARDS2025 616K posts
- 4. Huda 13.7K posts
- 5. #NationalCatDay N/A
- 6. South Korea 97.8K posts
- 7. #hazbinhotelseason2 14.4K posts
- 8. #PutThatInYourPipe N/A
- 9. Clippers 14.9K posts
- 10. Tapper 28.8K posts
- 11. Namjoon 204K posts
- 12. Jay Z 6,852 posts
- 13. Harden 11K posts
- 14. Olandria 28.8K posts
- 15. Eminem 12.4K posts
- 16. Nvidia 74.6K posts
- 17. Kawhi 3,886 posts
- 18. #RHOSLC 8,544 posts
- 19. Caden 9,284 posts
- 20. Western Union 12.9K posts