Tootfinder

My department is looking to hire a professor of practice in CS, with a focus on AI. Job posting below. If you have questions I'll do my best to answer them, else find someone who can! We are in Providence, easy commute access from Boston.
https://

Professor of the Practice in Computer Science with a Focus on Artificial Intelligence - Providence, Rhode Island job with Computer Science | 37970844
The Department of Computer Science at Brown University is seeking applicants for a Professor of the Practice position at the assistant, associate, ...

«KI-Modelle sind anfällig für wiederholte Angriffe:
Laut Forschern von Cisco versagen KI-Modelle bei realistischen Multi-Turn-Angriffen und lassen an Sicherheits-Benchmarks auf Basis weniger Prompts zweifeln.»
Der moderne Widerspruch ist die KI oder was ist es sonnst? So klug wie KI angeboten wird ist es einfach nicht.
🤖

KI-Modelle sind anfällig für wiederholte Angriffe
Laut Forschern von Cisco versagen KI-Modelle bei realistischen Multi-Turn-Angriffen und lassen an Sicherheits-Benchmarks auf Basis weniger Prompts zweifeln.

Should Demand Models Incorporate Competitor Prices? Oblivious Learning and Algorithmic Collusion
Yuhang Wu, Assaf Zeevi
https://arxiv.org/abs/2606.05363 https://arxiv.org/pdf/2606.05363 https://arxiv.org/html/2606.05363
arXiv:2606.05363v1 Announce Type: new
Abstract: On a platform with many sellers, should a pricing algorithm explicitly model competitors' prices when learning demand? Classical learning arguments suggest an affirmative answer: ignoring competitors induces model misspecification and inefficiency. In contrast, recent work on algorithmic collusion suggests that strategic obliviousness -- deliberately ignoring competitor prices -- may facilitate collusive outcomes and improve profits. We study this modeling choice in a stylized competitive market with unknown noisy demand, in which multiple sellers repeatedly set prices and estimate demand via iterated least squares, and either incorporate competitors' prices into their demand models (informed) or ignore them (oblivious). We first show that, relative to a monopolist, an oblivious seller in a competitive market must explore more aggressively to compensate for the loss of dynamic competitor information. Building on this insight, we characterize market dynamics when all sellers are oblivious and show that prices converge to the competitive outcome under sufficient exploration, while a continuum of pseudo-equilibria arises when exploration decays. Analyzing the resulting price trajectories, we uncover an excursion phenomenon that gives rise to transient collusive patterns that dissipate as learning progresses. In markets with both oblivious and informed sellers, the informed strictly out-earn the oblivious. Read as a strategy game, the modeling choice has a unique Nash equilibrium: the all-informed market, in which prices converge to the competitive outcome efficiently. Overall, our results indicate that collusive patterns are not robust and are not sustained by oblivious modeling; therefore, incorporating competitor information, together with sufficient price exploration, remains a reliable strategy for sellers in competitive markets.
toXiv_bot_toot

I think the government has identified that 2026 is a really bad year for replacing a large number of personal computers. Sorry, everyone.
https://questions-statements.parliament.uk/written-questions/detail/2026-04-13/hl16286

Written questions and answers - Written questions, answers and statements - UK Parliament
Information from UK Parliament on written questions & answers, written statements and daily reports.

from my link log —
42 is an answer to the question, what is the sum of three cubes?
https://aperiodical.com/2019/09/42-is-the-answer-to-the-question-what-is-80538738812075974³-8043575…

42 is the answer to the question “what is (-80538738812075974)³ + 80435758145817515³ + 12602123297335631³?”
We now know that the number 42 can be written as the sum of three cubes: \[ 42 = (-80538738812075974)^3 + 80435758145817515^3 + 12602123297335631^3 \] This computational breakthrough was achieved i…

Anker announces Thus, a compute-in-memory chip it says will bring on-device AI to its products and accessories, starting with its upcoming Soundcore earbuds (John Higgins/The Verge)
https://www.theverge.com/tech/916463/anker-thus-chip-announcement

Anker made its own chip to bring AI to all its products
The Thus chip will come to earbuds first before being rolled out to other Anker products.

Call the GI Rights Hotline at 1-877-447-4487.
Call for yourself or someone you care about

Free and confidential

One hotline for a nationwide network of counseling centers
https://girightshotline.org/

The Computer Science Fetish https://mail.cyberneticforests.com/the-computer-science-fetish/

The Computer Science Fetish
On the Valorization of Technical Authority 💡Nerd Alert: 1/5. Readable, but also mostly venting about weird encounters at a conference in San Francisco. The academic AI critic is often either bored or angry. Bored because everyone keeps saying the same things, or angry because all the same things need

The Brain That Goes Quiet: Serving a Large Model's Knowledge at 131 Tokens per Second on an 8 GB Laptop by Removing the Large Model from the Runtime Path
Myeong Jun Jo
https://arxiv.org/abs/2606.12154 https://arxiv.org/pdf/2606.12154 https://arxiv.org/html/2606.12154
arXiv:2606.12154v1 Announce Type: new
Abstract: In earlier work I showed that a 35B-class Mixture-of-Experts model can be loaded and executed on a consumer laptop with 8 GB of GPU memory. That result solved a placement problem and immediately exposed a different one: even correctly placed, the large model needed roughly four seconds to answer, because it was still being invoked at every query. This paper documents what happened when I stopped invoking it. During an offline phase, the large model reads source documents and writes verified answer entries into a structured knowledge store; at runtime, only a lightweight router, a deterministic renderer, and a 1B-class model are active. On the same 8 GB laptop, end-to-end response time fell from approximately 4,465 ms to 518 ms, effective end-to-end throughput rose from 15.7 to 131 tokens per second, and the small model's streaming decode rate held at 226-237 tokens per second with a time-to-first-token of 29-62 ms. The bottleneck is structural: three different large models (Qwen, Gemma, and GLM class) all showed the same multi-second runtime cost, and all three produced usable knowledge stores offline. On a 563-entry store built from seventeen real documents, keyword routing collapsed to 1.5% top-1 accuracy while BM25-based routing reached 92.8% (99.4% top-3), and a confidence gate raised effective top-1 to 98.0% by escalating 12.3% of queries. Exact-match fidelity of the small model ranged from 9/9 to 0/9 across envelope formats carrying identical content. A 16-case verification gate blocked all ten corrupted entries while admitting all six supported ones.
toXiv_bot_toot

Ask why, and know the answers are unsatisfying and contingent. It made sense at the time. Because we didn't understand yet. Because changing it would certainly break something else and we didn't know what yet and didn't have time to understand it. Yet.
Computers can be understood. Everything that is going on can be peeked at, prodded at, taken apart to know what is going on. Inside every complex system are simpler parts. The complexity comes from the combination of them, not mystery.
I cannot emphasize that enough. Just because you don't know, or nobody you know knows why something is how it is does not mean it is unknowable.