Fragile Reasoning: A Mechanistic Analysis of LLM Sensitivity to Meaning-Preserving Perturbations
April 4, 2026admin NUAI,Committee,Nachrichten,Uncategorized(0)arXiv:2604.01639v1 Announce Type: new Abstract: Large language models demonstrate strong performance on mathematical reasoning benchmarks...
Foxconn builds AI factory in partnership with Taiwan and Nvidia
Mai 19, 2025admin NUAI,Committee,Nachrichten,Uncategorized(0)Nvidia and Foxconn announced they are working with the Taiwan government to build an AI...

FourierMoE: Fourier Mixture-of-Experts Adaptation of Large Language Models
April 4, 2026admin NUAI,Committee,Nachrichten,Uncategorized(0)arXiv:2604.01762v1 Announce Type: cross Abstract: Parameter-efficient fine-tuning (PEFT) has emerged as a crucial paradigm for...
Four things we’d need to put data centers in space
April 5, 2026admin NUAI,Committee,Nachrichten,Uncategorized(0)MIT Technology Review Explains: Let our writers untangle the complex, messy world of technology to...
Four reasons to be optimistic about AI’s energy usage
Mai 20, 2025admin NUAI,Committee,Nachrichten,Uncategorized(0)The day after his inauguration in January, President Donald Trump announced Stargate, a $500 billion...

Four nuclear reactors hit a big milestone in the US
Juli 9, 2026admin NUAI,Committee,Nachrichten,Uncategorized(0)I was really looking forward to July 4, and not just because I love a...
Fostering breakthrough AI innovation through customer-back engineering
Mai 11, 2026admin NUAI,Committee,Nachrichten,Uncategorized(0)Despite years of digitization, organizations capture less than one-third of the value expected from digital...

Format-Adapter: Improving Reasoning Capability of LLMs by Adapting Suitable Format
Juli 1, 2025admin NUAI,Committee,Nachrichten,Uncategorized(0)arXiv:2506.23133v1 Announce Type: new Abstract: Generating and voting multiple answers is an effective method to...
Forget the hype — real AI agents solve bounded problems, not open-world fantasies
Juli 7, 2025admin NUAI,Committee,Nachrichten,Uncategorized(0)Event-driven multi-agent systems are a practical architecture for working with imperfect tools in a structured...
