Albert is a software engineer working at Meta, interested in security and machine learning.
Outside of work he researches argumentation, knowledge representation, and how to make reasoning by large language models verifiable.
Workshop on Personalized, Aligned, Long-term Memory for AI Systems (PALM 2026), co-located with Conference on Neural Information Processing Systems (NeurIPS 2026)
38th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2026)
Sixth International Workshop on Systems and Algorithms for Formal Argumentation (SAFA 2026), co-located with 11th International Conference on Computational Models of Argument (COMMA 2026)
30th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES 2026)
Workshop on Logical Aspects in Multi-Agent Systems and Strategic Reasoning (LAMAS&SR 2026), co-located with Federated Logic Conference (FLoC 2026)
8th International Workshop on EXplainable, Trustworthy, and Responsible AI and Multi-Agent Systems (EXTRAAMAS 2026), co-located with International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026)
Preprint, 2026
34th ACM International Conference on Information and Knowledge Management (CIKM 2025)
Proceedings of the 34th ACM International Conference on Information and Knowledge Management (2025) 2535-2545
29th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES 2025)
Procedia Computer Science 270 (2025) 2166-2175
A team proposes an architectural change to a production service. The proposal collects arguments about latency, operational complexity, deployment cost, and regression risk. Nobody …
I went to AAMAS this year. One talk, on neuro-argumentative learning, caught my attention. The pitch, roughly: a model reasons through a structured debate over past cases, with …
A few months ago I wrote about splitting legal reasoning into three steps: entity identification, property extraction, and symbolic rule application. It worked for hearsay …
Six months ago, your team gave a client a 15% discount to keep the deal alive. Now the client’s subsidiary wants the same terms. You ask your AI assistant what it knows about that …
It’s a fair question to ask in an era where LLMs seem to handle almost anything, why do I even bother with ontologies? To many, the word feels a bit 2005, a relic of the semantic …
In this blogpost, I wrote about making AI reasoning transparent by splitting a prompt into manual steps like entity identification and symbolic application. That approach worked …
The discourse around AI has recently unlocked a new fear: the potential for apps like Meta’s Vibes and OpenAI’s Sora to flood the internet with AI-generated slop. This anxiety …
LLMs give you answers but hide how they got there. Ask ChatGPT about a legal question and you get a conclusion with what looks like solid reasoning. Chain-of-Thought prompting has …
React recently introduced the useActionState hook, and while the official docs provide basic examples, I immediately wanted to tackle a common real-world scenario: preserving form …
TypeScript is an amazing technology, but it has been build under the assumption is compiles down to JavaScript. As a consequence, all the wonderful types one can enjoy during the …
The Model-View-Controller is usually a standard pattern for building web applications. Whether the view is expressed as HTML or perhaps just plain JSON, it helps to structure the …
Full text search is easy to use. But the inherent ambiguity of natural languages causes search results to be biased with false positives. To obtain more accurate results, we need …