--- id: "/foundation-models-labs/advanced/page/21/" title: "Foundation models & labs for advanced learners · page 21" url: "https://skillissue.dev/foundation-models-labs/advanced/page/21/" --- # Foundation models & labs for advanced learners [Beginner](/foundation-models-labs/beginner/index.md) [Intermediate](/foundation-models-labs/intermediate/index.md) [Advanced](/foundation-models-labs/advanced/index.md) [Expert](/foundation-models-labs/expert/index.md) 684 resources · page 21 of 22 - ### [DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters](/lab-practices/other-labs/danish-foundation-models-papers/index.md) by [Peter Schneider-Kamp et al. (University of Southern Denmark)](https://www.sdu.dk) arXiv paper (2026) describing DFM Mimir v1, a 1B-parameter Danish open model. Advanced–Expert learning_resource paper Denmark en - ### [DeepSeek-V3 Technical Report](/lab-practices/other-labs/deepseek-v3-technical-report/index.md) by [DeepSeek-AI](https://www.deepseek.com/) arXiv paper (2024) DeepSeek-V3 Technical Report: MLA attention, DeepSeekMoE and FP8 training. Advanced learning_resource paper China en - ### [deepset (German’s Next Language Model)](/lab-practices/other-labs/deepset-german-s-next-language-model/index.md) by [Branden Chan et al. (deepset)](https://www.deepset.ai/) arXiv HTML version of deepset's German's Next Language Model paper (2020) on GBERT and GELECTRA. Advanced–Expert learning_resource paper EU-DE en - ### [Eval awareness in Claude Opus 4.6's BrowseComp performance](/lab-practices/other-labs/eval-awareness-in-claude-opus-4-6-s-browsecomp-performance/index.md) by [Anthropic](https://anthropic.com/) Anthropic engineering post on Claude Opus 4.6 showing evaluation awareness on the BrowseComp benchmark. Advanced learning_resource paper US en - ### [ÚFAL at MultiLexNorm 2021: Improving Multilingual Lexical Normalization by Fine-tuning ByT5](/lab-practices/other-labs/fal-multilexnorm-paper/index.md) by [David Samuel and Milan Straka (ÚFAL, Charles University)](https://ufal.mff.cuni.cz/) arXiv paper (2021) describing ÚFAL's system for multilingual lexical normalisation at MultiLexNorm 2021. Advanced–Expert learning_resource paper Czechia en - ### [GLM-4.5 ARC tech report](/lab-practices/other-labs/glm-4-5-arc-tech-report/index.md) by [GLM-4.5 Team (Zhipu AI)](https://www.zhipuai.cn/) arXiv paper (2025) GLM-4.5 technical report on Z.ai's agentic, reasoning and coding foundation models. Advanced learning_resource paper China en - ### [Google DeepMind Publications](/lab-practices/other-labs/google-deepmind-papers/index.md) by [Google DeepMind](https://deepmind.google/) Google DeepMind publications index: research papers across AI and science. Advanced learning_resource paper West en - ### [Latxa: An Open Language Model and Evaluation Suite for Basque](/lab-practices/other-labs/hitz-latxa-paper/index.md) by [Julen Etxaniz et al. (University of the Basque Country)](https://www.ehu.eus) arXiv paper (2024) introducing Latxa, an open Basque language model and evaluation suite. Advanced–Expert learning_resource paper Spain en - ### [Meltemi: The first open Large Language Model for Greek](/lab-practices/other-labs/ilsp-meltemi-paper/index.md) by [Leon Voukoutis et al. (ILSP, Athena Research Center)](https://www.ilsp.gr/) arXiv paper (2024) introducing Meltemi, an open large language model for Greek. Advanced–Expert learning_resource paper Greece en - ### [Mitigating Catastrophic Forgetting in Language Transfer via Model Merging](/lab-practices/other-labs/insait-branch-and-merge-paper/index.md) by [Anton Alexandrov et al. (INSAIT)](https://insait.ai/) EMNLP 2024 Findings paper (INSAIT) on reducing catastrophic forgetting in language transfer via Branch-and-Merge model merging. Advanced–Expert learning_resource paper Bulgaria en - ### [GePpeTto Carves Italian into a Language Model](/lab-practices/other-labs/italianlp-geppetto-paper/index.md) by [Lorenzo De Mattei et al. (ItaliaNLP Lab, ILC-CNR)](https://www.ilc.cnr.it/) CEUR-WS paper (PDF) presenting GePpeTto, an Italian GPT-2 language model. Advanced–Expert learning_resource paper Italy - ### [jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval](/lab-practices/other-labs/jina-ai-jina-embeddings-v4-paper/index.md) by [Michael Günther et al. (Jina AI)](https://jina.ai/) arXiv paper (2025) introducing jina-embeddings-v4, universal multimodal multilingual embeddings. Advanced–Expert learning_resource paper EU-DE en - ### [Swedish Whispers: Leveraging a Massive Speech Corpus for Swedish Speech Recognition](/lab-practices/other-labs/kblab-papers/index.md) by [Leonora Vesterbacka et al. (National Library of Sweden)](https://kb.se) arXiv paper (2025) by KBLab on Swedish Whisper models trained on a large Swedish speech corpus. Advanced–Expert learning_resource paper Sweden en - ### [Kimi K3: Open Frontier Intelligence](/lab-practices/other-labs/kimi-k3-open-frontier-intelligence/index.md) by [Kimi Team (Moonshot AI)](https://www.moonshot.ai/) arXiv paper (2026) Kimi K3 technical report from Moonshot AI. Advanced learning_resource paper China en - ### [Kling-Omni Technical Report](/lab-practices/other-labs/kuaishou-kling-omni-technical-report/index.md) by [Kling Team (Kuaishou)](https://www.kuaishou.com/) arXiv HTML version of the Kling-Omni technical report (2025) on Kuaishou's video model architecture and training. Advanced learning_resource paper China en - ### [LAION-5B: An open large-scale dataset for training next generation image-text models](/lab-practices/other-labs/laion-laion-5b-paper/index.md) by [Christoph Schuhmann et al. (LAION)](https://laion.ai/) arXiv paper (2022) introducing LAION-5B, an open dataset of 5.85 billion image-text pairs. Advanced–Expert learning_resource paper EU-DE en - ### [LoRA Without Regret](/lab-practices/other-labs/lora-without-regret/index.md) by [Thinking Machines Lab](https://thinkingmachines.ai/) Thinking Machines Lab post LoRA Without Regret: when LoRA matches full fine-tuning, with hyperparameter guidance. Advanced learning_resource paper US en - ### [Luma Unified Models (Uni-1)](/lab-practices/other-labs/luma-unified-models-uni-1/index.md) by [Luma AI](https://lumalabs.ai/) Luma AI Uni-1 page: multimodal reasoning model that generates images. Advanced learning_resource paper US en - ### [MERA paper](/lab-practices/other-labs/mera-paper/index.md) by [Alena Fenogenova et al. (SberDevices)](https://www.sber.ru/) arXiv paper (2024) introducing MERA, a benchmark for evaluating LLMs in Russian. Advanced learning_resource paper Russia en - ### [MiniMax-M1 paper](/lab-practices/other-labs/minimax-m1-paper/index.md) by [MiniMax](https://www.minimax.io/) arXiv paper (2025) describing MiniMax-M1, scaling test-time compute with lightning attention. Advanced learning_resource paper China en - ### [MiniMax Sparse Attention](/lab-practices/other-labs/minimax-sparse-attention/index.md) by [Xunhao Lai et al. (Peking University)](https://www.pku.edu.cn/) arXiv paper (2026) describing MiniMax Sparse Attention, the attention design accompanying MiniMax-M3. Advanced learning_resource paper China en - ### [Mistral AI Research](/lab-practices/other-labs/mistral-ai-research-publications/index.md) by [Mistral AI](https://mistral.ai/) Mistral AI research page listing papers and technical reports on its language, speech, document and robotics models. Advanced learning_resource blog France en - ### [Muon is Scalable for LLM Training (Moonlight)](/lab-practices/other-labs/muon-is-scalable-for-llm-training-moonlight/index.md) by [Jingyuan Liu et al. (Moonshot AI)](https://www.moonshot.ai/) arXiv paper (2025) by Moonshot AI showing the Muon optimizer scales to LLM training, with the Moonlight model. Advanced learning_resource paper China en - ### [Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework](/lab-practices/other-labs/nask-polish-vision-language-adaptation-paper/index.md) by [Grzegorz Statkiewicz et al. (NASK National Research Institute)](https://www.nask.pl/) arXiv paper (2026) by NASK on annotation-efficient adaptation of vision-language models to Polish. Advanced–Expert learning_resource paper Poland en - ### [CLOVA Tech Blog](/lab-practices/other-labs/naver-clova-tech-blog-state-aligned/index.md) by [NAVER Cloud (CLOVA)](https://clova.ai/) NAVER CLOVA tech blog with technical posts on HyperCLOVA X models and their serving. Advanced learning_resource blog Korea en|ko - ### [NorwAI's Large Language Models: Technical Report](/lab-practices/other-labs/norwai-ntnu-papers/index.md) by [Jon Atle Gulla et al. (NTNU)](https://www.ntnu.edu/) arXiv paper (2026) technical report on NorwAI's Norwegian large language models. Advanced–Expert learning_resource paper Norway en - ### [Nous Research blog](/lab-practices/other-labs/nous-research-blog/index.md) by [Nous Research](https://nousresearch.com/) Nous Research blog with posts on Hermes models, the Psyche training network and research. Advanced learning_resource blog US en - ### [GlórIA: A Generative and Open Large Language Model for Portuguese](/lab-practices/other-labs/nova-lincs-gl-ria-paper/index.md) by [Ricardo Lopes et al. (NOVA LINCS, NOVA University Lisbon)](https://nova-lincs.di.fct.unl.pt/) arXiv paper (2024) introducing GlórIA, an open generative language model for European Portuguese. Advanced–Expert learning_resource paper Portugal en - ### [OpenLLM-Ro: Technical Report on Open-source Romanian LLMs](/lab-practices/other-labs/openllm-ro-technical-report/index.md) by [Mihai Masala et al. (National University of Science and Technology POLITEHNICA Bucharest)](https://upb.ro/) arXiv paper (2024) technical report on OpenLLM-Ro, open-source Romanian LLMs. Advanced–Expert learning_resource paper Romania en - ### [Preferred Networks 日本語 technology blog](/lab-practices/other-labs/preferred-networks-technology-blog/index.md) by [Preferred Networks, Inc.](https://preferred.jp/) Preferred Networks Japanese tech blog with research and engineering writeups by its staff. Advanced learning_resource guide Japan ja - ### [Prime Intellect blog](/lab-practices/other-labs/prime-intellect-blog/index.md) by [Prime Intellect](https://primeintellect.ai/) Prime Intellect blog with posts on decentralised training, INTELLECT models and its Environments Hub. Advanced learning_resource blog West en - ### [Qwen3 Technical Report](/frontier/model-releases/qwen3-technical-report/index.md) by [An Yang et al. (Qwen Team, Alibaba)](https://qwen.ai/) arXiv paper (2025) Qwen3 Technical Report describing the Qwen3 model family, architecture and training. Advanced learning_resource paper China en [Previous page](/foundation-models-labs/advanced/page/20/index.md) [Next page](/foundation-models-labs/advanced/page/22/index.md) [Plain-text Markdown copy](/foundation-models-labs/advanced/page/21/index.md)