Tootfinder

This https://arxiv.org/abs/2311.00445 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

A Systematic Comparison of Syllogistic Reasoning in Humans and Language Models
A central component of rational behavior is logical inference: the process of determining which conclusions follow from a set of premises. Psychologists have documented several ways in which humans' inferences deviate from the rules of logic. Do language models, which are trained on text generated by humans, replicate such human biases, or are they able to overcome them? Focusing on the case of syllogisms -- inferences from two simple premises -- we show that, within the PaLM2 family of transfo…

Height Filtrations and Base Loci on Flag Bundles over a Curve
Yangyu Fan, Wenbin Luo, Binggang Qu
https://arxiv.org/abs/2403.06808 https://

Height Filtrations and Base Loci on Flag Bundles over a Curve
Let $k$ be an algebraically closed field of characteristic zero. Let $C/k$ be a projective smooth curve with function field $K=k(C)$ and $G/k$ be a connected reductive group. Let $F$ be a principal $G$-bundle on $C$. Let $P \subseteq G$ be a parabolic subgroup and $λ: P \longrightarrow G$ be a strictly anti-dominant character. Then $F/P \longrightarrow C$ is a flag bundle and $\mathcal{L}_λ=F \times_P k_λ$ on $F/P$ is a relatively ample line bundle. Let $h_{\mathcal{L}_λ}$ be the height f…

"The mental growth of humanity, as revealed in scientific thinking, may be stated as a tendency to unite seemingly different categories into a complex unity into which previous concepts enter as component parts. The evolution of thought is a process of synthesizing concepts. Creation from principles should not be confused with imitation of appearances (mimicry)."
— Joseph Schillinger (1947)
#aiart

This https://arxiv.org/abs/2304.07531 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

Mining for Cost Awareness in the Infrastructure as Code Artifacts of Cloud-based Applications: an Exploratory Study
Context: The popularity of cloud computing as the primary platform for developing, deploying, and delivering software is largely driven by the promise of cost savings. Therefore, it is surprising that no empirical evidence has been collected to determine whether cost awareness permeates the development process and how it manifests in practice. Objective: This study aims to provide empirical evidence of cost awareness by mining open source repositories of cloud-based applications. The focus is o…

This https://arxiv.org/abs/2404.05104 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

Uniform approximation of Betti numbers
We prove that Lefschetz's principle of approximating the cohomology of a possibly singular affine scheme of finite type over a field by the cohomology of a suitable (thickening of a) hyperplane section can be made uniform: in the affine case, we can choose the hyperplane section independently of the cohomology. Using Jouanolou's trick, this gives a new way to bound the Betti numbers of quasi-projective schemes over a field, independently of the cohomology. This is achieved through a motivic ver…

Height Filtrations and Base Loci on Flag Bundles over a Curve
Yangyu Fan, Wenbin Luo, Binggang Qu
https://arxiv.org/abs/2403.06808 https://

Height Filtrations and Base Loci on Flag Bundles over a Curve
Let $k$ be an algebraically closed field of characteristic zero. Let $C/k$ be a projective smooth curve with function field $K=k(C)$ and $G/k$ be a connected reductive group. Let $F$ be a principal $G$-bundle on $C$. Let $P \subseteq G$ be a parabolic subgroup and $λ: P \longrightarrow G$ be a strictly anti-dominant character. Then $F/P \longrightarrow C$ is a flag bundle and $\mathcal{L}_λ=F \times_P k_λ$ on $F/P$ is a relatively ample line bundle. Let $h_{\mathcal{L}_λ}$ be the height f…

This https://arxiv.org/abs/2309.14556 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Art or Artifice? Large Language Models and the False Promise of Creativity
Researchers have argued that large language models (LLMs) exhibit high-quality writing capabilities from blogs to stories. However, evaluating objectively the creativity of a piece of writing is challenging. Inspired by the Torrance Test of Creative Thinking (TTCT), which measures creativity as a process, we use the Consensual Assessment Technique [3] and propose the Torrance Test of Creative Writing (TTCW) to evaluate creativity as a product. TTCW consists of 14 binary tests organized into the…

This https://arxiv.org/abs/2404.05104 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

Uniform approximation of Betti numbers
We prove that Lefschetz's principle of approximating the cohomology of a possibly singular affine scheme of finite type over a field by the cohomology of a suitable (thickening of a) hyperplane section can be made uniform: in the affine case, we can choose the hyperplane section independently of the cohomology. Using Jouanolou's trick, this gives a new way to bound the Betti numbers of quasi-projective schemes over a field, independently of the cohomology. This is achieved through a motivic ver…

Asymptotic Variation of Elementary Abelian p-Extensions over $P^1$
Hui June Zhu
https://arxiv.org/abs/2403.05453 https://arxiv.org/pd…

Asymptotic Variation of Elementary Abelian p-Extensions over $P^1$
In this paper, we prove that there exists a Zariski dense open subset $U$ in the parameter space of all elementary $p$-covers of the projective line that ramified at exactly one point, defined over the rationals, such that for every curve $X$ in $U(\overline{Q})$ and for any prime $p$ large enough, the reduction of $X$ at all primes lying over $p$ achieves its generic Newton slopes.

This https://arxiv.org/abs/2403.00835 has been replaced.
link: https://scholar.google.com/scholar?q=a

CLLMs: Consistency Large Language Models
Parallel decoding methods such as Jacobi decoding show promise for more efficient LLM inference as it breaks the sequential nature of the LLM decoding process and transforms it into parallelizable computation. However, in practice, it achieves little speedup compared to traditional autoregressive (AR) decoding, primarily because Jacobi decoding seldom accurately predicts more than one token in a single fixed-point iteration step. To address this, we develop a new approach aimed at realizing fas…