Tootfinder

An early morning alone on the banks of the (still young) Isar river some weeks ago... Really more "flow" rather than #WaterfallWednesday, but I just absolutely cherish places and moments like this & feeling so privileged being able to experience these after decades of living in large cities... (each time also a reminder how much we stand to lose...)

Apache Flink is uniquely positioned to serve as the backbone for AI agents, equipping them with the powerful new tool of stream processing. Join Steffen Hoellinger at this year's Berlin Buzzwords to explore how Flink jobs can be transformed into “Agents”—autonomous, goal-driven entities that dynamically interact with data streams, trigger actions, and adapt their behaviour based on real-time insights.
Learn more:

An illustration of formal moduli problems with differential graded Lie algebras
Ethan Eugene Wynner
https://arxiv.org/abs/2506.12977 https://

An illustration of formal moduli problems with differential graded Lie algebras
This article provides an exposition to the topic of formal moduli problems, emphasizing its connections with differential graded Lie algebras, and mainly following from Jacob Lurie's DAG X: Formal Moduli Problems. As such, this paper should not be viewed as a presentation of original work, but rather a concise introduction to the subject in the form of a set of organized notes. I hoped to make this paper feel welcoming and insightful for the non-expert enjoyer of derived algebraic geometry, lik…

MetaLint: Generalizable Idiomatic Code Quality Analysis through Instruction-Following and Easy-to-Hard Generalization
Atharva Naik, Lawanya Baghel, Dhakshin Govindarajan, Darsh Agrawal, Daniel Fried, Carolyn Rose
https://arxiv.org/abs/2507.11687

MetaLint: Generalizable Idiomatic Code Quality Analysis through Instruction-Following and Easy-to-Hard Generalization
Large Language Models, though successful in code generation, struggle with code quality analysis because they are limited by static training data and can't easily adapt to evolving best practices. We introduce MetaLint, a new instruction-following framework that formulates code quality analysis as the task of detecting and fixing problematic semantic code fragments or code idioms based on high-level specifications. Unlike conventional approaches that train models on static, rule-based data, Met…

FormGym: Doing Paperwork with Agents
Matthew Toles, Rattandeep Singh, Isaac Song Zhou Yu
https://arxiv.org/abs/2506.14079 https://arx…

FormGym: Doing Paperwork with Agents
Completing paperwork is a challenging and time-consuming problem. Form filling is especially challenging in the pure-image domain without access to OCR, typeset PDF text, or a DOM. For computer agents, it requires multiple abilities, including multi-modal understanding, information retrieval, and tool-use. We present a novel form-filling benchmark consisting of 432 fields spread across 55 documents and 3 tasks, requiring knowledge of 236 features per user. We find that baseline VLAs achieve les…

ChatGPT didn't object to putting a red nose on the dog or painting the rabbits green, but at least it didn't suggest I use staplers on the mice.
#scrooged #chatgpt

Meta says it has taken down around 10M Facebook profiles so far this year that were impersonating or repeatedly reusing content from large content producers (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/07/14/following-youtu…

Following YouTube, Meta announces crackdown on 'unoriginal' Facebook content | TechCrunch
Like YouTube, Meta says it won't penalize users who are engaging with other people's content, doing things like making reaction videos, joining a trend, or adding their own takes.

High-gain model-following control for trajectory tracking
Nicals Tietze, Kai Wulff, Johann Reger
https://arxiv.org/abs/2506.13463 https://

High-gain model-following control for trajectory tracking
We consider trajectory tracking for minimum-phase nonlinear systems in Byrnes-Isidori form using the model-following control (MFC) architecture. The tracking problem is motivated by a hierarchical control concept where a higher-level instance provides the reference trajectory at run-time. We present a computational efficient implementation of the feedback linearisation MFC design, and apply high-gain feedback in the process control loop (PCL) to achieve practical tracking in presence of Lipschi…

As data speeds increase, it has become crucial to detect problems as they happen. At this year's Berlin Buzzwords, Olena Kutsenko explained how to build a real-time anomaly detection system using Apache Kafka for streaming, Apache Flink for processing, and AI for pattern recognition, covering Apache Iceberg for storing historical data to improve models.
Watch the full session:

Just received a friend request on Facebook by someone named Peter Waldmeier.
https://www.facebook.com/peter.waldmeier.865
He seemed such a nice person, I think he is a model in a Scandinavian country.
When I asked him why he was sending me a friend request he go…