https://matheus.wiki/ Research papers and blog posts by Matheus Kunzler Maldaner: neurosymbolic AI, AI agents, and human-computer interaction. en-us https://matheus.wiki/papers/sentinelbench/ https://matheus.wiki/papers/sentinelbench/ Wed, 03 Jun 2026 00:00:00 GMT AI agents are increasingly asked to carry out work that spans minutes, hours, or longer. Yet the default model of agent behavior is continuous action: issuing tool calls, refreshing pages, searching for alternatives, or otherwise trying to force progress. This is the wrong approach for many long-running tasks, which are better served by a strategy of sustained attention. Instead, agents should mon https://matheus.wiki/pages/blog/anki-screen.html https://matheus.wiki/pages/blog/anki-screen.html Fri, 24 Apr 2026 00:00:00 GMT How I built a system that watches your screen while you study, extracts key concepts, and generates quizzes and flashcards -- all inside a containerized browser. https://matheus.wiki/pages/blog/anki-garmin.html https://matheus.wiki/pages/blog/anki-garmin.html Sat, 14 Mar 2026 00:00:00 GMT How I built a system to review Anki flashcards on a Garmin watch, a desktop overlay, and a ChatGPT integration -- all from a single API bridge. https://matheus.wiki/papers/platos-cave/ https://matheus.wiki/papers/platos-cave/ Thu, 01 Jan 2026 00:00:00 GMT The growing publication rate of research papers has created an urgent need for better ways to fact-check information, assess writing quality, and identify unverifiable claims. We present Plato's Cave as an open-source, human-centered research verification system that (i) creates a directed acyclic graph (DAG) from a document, (ii) leverages web agents to assign credibility scores to nodes and edge https://matheus.wiki/papers/mm-scale/ https://matheus.wiki/papers/mm-scale/ Thu, 01 Jan 2026 00:00:00 GMT Vision-Language Models (VLMs) continue to struggle to make morally salient judgments in multimodal and socially ambiguous contexts. Prior works typically rely on binary or pairwise supervision, which often fail to capture the continuous and pluralistic nature of human moral reasoning. We present MM-SCALE (Multimodal Moral Scale), a large-scale dataset for aligning VLMs with human moral preferences https://matheus.wiki/papers/magentic-ui/ https://matheus.wiki/papers/magentic-ui/ Wed, 01 Jan 2025 00:00:00 GMT AI agents powered by large language models are increasingly capable of autonomously completing complex, multi-step tasks using external tools. Yet, they still fall short of human-level performance in most domains including computer use, software development, and research. Their growing autonomy and ability to interact with the outside world, also introduces safety and security risks including pote https://matheus.wiki/papers/seeing-twice/ https://matheus.wiki/papers/seeing-twice/ Wed, 01 Jan 2025 00:00:00 GMT While generative AI systems have gained popularity in diverse applications, their potential to produce harmful outputs limits their trustworthiness and utility. A small but growing line of research has explored tools and processes to better engage non-AI expert users in auditing generative AI systems. In this work, we present the design and evaluation of MIRAGE, a web-based tool exploring a 'contr https://matheus.wiki/papers/explogic/ https://matheus.wiki/papers/explogic/ Wed, 01 Jan 2025 00:00:00 GMT Constraining deep neural networks (DNNs) to learn individual logic types per node, as performed using the DiffLogic network architecture, opens the door to model-specific explanation techniques that quell the complexity inherent to DNNs. Inspired by principles of circuit analysis from computer engineering, this work presents an algorithm (eXpLogic) for producing saliency maps which explain input p https://matheus.wiki/papers/mirage/ https://matheus.wiki/papers/mirage/ Mon, 01 Jan 2024 00:00:00 GMT While generative AI systems have gained popularity in diverse applications, their potential to produce harmful outputs limits their trustworthiness and usability in different applications. Recent years have seen growing interest in engaging diverse AI users in auditing generative AI that might impact their lives. To this end, we propose MIRAGE as a web-based tool where AI users can compare outputs https://matheus.wiki/papers/xai-syntax/ https://matheus.wiki/papers/xai-syntax/ Mon, 01 Jan 2024 00:00:00 GMT Policy makers, healthcare providers, and defense contractors need to understand many types of machine learning model behaviors. While eXplainable Artificial Intelligence (XAI) provides tools for interpreting these behaviors, few frameworks, surveys, and taxonomies produce succinct yet general notation to help researchers and practitioners describe their explainability needs and quantify whether th