Tootfinder

Can confirm this works on macOS 26, thank you Greg
#retrocomputing #apple #flyingtoasters

StyleAdaptedLM: Enhancing Instruction Following Models with Efficient Stylistic Transfer
Pritika Ramu, Apoorv Saxena, Meghanath M Y, Varsha Sankar, Debraj Basu
https://arxiv.org/abs/2507.18294

StyleAdaptedLM: Enhancing Instruction Following Models with Efficient Stylistic Transfer
Adapting LLMs to specific stylistic characteristics, like brand voice or authorial tones, is crucial for enterprise communication but challenging to achieve from corpora which lacks instruction-response formatting without compromising instruction adherence. We introduce StyleAdaptedLM, a framework that efficiently transfers stylistic traits to instruction-following models using Low-Rank Adaptation (LoRA). LoRA adapters are first trained on a base model with diverse unstructured stylistic corpor…

“A string of bands have pulled out of a UK music festival hours before they were due to perform after Irish band The Mary Wallopers said they were ‘cut off’ for displaying a Palestinian flag.
The Last Dinner Party, Cliffords and The Academic announced on Saturday that they would no longer be performing at the Victorious festival in Portsmouth following Friday’s incident.
…
Rock band The Last Dinner Party said they would boycott the festival in a statement shared on their Inst…

Bands pull out of UK music festival after The Mary Wallopers say they were ‘cut off’
Irish folk group displayed Palestinian flag on stage in Portsmouth

A surge in AI company valuations is driving the popularity of special purpose vehicles, but some involve high fees, opaque structures, and layers of middlemen (Business Insider)
https://africa.businessinsider.com/mar

Investors warn AI FOMO could be fueling a risky bubble in companies like OpenAI and Anthropic
As AI valuations soar into the hundreds of billions, a parallel market of temporary financial shells is proliferating to meet investor FOMO, and many investors are starting to sound the alarm bells that it portends a bubble in the AI gold rush

ComposableNav: Instruction-Following Navigation in Dynamic Environments via Composable Diffusion
Zichao Hu, Chen Tang, Michael J. Munje, Yifeng Zhu, Alex Liu, Shuijing Liu, Garrett Warnell, Peter Stone, Joydeep Biswas
https://arxiv.org/abs/2509.17941

ComposableNav: Instruction-Following Navigation in Dynamic Environments via Composable Diffusion
This paper considers the problem of enabling robots to navigate dynamic environments while following instructions. The challenge lies in the combinatorial nature of instruction specifications: each instruction can include multiple specifications, and the number of possible specification combinations grows exponentially as the robot's skill set expands. For example, "overtake the pedestrian while staying on the right side of the road" consists of two specifications: "overtake the pedestrian" and…

In-person and telephone town halls used to be a common way for members of Congress to connect with constituents following a busy session.
Earlier this year, though, the NRCC urged Republicans to avoid town halls altogether
— and those lawmakers seem to be listening.
Take the tense scene inside a packed auditorium in Lincoln, Nebraska, recently,
where Republican Rep. Mike Flood faced hundreds of constituents during an August 4 town hall.
"How can you justi…

Just reported that I found a couple of dead Salamander at beginning of August in the #Karwendel . 😞
Luckily I was following the Blog of karwendel.org. Otherwise I wouldn't have known that reports about such incidents should be reported:

Tatterfrau, Tattermann und der Salamander-Fresser - Naturpark Karwendel ⛰️
Eine Pilz-Krankheit breitet sich unter heimischen Salamander- und Molch-Arten aus. Deshalb sollen kranke Alpensalamander gemeldet werden.

Found an app called “Octave Coffee”. It’s an opinionated timer for pourover coffee. Tells you how much to pour, then how much to wait, and so on. And it has a cute pixel art kitty.
The problem is… it negs you while doing it, and your reward at the end for following the instructions is the message, “You actually listened. Shocking.”
I did it twice. It doesn’t vary.
So… I’m deleting it, even though I found it useful, because I don’t want to be insulted first thing in the morn…

From #LizOyer @lawyer_oyer@instagram.com
We have a five-alarm fire inside the #DepartmentOfJustice. I’ve been telling you about the dominoes falling over the last few months. Now it’s all coming to a head. We must stay alert, informed, and outspoken.

Checklists Are Better Than Reward Models For Aligning Language Models
Vijay Viswanathan, Yanchao Sun, Shuang Ma, Xiang Kong, Meng Cao, Graham Neubig, Tongshuang Wu
https://arxiv.org/abs/2507.18624

Checklists Are Better Than Reward Models For Aligning Language Models
Language models must be adapted to understand and follow user instructions. Reinforcement learning is widely used to facilitate this -- typically using fixed criteria such as "helpfulness" and "harmfulness". In our work, we instead propose using flexible, instruction-specific criteria as a means of broadening the impact that reinforcement learning can have in eliciting instruction following. We propose "Reinforcement Learning from Checklist Feedback" (RLCF). From instructions, we extract checkl…