<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Blog on Alfonso Amayuelas</title><link>https://www.amayuelas.me/blog/</link><description>Recent content in Blog on Alfonso Amayuelas</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 20 Nov 2026 18:54:08 -0800</lastBuildDate><atom:link href="https://www.amayuelas.me/blog/index.xml" rel="self" type="application/rss+xml"/><item><title>Planning to Explore: Curiosity-Driven Planning for LLM Test Generation</title><link>https://www.amayuelas.me/blog/planning-to-explore/</link><pubDate>Tue, 14 Apr 2026 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/planning-to-explore/</guid><description>How curiosity-driven planning, grounded in a 35-year thread of information-theoretic exploration, helps LLMs break through the coverage plateau in automated test generation.</description></item><item><title>About a Research Visit to KAUST</title><link>https://www.amayuelas.me/blog/kaust-research-visit/</link><pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/kaust-research-visit/</guid><description>Reflections on a research visit to KAUST in Saudi Arabia, working with Jurgen Schmidhuber&amp;rsquo;s lab on curiosity-driven exploration and its application to LLM-based test generation.</description></item><item><title>Grounding LLM Reasoning with Knowledge Graphs</title><link>https://www.amayuelas.me/blog/grounding-llm-reasoning-with-kgs/</link><pubDate>Sat, 01 Mar 2025 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/grounding-llm-reasoning-with-kgs/</guid><description>We discuss the integration of LLMs with Knowledge Graphs (KGs) to ground their reasoning capabilities. It highlights the importance of KGs in providing structured knowledge that is interpretable and computationally efficient. The integration of KGs with LLMs helps in reducing hallucinations and ensuring that the generated outputs are grounded in verifiable data.</description></item><item><title>Towards a Reliable and Trustworthy MultiAgent Debate</title><link>https://www.amayuelas.me/blog/reliable-multiagent-debate/</link><pubDate>Thu, 28 Nov 2024 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/reliable-multiagent-debate/</guid><description>We study adversarial attacks in multi-agent LLM debate, showing a single adversarial agent can persuade other models toward incorrect answers, and explore uncertainty-based methods to make multi-agent communication more reliable.</description></item><item><title>Exploring the Frontier of Uncertainty with Large Language Models: The Role of Known-Unknown Questions</title><link>https://www.amayuelas.me/blog/knowledge-of-knowledge/</link><pubDate>Thu, 20 Jun 2024 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/knowledge-of-knowledge/</guid><description>LLMs tend to be overconfident, always providing an answer to the user&amp;rsquo;s query, even when the model is unsure about the answer. What happens when we ask the LLMs questions for which the answer is unclear even to humans?</description></item><item><title>Neural Methods for Logical Reasoning over Knowledge Graphs</title><link>https://www.amayuelas.me/blog/neural-methods-for-logical-reasoning-over-knowledge-graphs/</link><pubDate>Sun, 13 Nov 2022 00:00:00 +0000</pubDate><guid>https://www.amayuelas.me/blog/neural-methods-for-logical-reasoning-over-knowledge-graphs/</guid><description>In this work, we explore and try to improve how we can answer complex logical queries over knowledge graphs. We call them complex logical queries because the questions can be decomposed into multiple logical statements to reach the final answer.</description></item></channel></rss>