Proceedings Eighth International Conference on Applied Category Theory

This paper presents the proceedings of the Eighth International Conference on Applied Category Theory (ACT2025), held at the University of Florida in June 2025, which featured a diverse collection of contributions spanning pure and applied disciplines such as computer science, quantum computation, and chemistry.

Amar Hadzihasanovic (Tallinn University of Technology), Jean-Simon Pacaud Lemay (Macquarie University)Tue, 10 Ma🔢 math

BemaGANv2: Discriminator Combination Strategies for GAN-based Vocoders in Long-Term Audio Generation

BemaGANv2 is an advanced GAN-based vocoder that enhances long-term audio generation for Text-to-Music and Text-to-Audio applications by integrating Anti-aliased Multi-Periodicity composition modules in the generator and systematically evaluating novel discriminator combination strategies, including the Multi-Envelope Discriminator, to achieve high-fidelity and temporally coherent results.

Taesoo Park, Mungwi Jeong, Mingyu Park, Narae Kim, Junyoung Kim, Mujung Kim, Jisang Yoo, Hoyun Lee, Sanghoon Kim, Soonchul KwonTue, 10 Ma🤖 cs.LG

FATE: A Formal Benchmark Series for Frontier Algebra of Multiple Difficulty Levels

The paper introduces FATE, a new formal algebra benchmark series spanning from undergraduate exercises to PhD-level research problems, which reveals that current state-of-the-art LLMs struggle significantly with formalizing advanced mathematical reasoning, achieving near-zero accuracy on the most difficult tasks despite stronger natural-language performance.

Jiedong Jiang, Wanyi He, Yuefeng Wang, Guoxiong Gao, Yongle Hu, Jingting Wang, Nailin Guan, Peihao Wu, Chunbo Dai, Liang Xiao, Bin DongTue, 10 Ma🤖 cs.LG

Consistency-based Abductive Reasoning over Perceptual Errors of Multiple Pre-trained Models in Novel Environments

This paper proposes a consistency-based abductive reasoning framework that integrates predictions from multiple pre-trained models at test-time by formulating error management as a logic-constrained optimization problem, thereby significantly improving robustness and accuracy in novel environments with distributional shifts compared to individual models and standard ensembles.

Mario Leiva, Noel Ngu, Joshua Shay Kricheli, Aditya Taparia, Ransalu Senanayake, Paulo Shakarian, Nathaniel Bastian, John Corcoran, Gerardo SimariThu, 12 Ma🤖 cs.AI

Diagonalizing Through the ω\omega-Chain: Iterated Self-Certification on Bounded Turing Machines and its Least Fixed Point

This paper demonstrates that while bounded Turing machines cannot achieve self-certification due to temporal overhead, the iterative advancement of finite halting observations forms an ascending ω\omega-chain whose Scott limit yields the least fixed point, effectively resolving the halting problem through the continuous deferral of diagonalization.

Miara SungMon, 09 Ma💻 cs

LTLGuard: Formalizing LTL Specifications with Compact Language Models and Lightweight Symbolic Reasoning

LTLGuard is a modular framework that enables resource-efficient open-weight language models (4B–14B parameters) to generate correct and conflict-free Linear Temporal Logic (LTL) specifications from informal requirements by combining constrained generation with lightweight symbolic reasoning for iterative consistency checking and refinement.

Medina Andresel, Cristinel Mateis, Dejan Nickovic, Spyridon Kounoupidis, Panagiotis Katsaros, Stavros TripakisMon, 09 Ma🤖 cs.AI