Curriculum

Public work

List of publications, conferences and preprints in reverse chronological order.

• [ArXiV 2026] K. Brilliantov*, A. Hernández-Cano*, and E. Abbé. How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?.

  > pdf <

• [ArXiV 2026] A. Hägele*, A. Hernández-Cano*, A. Kosson, and M. Jaggi. Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors.

  > pdf <

• [ACL 2026] A. Hernández-Cano*, A. Hägele*, A. H. Huang*, A. Romanou*, et. al.* Apertus: Democratizing Open and Compliant LLMs for Global Language Environments.

  > pdf <

• [ICLR 2026 (Workshop)] S. Mehra, A. Hernández-Cano, and M. Jaggi. Can the Future Inform the Present? Investigating Latent Lookahead Refinement via Multi-Token Prediction.

  > pdf <

• [NeurIPS 2025] A. Hernández-Cano*, D. Garbaya*, I. Schlag, M. Jaggi. Towards Fully FP8 GEMM LLM Training at scale.

  > pdf | presentation <

• [Front. AI 2024] A. Hernández-Cano, Y. Ni, Z. Zou, A. Zakeri, M. Imani. Hyperdimensional Computing with Holographic and Adaptive Encoder.

  > pdf <

• [IEEE Acess 2024] Y. Ni, D. Abraham, M. Issa, A. Hernández-Cano, et al. Dynamic MAC Protocol for Wireless Spectrum Sharing via Hyperdimensional Self-Learning.

  > pdf <

• [ArXiV 2023] Z. Chen, A. Hernández-Cano*, A. Romanou*, A. Bonnet*, K. Matoba*, et. al. “MediTron-70B: Scaling Medical Pretraining for Large Language Models”.

  > pdf | data & finetuning code | trainer code | huggingface team <

• [PLoS One 2023] E. Correa-Padilla, A. Hernández-Cano, G. Cuevas, Y. Acevedo-Betancur, F. Esquivel-Guadarrama, K. Martinez-Mayorga. “Modifications in the piperazine ring of nucleozin affect anti-influenza activity”.

  > pdf <

• [DATE 2023] J.R. Rubalcava-Cortés, A. Hernández-Cano, A.C. Pacheco-Tovar, F. Imani, R. Cammarota, M. Imani. “Privacy-Preserving Neural Representation for Brain-Inspired Learning”.

  > code <

• [ACS Fall 2021] A. Hernández-Cano, J. Naveja-Romero, A. Madariaga-Mazon, K. Martinez-Mayorga. “Molecular de novo design using context-free grammars”.

• [DAC 2021] A. Hernández-Cano, R. Cammarota, M. Imani. “PRID: Model Inversion Privacy Attacks in Hyperdimensional Learning Systems”.

  > pdf <

• [DAC 2021] A. Hernández-Cano, M. Zhou, C. Zhou, X. Yin, M. Imani. “RegHD: Robust and Efficient Regression in Hyper-Dimensional Learning System”.

  > pdf <

• [DATE 2021] A. Hernández-Cano, Y. Kim, M. Imani. “A Framework for Efficient and Binary Clustering in High-Dimensional Space”.

  > pdf | code <

• [GLSVLSI 2021] A. Hernández-Cano, C. Zhuo, X. Yin, M. Imani. “Real-Time and Robust Hyperdimensional Classification”.

  > pdf | code <

• [DATE 2021] A. Hernández-Cano, N. Matsumoto, E. Ping, M. Imani. “OnlineHD: Robust, Efficient, and Single-Pass Online Learning Using Hyperdimensional System”.

  Erratum: The encoding equation is wrong, please refer to equation 1 in “Real-Time and Robust Hyperdimensional Classification”.

  > pdf | code <

• [AIP 2020] L. Hernández-Cano, A. Hernández-Cano, A. Leonor-Rivera. “Sleep staging by hyperdimensional dense networks”.

  > pdf <

(*) = Equal contributions.

Hackathons

• [LauzHack-LLMs 2025] 🏅 Tralala: From text to prototype: effortless 3D design with AI. Organizer’s prize in the APIs track.
• [LauzHack 2024] 🥈 Chronica: Create, manage, and visualize notes effectively and securely. Second Place overall.
• [HackUPC 2024] 🥇🥇 WALL-M: Bringing structure to your e-mail inbox using RAG and hybrid search. First Place: Use of Taipy. First Place: Best use of GenAI using Vector Search.
• [LauzHack 2023] Legacy LM: A personalized conversational AI model, preserving voices and legacies for future connection
• [LauzHack 2022] 🥈 Q0: An online platform for personalized study maps and adaptive learning content. Second Place overall.