Wieland Brendel
@wielandbrendel.bsky.social
Machine Learning Researcher and Social Entrepreneur | Group Leader at ELLIS Institute Tübingen & Max Planck Institute for Intelligent Systems robustml.is.mpg.de | Co-Founder maddox.ai | Co-Initiator bw-ki.de | @ellis.eu scholar
Pinned
🚀 We’re hiring! Join Bernhard Schölkopf & me at @ellisinsttue.bsky.social to push the frontier of #AI in education!
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
We are super happy about this milestone - a great validation for our three-year-long project Polybot that we are on the right track towards a future in farming automation that can scale up novel regenerative farming practices!
Polybot, founded by @wielandbrendel.bsky.social has secured validation grant from SPRIND to advance its AI-driven agricultural robot. @tuebingen-ai.bsky.social @maxplanckcampus.bsky.social
Find out more: institute-tue.ellis.eu/en/news/from...
Find out more: institute-tue.ellis.eu/en/news/from...
From research to startup: Polybot receives order for AI-based harvesting robotics
institute-tue.ellis.eu
February 25, 2025 at 4:54 PM
We are super happy about this milestone - a great validation for our three-year-long project Polybot that we are on the right track towards a future in farming automation that can scale up novel regenerative farming practices!
Reposted by Wieland Brendel
New preprint out! 🎉
How does LLM training loss translate to downstream performance?
We show that pretraining data and tokenizer shape loss-to-loss scaling, while architecture and other factors play a surprisingly minor role!
brendel-group.github.io/llm-line/ 🧵1/8
How does LLM training loss translate to downstream performance?
We show that pretraining data and tokenizer shape loss-to-loss scaling, while architecture and other factors play a surprisingly minor role!
brendel-group.github.io/llm-line/ 🧵1/8
February 18, 2025 at 2:09 PM
New preprint out! 🎉
How does LLM training loss translate to downstream performance?
We show that pretraining data and tokenizer shape loss-to-loss scaling, while architecture and other factors play a surprisingly minor role!
brendel-group.github.io/llm-line/ 🧵1/8
How does LLM training loss translate to downstream performance?
We show that pretraining data and tokenizer shape loss-to-loss scaling, while architecture and other factors play a surprisingly minor role!
brendel-group.github.io/llm-line/ 🧵1/8
Reposted by Wieland Brendel
CuratedThoughts: Data Curation for RL Datasets 🚀
Since DeepSeek-R1 introduced reasoning-based RL, datasets like Open-R1 & OpenThoughts emerged for fine-tuning & GRPO. Our deep dive found major flaws — 25% of OpenThoughts needed elimination by data curation.
Here's why 👇🧵
Since DeepSeek-R1 introduced reasoning-based RL, datasets like Open-R1 & OpenThoughts emerged for fine-tuning & GRPO. Our deep dive found major flaws — 25% of OpenThoughts needed elimination by data curation.
Here's why 👇🧵
February 17, 2025 at 6:22 PM
CuratedThoughts: Data Curation for RL Datasets 🚀
Since DeepSeek-R1 introduced reasoning-based RL, datasets like Open-R1 & OpenThoughts emerged for fine-tuning & GRPO. Our deep dive found major flaws — 25% of OpenThoughts needed elimination by data curation.
Here's why 👇🧵
Since DeepSeek-R1 introduced reasoning-based RL, datasets like Open-R1 & OpenThoughts emerged for fine-tuning & GRPO. Our deep dive found major flaws — 25% of OpenThoughts needed elimination by data curation.
Here's why 👇🧵
🚀 We’re hiring! Join Bernhard Schölkopf & me at @ellisinsttue.bsky.social to push the frontier of #AI in education!
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
February 11, 2025 at 4:34 PM
🚀 We’re hiring! Join Bernhard Schölkopf & me at @ellisinsttue.bsky.social to push the frontier of #AI in education!
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
Does anyone know how OpenAI gets o3-mini to exceed 700 tokens/sec? I’ve only seen such speeds on specialized chips from @cerebrassystems.bsky.social, @sambanova.bsky.social, or @groq.com—but not on standard Nvidia GPUs, which I assumed power OpenAI’s inference.
February 11, 2025 at 3:57 PM
Does anyone know how OpenAI gets o3-mini to exceed 700 tokens/sec? I’ve only seen such speeds on specialized chips from @cerebrassystems.bsky.social, @sambanova.bsky.social, or @groq.com—but not on standard Nvidia GPUs, which I assumed power OpenAI’s inference.
Reposted by Wieland Brendel
As many areas, science itself will be transformed by AI. We have now released a state-of-the-art survey on this emerging field. We discuss methods and tools for AI-enhanced search, ideation, experimentation, content generation and peer review.
arxiv.org/pdf/2502.05151
arxiv.org/pdf/2502.05151
February 10, 2025 at 8:09 AM
As many areas, science itself will be transformed by AI. We have now released a state-of-the-art survey on this emerging field. We discuss methods and tools for AI-enhanced search, ideation, experimentation, content generation and peer review.
arxiv.org/pdf/2502.05151
arxiv.org/pdf/2502.05151
Reposted by Wieland Brendel
🏹 Job: The Tübingen AI Center, a partner in the OpenEuroLLM project, is seeking a Scientific Project Coordinator to oversee efforts in developing open-source, trustworthy foundation models and leading community engagement.
📍 Tübingen
📅 Apply by March 3rd, 2025
📌 Get details: tuebingen.ai/careers
📍 Tübingen
📅 Apply by March 3rd, 2025
📌 Get details: tuebingen.ai/careers
Careers
Join the Tübingen AI Center and shape the next generation of robust, efficient and accountable machine learning systems.
tuebingen.ai
February 10, 2025 at 3:16 PM
🏹 Job: The Tübingen AI Center, a partner in the OpenEuroLLM project, is seeking a Scientific Project Coordinator to oversee efforts in developing open-source, trustworthy foundation models and leading community engagement.
📍 Tübingen
📅 Apply by March 3rd, 2025
📌 Get details: tuebingen.ai/careers
📍 Tübingen
📅 Apply by March 3rd, 2025
📌 Get details: tuebingen.ai/careers
Reposted by Wieland Brendel
📢 Join our Bundeswettbewerb Künstliche Intelligenz (BWKI) at Didacta 2025 in Stuttgart!
From tomorrow, February 11, to Saturday, February 15, the Didacta education fair will take place in Stuttgart:
📍 Find BWKI in Hall 7, Stand 7D75 as part of the AI & Digital Workshop area - workshops offered!
From tomorrow, February 11, to Saturday, February 15, the Didacta education fair will take place in Stuttgart:
📍 Find BWKI in Hall 7, Stand 7D75 as part of the AI & Digital Workshop area - workshops offered!
February 10, 2025 at 4:00 PM
📢 Join our Bundeswettbewerb Künstliche Intelligenz (BWKI) at Didacta 2025 in Stuttgart!
From tomorrow, February 11, to Saturday, February 15, the Didacta education fair will take place in Stuttgart:
📍 Find BWKI in Hall 7, Stand 7D75 as part of the AI & Digital Workshop area - workshops offered!
From tomorrow, February 11, to Saturday, February 15, the Didacta education fair will take place in Stuttgart:
📍 Find BWKI in Hall 7, Stand 7D75 as part of the AI & Digital Workshop area - workshops offered!