Tootfinder

#Windows zu #Linux von einem nicht-technichen Wissenschaftler

Consciously avoiding people using the term “subconscious”
A while ago I read an article where the author used the term "subconscious" and it miffed me. In essence, it goes like this: Serious psychologists do not use the term because it is not possible to determine if the unconscious is above, below, left, right, in-front of, or behind us. You know, because it's not a thing, and it's not sent from the devil or hell, whatever Freud might have hypothesized.

Rethinking Knowledge Distillation: A Data Dependent Regulariser With a Negative Asymmetric Payoff
Israel Mason-Williams, Gabryel Mason-Williams, Helen Yannakoudakis
https://arxiv.org/abs/2510.12615

Rethinking Knowledge Distillation: A Data Dependent Regulariser With a Negative Asymmetric Payoff
Knowledge distillation is often considered a compression mechanism when judged on the resulting student's accuracy and loss, yet its functional impact is poorly understood. In this work, we quantify the compression capacity of knowledge distillation and the resulting knowledge transfer from a functional perspective, decoupling compression from architectural reduction, which provides an improved understanding of knowledge distillation. We employ hypothesis testing, controls, and random control d…

Erfolgreiche Flucht aus dem goldenen Käfig.
Erfahrungsberichts eines Wissenschaftlers über den Wechsel von Windows zu Linux im wissenschaftlichen Alltag
#linux <…

And more to read in your freetime, if you are interested in AI, from Bluesky, Phillip Isola.
"Over the past year, my lab has been working on fleshing out theory applications of the Platonic Representation Hypothesis.
Today I want to share two new works on this topic:"
Eliciting higher alignment:
https://arxiv.org/…

Words That Make Language Models Perceive
Large language models (LLMs) trained purely on text ostensibly lack any direct perceptual experience, yet their internal representations are implicitly shaped by multimodal regularities encoded in language. We test the hypothesis that explicit sensory prompting can surface this latent structure, bringing a text-only LLM into closer representational alignment with specialist vision and audio encoders. When a sensory prompt tells the model to 'see' or 'hear', it cues the model to resolve its next…

Engineering Dominating Patterns: A Fine-grained Case Study
Jonathan Dransfeld, Marvin K\"unnemann, Mirza Redzic, Marcus Wunderlich
https://arxiv.org/abs/2510.12232 https://…

Engineering Dominating Patterns: A Fine-grained Case Study
The \emph{Dominating $H$-Pattern} problem generalizes the classical $k$-Dominating Set problem: for a fixed \emph{pattern} $H$ and a given graph $G$, the goal is to find an induced subgraph $S$ of $G$ such that (1) $S$ is isomorphic to $H$, and (2) $S$ forms a dominating set in $G$. Fine-grained complexity results show that on worst-case inputs, any significant improvement over the naive brute-force algorithm is unlikely, as this would refute the Strong Exponential Time Hypothesis. Nevertheless…

Two-cardinal Kurepa Hypotheses
Fanxin Wu
https://arxiv.org/abs/2510.08860 https://arxiv.org/pdf/2510.08860

Two-cardinal Kurepa Hypotheses
We consider the two-cardinal Kurepa Hypothesis $\mathsf{KH}(κ,λ)$. We observe that if $κ\leqλ<μ$ are infinite cardinals then $\lnot\mathsf{KH}(κ,λ)\land\mathsf{KH}(κ,μ)\rightarrow\mathsf{KH}(λ^+,μ)$, and show that in some sense this is the only $\mathsf{ZFC}$ constraint. The case of singular $λ$ and its relation to Chang's Conjecture and scales is discussed. We also extend an independence result about Kurepa and Aronszajn trees due to Cummings to the case of successors of singular c…

AI Agents as Universal Task Solvers
Alessandro Achille, Stefano Soatto
https://arxiv.org/abs/2510.12066 https://arxiv.org/pdf/2510.12066

AI Agents as Universal Task Solvers
AI reasoning agents are already able to solve a variety of tasks by deploying tools, simulating outcomes of multiple hypotheses and reflecting on them. In doing so, they perform computation, although not in the classical sense -- there is no program being executed. Still, if they perform computation, can AI agents be universal? Can chain-of-thought reasoning solve any computable task? How does an AI Agent learn to reason? Is it a matter of model size? Or training dataset size? In this work, w…

Hypothesis testing for the dimension of random geometric graph
Mingao Yuan, Feng Yu
https://arxiv.org/abs/2510.11844 https://arxiv.org/pdf/2510.11844

Hypothesis testing for the dimension of random geometric graph
Random geometric graphs (RGGs) offer a powerful tool for analyzing the geometric and dependence structures in real-world networks. For example, it has been observed that RGGs are a good model for protein-protein interaction networks. In RGGs, nodes are randomly distributed over an $m$-dimensional metric space, and edges connect the nodes if and only if their distance is less than some threshold. When fitting RGGs to real-world networks, the first step is probably to input or estimate the dimens…

A compressed code for memory discrimination
Dale Zhou, Sharon Mina Noh, Nora C Harhen, Nidhi V Banavar, C. Brock Kirwan, Michael A Yassa, Aaron M Bornstein
https://arxiv.org/abs/2510.10791

A compressed code for memory discrimination
The ability to discriminate similar visual stimuli is an important index of memory function. This ability is widely thought to be supported by expanding the dimensionality of relevant neural codes, such that neural representations for similar stimuli are maximally distinct, or ``separated.'' An alternative hypothesis is that discrimination is supported by lossy compression of visual inputs, efficiently coding sensory information by discarding seemingly irrelevant details. A benefit of compressi…