Tootfinder

Lachlan Murdoch completes a deal to secure control of his family's media empire, ensuring that it remains conservative; source: the deal is valued at $3.3B (New York Times)
https://www.nytimes.com/2025/09/08/bu…

Lachlan Murdoch Secures Control of Fox and News Corp, Ending Succession Fight
Lachlan Murdoch will take control of a new family trust in a deal worth $3.3 billion, ensuring that his father’s media empire will retain its conservative slant.

Is It Time To Treat Prompts As Code? A Multi-Use Case Study For Prompt Optimization Using DSPy
Francisca Lemos (ALGORITMI Research Centre/LASI, University of Minho), Victor Alves (ALGORITMI Research Centre/LASI, University of Minho), Filipa Ferraz (ALGORITMI Research Centre/LASI, University of Minho)
https://arxiv.org/abs/2507.0…

Is It Time To Treat Prompts As Code? A Multi-Use Case Study For Prompt Optimization Using DSPy
Although prompt engineering is central to unlocking the full potential of Large Language Models (LLMs), crafting effective prompts remains a time-consuming trial-and-error process that relies on human intuition. This study investigates Declarative Self-improving Python (DSPy), an optimization framework that programmatically creates and refines prompts, applied to five use cases: guardrail enforcement, hallucination detection in code, code generation, routing agents, and prompt evaluation. Each …

this #seeedstudio #xiao #rp2350 looks cute!
https://www.

Seeed Studio XIAO RP2350
The XIAO RP2350 packs the power of the Raspberry Pi RP2350 (switchable architecture of dual Arm Cortex-M33 cores running at 150MHz with FPU, and dual open-hardware Hazard3 RISC‑V cores, enhanced security and encryption) into the classic XIAO form factor. Measuring just 21x17.8mm, it features 19 multifunction GPIOs, an RGB LED, and a Battery Management System with ultra-low power consumption of 27μA, battery power supply, and direct battery voltage measurement. Thanks to the XIAO ecosystem, t…

A Clinically-Grounded Two-Stage Framework for Renal CT Report Generation
Renjie Liang, Zhengkang Fan, Jinqian Pan, Chenkun Sun, Russell Terry, Jie Xu
https://arxiv.org/abs/2506.23584

A Clinically-Grounded Two-Stage Framework for Renal CT Report Generation
Generating radiology reports from CT scans remains a complex task due to the nuanced nature of medical imaging and the variability in clinical documentation. In this study, we propose a two-stage framework for generating renal radiology reports from 2D CT slices. First, we extract structured abnormality features using a multi-task learning model trained to identify lesion attributes such as location, size, enhancement, and attenuation. These extracted features are subsequently combined with the…

FLM-Audio: Natural Monologues Improves Native Full-Duplex Chatbots via Dual Training
Yiqun Yao, Xiang Li, Xin Jiang, Xuezhi Fang, Naitong Yu, Wenjia Ma, Aixin Sun, Yequan Wang
https://arxiv.org/abs/2509.02521

FLM-Audio: Natural Monologues Improves Native Full-Duplex Chatbots via Dual Training
Full-duplex dialog models are designed to listen and speak simultaneously with rapid responses to fast-changing user input. Among existing approaches, native full-duplex models merges different channels (e.g. listen and speak) in a single time step, overcoming the high response latency inherent to time-division multiplexing time-division multiplexing (TDM) alternatives. Yet, a key challenge remains: aligning textual monologues with audio streams that operate at different bitrates. The prevailin…

So much of the work we do in software is about asking not just what •does• happen when the developer runs their own code, but what •could• happen when the code is running in the wild, out from under the watchful eye of its author.
That’s really, really hard. It’s a large portion of what makes development time-consuming and labor-intensive.
And it’s something you can only do if you actually understand what the code •means• — both to the computers and to the humans.
4/

QuickSilver -- Speeding up LLM Inference through Dynamic Token Halting, KV Skipping, Contextual Token Fusion, and Adaptive Matryoshka Quantization
Danush Khanna, Aditya Kumar Guru, Srivarshinee Sridhar, Zidan Ahmed, Rubhav Bahirwani, Meetu Malhotra, Vinija Jain, Aman Chadha, Amitava Das, Kripabandhu Ghosh
https://arxiv.org/abs/2…

QuickSilver -- Speeding up LLM Inference through Dynamic Token Halting, KV Skipping, Contextual Token Fusion, and Adaptive Matryoshka Quantization
Inference accounts for the majority of latency and energy consumption in large language model (LLM) deployments, often exceeding 90% of total cost. While training-time efficiency has seen extensive progress, runtime optimization remains a key bottleneck, particularly under autoregressive decoding. Existing approaches -- such as pruning, quantization, early exits, and speculative decoding -- often require retraining, architectural changes, or disrupt decoding compatibility. We introduce QuickSil…

Each NFL team's biggest challenge: Will Bengals play defense in '25? Did Chiefs, Texans fix offensive line? https://www.nfl.com/news/each-nfl-team-s-biggest-challenge-will-bengals-play-defense-in-25-did-chiefs-…

Hi friends of research software 👋
The @… is running a free workshop on test-driven development on Monday.
https://events.digital-research.academ

Embodied Intelligence for Sustainable Flight: A Soaring Robot with Active Morphological Control
Ghadeer Elmkaiel, Syn Schmitt, Michael Muehlebach
https://arxiv.org/abs/2508.19684

Embodied Intelligence for Sustainable Flight: A Soaring Robot with Active Morphological Control
Achieving both agile maneuverability and high energy efficiency in aerial robots, particularly in dynamic wind environments, remains challenging. Conventional thruster-powered systems offer agility but suffer from high energy consumption, while fixed-wing designs are efficient but lack hovering and maneuvering capabilities. We present Floaty, a shape-changing robot that overcomes these limitations by passively soaring, harnessing wind energy through intelligent morphological control inspired by…