The ELIAS Lab builds efficient language-model methods, systems, and orchestrations — smaller, faster, cheaper, greener — to make advanced AI more accessible and easier to deploy. We study the full stack: synthetic data, distillation, adaptive inference, cost-aware algorithms, and agentic retrieval.
Production LLM stacks pay for a full forward pass to generate, then pay again for separate guard and classifier models. We reuse the serving model's hidden states: lightweight probes (as few as 100K parameters) aggregate signals across tokens and layers to match guard-model accuracy on ToxicChat and WildGuardMix — inside a single forward pass. Validated across Llama-3.2, GPT-OSS-20B, and Qwen3-30B.
A training-free agentic retriever that gives a language model control over the breadth–depth tradeoff on knowledge graphs, using only two tools: global lexical search and one-hop neighborhood expansion. Reaches 59.1% avg Hit@1 on STaRK, improving over prior retrieval-based and agentic baselines by up to 31.4 points. Distilled into an 8B model via label-free imitation. Joint work with Harvard Medical School and Oxford.
A cheap inference-time signal for production reliability: entropy trajectories from top-k log-probabilities, summarized into a compact feature vector, feed a lightweight classifier that estimates slice-level accuracy under domain shift. Evaluated across ten STEM reasoning benchmarks and nine LLMs (3B–20B). Tracks held-out accuracy well enough to prioritize data acquisition toward the worst-performing slices.
Pairwise ranking prompting pairs an LLM judge with a classical sorting algorithm — but LLM comparisons are noisy, order-sensitive, and sometimes intransitive, so sorting wastes budget. We reframe reranking as active learning from noisy pairwise comparisons, and introduce a one-call randomized-direction oracle that converts position bias into zero-mean noise. 3–9× fewer LLM calls than sorting PRP at matched NDCG@10.
Luciano is Associate Professor at Universidad de San Andrés, where he directs the ELIAS Lab (Efficient Language Intelligence for Autonomous Systems), focused on applied research in large language models and agentic AI systems. He teaches NLP at the graduate and undergraduate level. He holds a Ph.D. in Computer Science from the Max Planck Institute for Informatics, and his work has been published in top NLP venues including ACL and EMNLP.
With over 15 years across research and industry, he has worked extensively on driving state-of-the-art AI into production at scale. At Microsoft Research (AI Frontiers), he worked on the next generation of efficient LLMs as a senior research engineer — distillation, adaptive computation, and dynamic inference — and contributed to research on LLM orchestration and agentic systems for synthetic data generation and model post-training. At Goldman Sachs, he led an NLP research team applying AI to complex financial and legal data.
He co-founded Lumina Labs, building autonomous systems for investment research, and Ambiverse, a Max Planck Institute spin-off for industrial applications of language technology, where he served as VP of Innovation. He also holds a master's in Computer Science from Imperial College London, a master's in Economics from Universidad de San Andrés, and an advanced bachelor's in Economics from Universidad de Buenos Aires.
Matias Antenaza · Alex Bodner · Santiago Carrillo · Josefina Dehan · Nataly Sol Hofkamp · Segundo Santos · Ignacio Shuemer · Antonio Tepsich · Agustina Videla Rivero · Mora Vigo Malusardi · Facundo Vulcano
ARK (adaptive breadth-depth retrieval on knowledge graphs, with Harvard Medical School and Oxford) and Hidden-State Probes (single-pass classification, Meyoyan & Del Corro) — both main conference, long papers.
The first edition sold out. A second edition of the UdeSA executive program starts September 24.
Eight sessions (Aug–Nov) where practitioners from Microsoft, Mercado Libre, Roboflow, Tiendanube and more share real AI applications. Opens Aug 13 with Martín Sciarrillo (Microsoft).
A talk series by the Department of Engineering and ELIAS Lab: over eight sessions, people from leading companies share real-world AI applications — from design and implementation to the technical and business challenges. Coordinated by Franco Zan. Campus UdeSA (Victoria), Aula Magna · 14:40–16:10.
Programa para profesionales que quieren construir sistemas agénticos en producción. Cubre prompt engineering, RAG basado en agentes, function calling, el protocolo MCP, sistemas multi-agente y despliegue, con un hackatón final sobre un caso de negocio real. Coordinación académica: Luciano Del Corro. Profesor: Juan Wisznia.
Final-year projects (tesinas) from Luciano's courses have repeatedly grown into papers at venues like ACL and EMNLP — most of the lab's publications are first-authored by students.
We're a small group of grad researchers and final-year project students working on agentic AI and LLM efficiency. If you're a student looking to start a research project — a thesis, an internship, your first paper — this is a good place to do it. Students here have published their first paper at NLP venues like ACL and EMNLP, and most of our papers are first-authored by students. Prior ML experience helps, but curiosity matters more.