Tootfinder

What are Cowboys thinking? Lack of free-agency moves, futures of Dak Prescott, CeeDee Lamb https://theathletic.com/5440132/2024/04/23/cowboys-future-dak-prescott-ceedee-lamb-free-agency/

What are Cowboys thinking? Lack of free-agency moves, futures of Dak Prescott, CeeDee Lamb
Jerry Jones had plenty of non-draft questions to answer at the Cowboys' pre-draft news conference Tuesday.

Mahmood supports TogetherSF's cynical #MayorDictator measure, which claims to be about reforming commissions but really gives the mayor 2/3 appointing power with no confirmation to the powerful commissions it retains, making them a puppet show.
Probably a good idea to vote for @…

Series D, Episode 02 - Power
TARRANT: We're not going anywhere, Orac, until you give me straight answers to two questions. One, how did the Seska woman get out of here? And two, how does that method enable us to deactivate the nuclear compression charge?
ORAC: Answers. One, the woman made her exit via the secondary hatch.
https://blake.torpidity.net/m/402/441 B7B5

SELF-[IN]CORRECT: LLMs Struggle with Refining Self-Generated Responses
Dongwei Jiang, Jingyu Zhang, Orion Weller, Nathaniel Weir, Benjamin Van Durme, Daniel Khashabi
https://arxiv.org/abs/2404.04298

SELF-[IN]CORRECT: LLMs Struggle with Refining Self-Generated Responses
Can LLMs continually improve their previous outputs for better results? An affirmative answer would require LLMs to be better at discriminating among previously-generated alternatives, than generating initial responses. We explore the validity of this hypothesis in practice. We first introduce a unified framework that allows us to compare the generative and discriminative capability of any model on any task. Then, in our resulting experimental analysis of several LLMs, we do not observe the per…

This https://arxiv.org/abs/2311.12180 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

cuPDLP.jl: A GPU Implementation of Restarted Primal-Dual Hybrid Gradient for Linear Programming in Julia
In this paper, we provide an affirmative answer to the long-standing question: Are GPUs useful in solving linear programming? We present cuPDLP.jl, a GPU implementation of restarted primal-dual hybrid gradient (PDHG) for solving linear programming (LP). We show that this prototype implementation in Julia has comparable numerical performance on standard LP benchmark sets to Gurobi, a highly optimized implementation of the simplex and interior-point methods. This demonstrates the power of using G…

Comparison results for Markov tree distributions
Jonathan Ansari, Moritz Ritter
https://arxiv.org/abs/2404.17441 https://arxiv.org/pd…

Comparison results for Markov tree distributions
We develop comparison results for Markov tree distributions extending ordering results from the literature on discrete time Markov processes and recently studied ordering results for conditionally independent factor models to tree structures. Based on fairly natural positive dependence conditions, our main contribution is a comparison result with respect to the supermodular order. Since this order is a pure dependence order, it has many applications in optimal transport, finance, and insurance.…

Can Large Language Models put 2 and 2 together? Probing for Entailed Arithmetical Relationships
D. Panas, S. Seth, V. Belle
https://arxiv.org/abs/2404.19432 https://arxiv.org/pdf/2404.19432
arXiv:2404.19432v1 Announce Type: new
Abstract: Two major areas of interest in the era of Large Language Models regard questions of what do LLMs know, and if and how they may be able to reason (or rather, approximately reason). Since to date these lines of work progressed largely in parallel (with notable exceptions), we are interested in investigating the intersection: probing for reasoning about the implicitly-held knowledge. Suspecting the performance to be lacking in this area, we use a very simple set-up of comparisons between cardinalities associated with elements of various subjects (e.g. the number of legs a bird has versus the number of wheels on a tricycle). We empirically demonstrate that although LLMs make steady progress in knowledge acquisition and (pseudo)reasoning with each new GPT release, their capabilities are limited to statistical inference only. It is difficult to argue that pure statistical learning can cope with the combinatorial explosion inherent in many commonsense reasoning tasks, especially once arithmetical notions are involved. Further, we argue that bigger is not always better and chasing purely statistical improvements is flawed at the core, since it only exacerbates the dangerous conflation of the production of correct answers with genuine reasoning ability.