Tootfinder

Following the expected mention of technology neutrality with regard to the 2035 review of CO₂ standards for cars, the automotive section of Von der Leyen's State of the Union speech ends on a positive note:
https://ec.europa.eu/commission/presscorner/detail/en/SPEECH…

Here’s the lightning sketch of Paul’s Treatise Against Efficiency that I’ve never written:
1. Efficiency is asymptotically inefficient: as costs approach zero, the cost of further reducing them approaches infinity.
2. Efficiency prioritizes the measurable over the difficult-to-measure.
3. Efficiency prioritizes what those in power see (or imagine) over on-the-ground reality.
4. Following from 2 and 3, efficiency reduces the amount and quality of information flowing into a human system.
5. Efficiency foments institutional inflexibility.
6. By removing slack, efficiency causes small failures to cascade more readily and increases the risk of catastrophic failure.
7. Following rom 4, 5, and 6, efficiency trades small costs for massive risks: from failures, from missed opportunities, and from inability to adjust.
8. Efficiency, when pushed, strangles the emergent phenomena that in the long term create all new things of value.
9. Thus, although it can be a by-product of evolution, efficiency as a goal in itself strangles evolution.
10. Efficiency as a goal strangles joy.

Learning to Walk with Less: a Dyna-Style Approach to Quadrupedal Locomotion
Francisco Affonso, Felipe Andrade G. Tommaselli, Juliano Negri, Vivian S. Medeiros, Mateus V. Gasparino, Girish Chowdhary, Marcelo Becker
https://arxiv.org/abs/2509.06296

Learning to Walk with Less: a Dyna-Style Approach to Quadrupedal Locomotion
Traditional RL-based locomotion controllers often suffer from low data efficiency, requiring extensive interaction to achieve robust performance. We present a model-based reinforcement learning (MBRL) framework that improves sample efficiency for quadrupedal locomotion by appending synthetic data to the end of standard rollouts in PPO-based controllers, following the Dyna-Style paradigm. A predictive model, trained alongside the policy, generates short-horizon synthetic transitions that are gra…

🚁 Small but mighty: A seed-inspired monocopter idea takes flight
#drones

Small but mighty: A seed-inspired monocopter idea takes flight
From a seed-inspired design to a 26-minute flight time on a single rotor, a new monocopter developed by SUTD researchers marks a 10-year journey towards redefining how efficient small flying robots can be.

MST-Distill: Mixture of Specialized Teachers for Cross-Modal Knowledge Distillation
Hui Li, Pengfei Yang, Juanyang Chen, Le Dong, Yanxin Chen, Quan Wang
https://arxiv.org/abs/2507.07015

MST-Distill: Mixture of Specialized Teachers for Cross-Modal Knowledge Distillation
Knowledge distillation as an efficient knowledge transfer technique, has achieved remarkable success in unimodal scenarios. However, in cross-modal settings, conventional distillation methods encounter significant challenges due to data and statistical heterogeneities, failing to leverage the complementary prior knowledge embedded in cross-modal teacher models. This paper empirically reveals two critical issues in existing approaches: distillation path selection and knowledge drift. To address …

All-optical polarization control in time-varying low-index films via plasma symmetry breaking
Wallace Jaffray, Sven Stengel, Alexandra Boltasseva, Vladimir M. Shalaev, Maria Antonietta Vincenti, Domenico De Ceglia, Michael Scalora, Carlo Rizza, Marcello Ferrera
https://arxiv.org/abs/2510.06985

All-optical polarization control in time-varying low-index films via plasma symmetry breaking
Controlling the polarization state of light with sub-picosecond speed and subwavelength precision remains a key challenge for next-generation nanophotonic devices. Conventional methods such as birefringent crystals, liquid crystals, or electro-optic Pockels cells are limited in speed, compactness, and energy efficiency. While structured materials and two-dimensional heterostructures offer potential for on-chip ultrafast performance, achieving all-optical control remains an open problem. Here we…

Thickness-Dependent Spin Pumping in YIG/W$_{90}$Ti$_{10}$ Bilayers
Marielle Hachem, Zeinab Harajli, Samih Isber, Mohammad Haidar
https://arxiv.org/abs/2507.06831

Thickness-Dependent Spin Pumping in YIG/W$_{90}$Ti$_{10}$ Bilayers
We investigate the spin pumping efficiency in YIG/YIG/W$_{90}$Ti$_{10}$ bilayers by measuring the thickness dependence of both the YIG and WTi layers using broadband ferromagnetic resonance (FMR) spectroscopy. The deposition of a 5-nm WTi layer leads to enhanced Gilbert damping in thinner YIG films, indicating efficient spin current injection. From the spin pumping contribution to the damping of the YIG/WTi bilayer, we determine an effective spin mixing conductance of $ 3.3 \times 10^{18}~\math…

Robust and Adaptive Spectral Method for Representation Multi-Task Learning with Contamination
Yian Huang, Yang Feng, Zhiliang Ying
https://arxiv.org/abs/2509.06575 https://

Robust and Adaptive Spectral Method for Representation Multi-Task Learning with Contamination
Representation-based multi-task learning (MTL) improves efficiency by learning a shared structure across tasks, but its practical application is often hindered by contamination, outliers, or adversarial tasks. Most existing methods and theories assume a clean or near-clean setting, failing when contamination is significant. This paper tackles representation MTL with an unknown and potentially large contamination proportion, while also allowing for heterogeneity among inlier tasks. We introduce …

IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards
Xu Guo, Tianyi Liang, Tong Jian, Xiaogui Yang, Ling-I Wu, Chenhui Li, Zhihui Lu, Qipeng Guo, Kai Chen
https://arxiv.org/abs/2508.04632

IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards
Reinforcement Learning with Verifiable Rewards (RLVR) improves instruction following capabilities of large language models (LLMs), but suffers from training inefficiency due to inadequate difficulty assessment. Moreover, RLVR is prone to over-optimization, where LLMs exploit verification shortcuts without aligning to the actual intent of user instructions. We introduce Instruction Following Decorator (IFDecorator}, a framework that wraps RLVR training into a robust and sample-efficient pipeline…

Hierarchically Engineered Titanium Suboxide Films for High-Efficiency Solar Thermal Conversion
Silpa S, Ann Eliza Joseph, Srinivas G, Harish C Barshilia, Vinayak B Kamble
https://arxiv.org/abs/2510.03710

Hierarchically Engineered Titanium Suboxide Films for High-Efficiency Solar Thermal Conversion
We report the development of broadband solar absorber coatings based on titanium suboxide composite thin films on aluminium substrates. The films are fabricated via scalable DC magnetron sputtering using a Ti target, followed by post-annealing in a fixed $O_2$ partial pressure of 0.45 mbar. By tuning deposition time and annealing temperature, a composite phase of $Ti_2O_3$ and $TiO_2$ was achieved. The Raman mapping of the films substantiates the distribution and coexistence of the two phases. …