Tootfinder

CLARity: Reasoning Consistency Alone Can Teach Reinforced Experts
Jiuheng Lin, Cong Jiang, Zirui Wu, Jiarui Sun, Yansong Feng
https://arxiv.org/abs/2510.09278 https://

CLARity: Reasoning Consistency Alone Can Teach Reinforced Experts
Training expert LLMs in domains with scarce data is difficult, often relying on multiple-choice questions (MCQs). However, standard outcome-based reinforcement learning (RL) on MCQs is risky. While it may improve accuracy, we observe it often degrades reasoning quality such as logical consistency. Existing solutions to supervise reasoning, such as large-scale Process Reward Models (PRMs), are prohibitively expensive. To address this, we propose CLARity, a cost-effective RL framework that enhanc…

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning
Haozhan Li, Yuxin Zuo, Jiale Yu, Yuhao Zhang, Zhaohui Yang, Kaiyan Zhang, Xuekai Zhu, Yuchen Zhang, Tianxing Chen, Ganqu Cui, Dehui Wang, Dingxiang Luo, Yuchen Fan, Youbang Sun, Jia Zeng, Jiangmiao Pang, Shanghang Zhang, Yu Wang, Yao Mu, Bowen Zhou, Ning Ding
https://arxiv.org/a…

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning
Vision-Language-Action (VLA) models have recently emerged as a powerful paradigm for robotic manipulation. Despite substantial progress enabled by large-scale pretraining and supervised fine-tuning (SFT), these models face two fundamental challenges: (i) the scarcity and high cost of large-scale human-operated robotic trajectories required for SFT scaling, and (ii) limited generalization to tasks involving distribution shift. Recent breakthroughs in Large Reasoning Models (LRMs) demonstrate tha…

Retrocedamos solo un año, hasta 1983, cuando R.E.M. publicaba su primer larga duración (antes habían publicado el EP Chronic Town). ¡Cómo no la iban a romper en los círculos universitarios y underground con este discazo que aún sigue sonando fresco y original!
Murmur
#REM #NowPlaying

Inversion-Raumfrachter soll jeden Ort der Welt in 60 Minuten erreichen
Inversions Raumfrachter sollen militärische Güter im All lagern und in kürzester Zeit weltweit ausliefern – das Projekt wirft weltraumpolitische Fragen auf.

Knapp 30 Jahre davor hatte ich als damaliger Umweltreferent des u-asta auch schon Aktionen zum gleichen Thema. Eine der langsam mahlenden Mühlen in #freiburg
https://www.

Mittagessen statt Autoverkehr: Aktion in Freiburg fordert Neudenken der Rempartstraße
Die Rempartstraße soll autofrei werden. Dafür haben Greenpeace Freiburg und Students for Future eine Aktion gestartet, bei der der gesperrte Straßenabschnitt zum gemeinsamen Mittagessen einlud.

Reinforcing Diffusion Models by Direct Group Preference Optimization
Yihong Luo, Tianyang Hu, Jing Tang
https://arxiv.org/abs/2510.08425 https://arxiv.org/…

Reinforcing Diffusion Models by Direct Group Preference Optimization
While reinforcement learning methods such as Group Relative Preference Optimization (GRPO) have significantly enhanced Large Language Models, adapting them to diffusion models remains challenging. In particular, GRPO demands a stochastic policy, yet the most cost-effective diffusion samplers are based on deterministic ODEs. Recent work addresses this issue by using inefficient SDE-based samplers to induce stochasticity, but this reliance on model-agnostic Gaussian noise leads to slow convergenc…

Elterntaxis - Lasst sie laufen!
https://www.sueddeutsche.de/bayern/elterntaxis-bayern-nuernberg-schulen-kampagne-li.3335344

Umstrittene Elterntaxis: Stadt Nürnberg startet neue Kampagne zur Verkehrssicherheit
In Nürnberg startet ein Pilotprojekt zur Reduzierung von Elterntaxis an Grundschulen. Es will Kinder ermutigen, selbständig zur Schule zu gehen.

Can SSD-Mamba2 Unlock Reinforcement Learning for End-to-End Motion Control?
Gavin Tao, Yinuo Wang, Jinzhao Zhou
https://arxiv.org/abs/2509.07593 https://ar…

Can SSD-Mamba2 Unlock Reinforcement Learning for End-to-End Motion Control?
End-to-end reinforcement learning for motion control promises unified perception-action policies that scale across embodiments and tasks, yet most deployed controllers are either blind (proprioception-only) or rely on fusion backbones with unfavorable compute-memory trade-offs. Recurrent controllers struggle with long-horizon credit assignment, and Transformer-based fusion incurs quadratic cost in token length, limiting temporal and spatial context. We present a vision-driven cross-modal RL fra…

In-Context Policy Adaptation via Cross-Domain Skill Diffusion
Minjong Yoo, Woo Kyung Kim, Honguk Woo
https://arxiv.org/abs/2509.04535 https://arxiv.org/pdf…

In-Context Policy Adaptation via Cross-Domain Skill Diffusion
In this work, we present an in-context policy adaptation (ICPAD) framework designed for long-horizon multi-task environments, exploring diffusion-based skill learning techniques in cross-domain settings. The framework enables rapid adaptation of skill-based reinforcement learning policies to diverse target domains, especially under stringent constraints on no model updates and only limited target domain data. Specifically, the framework employs a cross-domain skill diffusion scheme, where domai…