Optimizing Length Compression in Large Reasoning Models
9 月 12, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2506.14755v2 Announce Type: replace-cross Abstract: Large Reasoning Models (LRMs) have achieved remarkable success, yet they...
Optimizing Assembly Code with LLMs: Reinforcement Learning Outperforms Traditional Compilers
5 月 25, 2025admin NUAI,Committee,新闻,Uncategorized(0)LLMs have shown impressive capabilities across various programming tasks, yet their potential for program optimization...
Operationalizing AI for Scale and Sovereignty
5 月 1, 2026admin NUAI,Committee,新闻,Uncategorized(0)Companies are taking control of their own data to tailor AI for their needs. The...
OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles
11 月 12, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2503.17352v3 Announce Type: replace-cross Abstract: We introduce OpenVLThinker, one of the first open-source large vision-language...
OpenThoughts: A Scalable Supervised Fine-Tuning SFT Data Curation Pipeline for Reasoning Models
6 月 14, 2025admin NUAI,Committee,新闻,Uncategorized(0)The Growing Complexity of Reasoning Data Curation Recent reasoning models, such as DeepSeek-R1 and o3...
OpenSIR: Open-Ended Self-Improving Reasoner
1 月 1, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2511.00602v2 Announce Type: replace Abstract: Recent advances in large language model (LLM) reasoning through reinforcement...
OpenGloss: A Synthetic Encyclopedic Dictionary and Semantic Knowledge Graph
11 月 25, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2511.18622v1 Announce Type: new Abstract: We present OpenGloss, a synthetic encyclopedic dictionary and semantic knowledge...
OpenCUA’s open source computer-use agents rival proprietary models from OpenAI and Anthropic
8 月 23, 2025admin NUAI,Committee,新闻,Uncategorized(0)The open source framework provides the data and training recipe for building powerful computer-use agents...
OpenAI’s GPT-5 rollout is not going smoothly
8 月 9, 2025admin NUAI,Committee,新闻,Uncategorized(0)It also failed on a simple algebra arithmetic problem that elementary schoolers could probably nail...
