Stars
Qwen3.5 is the large language model series developed by Qwen team, Alibaba Cloud.
BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution
Training Language Model Agents to Find Vulnerabilities with CTF-Dojo
Cyber-Zero: Training Cybersecurity Agents Without Runtime
X-Repo2Run: Configuraing Multilingual Docker Environment via Code Agent
Making code edting up to 7.7x faster using multi-layer speculation
CTF Archives: Collection of CTF Challenges.
Small, simple agent task environments for training and evaluation
[ICLR 2025] 🧬 RegMix: Data Mixture as Regression for Language Model Pre-training (Spotlight)
[ICLR'25] BigCodeBench: Benchmarking Code Generation Towards AGI
XFT: Unlocking the Power of Code Instruction Tuning by Simply Merging Upcycled Mixture-of-Experts
Astraios: Parameter-Efficient Instruction Tuning Code Language Models
A framework for the evaluation of autoregressive code generation language models.
DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of Machine-Generated Text
🐙 OctoPack: Instruction Tuning Code Large Language Models
Source Code Data Augmentation for Deep Learning: A Survey.
Home of StarCoder: fine-tuning & inference!
[EACL 2024] ICE-Score: Instructing Large Language Models to Evaluate Code
💩State-of-the-art shitcode principles your project should follow to call it a proper shitcode
PyArmadillo: an alternative approach to linear algebra in Python
A corpus and code for understanding norms and subjectivity. 🤖
Stacked hierarchical attention for text-based games
🦾 A list of reported app support for Apple Silicon as well as Apple M4 and M3 Ultra Macs