Tootfinder

This https://arxiv.org/abs/2312.08901 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Fewer is More: Boosting LLM Reasoning with Reinforced Context Pruning
Large Language Models (LLMs) have shown impressive capabilities, yet they still struggle with math reasoning. In this work, we propose CoT-Influx, a novel approach that pushes the boundary of few-shot Chain-of-Thoughts (CoT) learning to improve LLM mathematical reasoning. Motivated by the observation that adding more concise CoT examples in the prompt can improve LLM reasoning performance, CoT-Influx employs a coarse-to-fine pruner to maximize the input of effective and concise CoT examples. Th…

Context is For Trying to Understand Reality, Not Winning Arguments. That's the Problem., by @… https://www.readtpa.com/p/context-is-for-t

Context is For Trying to Understand Reality, Not Winning Arguments. That's the Problem.
A smart piece at The Atlantic has me thinking about our too-short attention spans and motivated reasoning.

Motivated reasoning: Weshalb Menschen den #Klimawandel leugnen.
#motivatedreasoning
https://…

Weshalb leugnen Menschen den Klimawandel? » latinapress Nachrichten
Der Klimawandel ist kein neues Phänomen. Er beschreibt die langfristigen Veränderungen von Faktoren wie der Temperatur, des Niederschlages und der Meeresströmungen. Diese Veränderungen wurden in den letzten Jahrzehnten durch zahlreiche Studien und Beobachtungen weltweit nachgewiesen. Aktuell leidet Südamerika unter einer ungewöhnlichen Hitzewelle, in Brasilien, Paraguay, Argentinien, Uruguay und Chile werden Rekord Temperaturen registriert. Seit Mittwoch

This https://arxiv.org/abs/2310.03094 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning
Large language models (LLMs) such as GPT-4 have exhibited remarkable performance in a variety of tasks, but this strong performance often comes with the high expense of using paid API services. In this paper, we are motivated to study building an LLM cascade to save the cost of using LLMs, particularly for performing reasoning (e.g., mathematical, causal) tasks. Our cascade pipeline follows the intuition that simpler questions can be addressed by a weaker but more affordable LLM, whereas only t…

This https://arxiv.org/abs/2305.02925 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDM_…

Maximum values of the Sombor-index-like graph invariants of trees and connected graphs
A set of novel vertex-degree-based invariants was introduced by Gutman, denoted by \newline $SO_1, SO_2, \ldots,SO_6$. These invariants were constructed through geometric reasoning based on a new graph invariant framework. Motivated by proposed open problems in [Z. Tang, Q. Li, H. Deng, \textit{Trees with Extremal Values of the Sombor-Index-Like Graph Invariants}, MATCH Commun. Math. Comput. Chem. \textbf{90} (2023) 203-222], we have found the maximum values of $SO_5$ and $SO_6$ in the set of m…

This https://arxiv.org/abs/2310.03094 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning
Large language models (LLMs) such as GPT-4 have exhibited remarkable performance in a variety of tasks, but this strong performance often comes with the high expense of using paid API services. In this paper, we are motivated to study building an LLM cascade to save the cost of using LLMs, particularly for performing reasoning (e.g., mathematical, causal) tasks. Our cascade pipeline follows the intuition that simpler questions can be addressed by a weaker but more affordable LLM, whereas only t…