Meta AI
Open-Source LLM Entwickler (Llama)
Nougat
docprocessingVisual Transformer OCR (Meta) fuer wissenschaftliche Dokumente (Mathe, Formeln).
WebsiteMeta Llama 4
modelsLlama 4: Scout (10M Kontext!), Maverick, Behemoth. 200 Sprachen. Open Source.
WebsitearXiv:2610.00320v1 Announce Type: new Abstract: Fine-tuning adapts aligned large language models (LLMs) to downstream t...
arXiv:2610.00309v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in privacy-cri...
arXiv:2610.00025v1 Announce Type: new Abstract: Agent harnesses increasingly want to run small language models (SLMs) o...
arXiv:2610.00047v1 Announce Type: new Abstract: Diversity collapse in parallel chain-of-thought has motivated inference...
arXiv:2610.00054v1 Announce Type: new Abstract: Reading an LLM judge's verdict from the logits of its first generated t...
arXiv:2610.00833v1 Announce Type: new Abstract: Fluent LLM explanations may not follow the evidence from a structured s...
arXiv:2610.00910v1 Announce Type: new Abstract: Human reasoning depends on how objects are related within propositions....
arXiv:2610.01230v1 Announce Type: new Abstract: Efficient long-context inference is essential for large language models...
arXiv:2610.01493v1 Announce Type: new Abstract: Iterative fine-tuning on synthetic data causes \emph{model collapse}: o...
arXiv:2610.01763v1 Announce Type: new Abstract: Activation sparsity speeds up large language model (LLM) inference by s...