Full papers
5-page papers (excluding references)
Machine Learning/CORE Unranked
NFAM 2026
New Frontiers in Associative Memories - Workshop at ICLR 2026
Apr 26 2026Official workshop site Site reachable
The NFAM 2026 workshop, held in conjunction with ICLR 2026 in Rio de Janeiro, brings together researchers to advance a unified understanding of associative memory as a foundational computational paradigm in artificial intelligence. It bridges classical energy-based models with modern architectures like transformers and diffusion models, focusing on memory's role in attention, inference, reasoning, and agentic AI systems.
Key deadlines
Verified September 9, 2026Full paper
February 15, 2026
AoE
Workshop timeline
Submission and decisions
Full paperKey deadline
February 15, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Tiny papers
3-page papers (excluding references)
Research areas in scope
01
Scope and Related Work
Hopfield networks and Dense Associative MemoriesEnergy-based models, attractor dynamics, and score-based or diffusion-based generative modelsAssociative memory as attention, inference, or retrieval in transformers and deep architecturesEnergy TransformersCapacity, stability, convergence, and memorization–generalization tradeoffs in high-dimensional associative systemsOptimization dynamics, test-time training, and adaptation viewed as associative processesAssociative memory for generative modeling in non-Euclidean domains such as manifolds, graphs, and distributionsMemory-augmented architectures for agentic AI, persistent reasoning, tool use, and long-horizon decision makingLifelong learning, continual adaptation, and mitigation of catastrophic forgetting via associative mechanismsMultimodal and structured reasoning using associative recall across language, vision, and embodied settingsNeuroscience- and physics-inspired perspectives on memory, energy landscapes, and transient dynamicsScalable, hardware-efficient, and biologically plausible implementations of associative memoryAnalog and digital hardware design for associative memoryBenchmarks, evaluation protocols, and empirical studies of memory-augmented models
Policies worth checking twice
- Submissions must be anonymized for double-blind review
- Reviews will not be shared publicly
- Accepted papers will be made public on OpenReview
- Supplementary materials after references are allowed and do not count toward the page limit
- Preference given to new, unpublished work
- Standard ICLR conflict of interest rules apply
- Camera-ready submissions must include author names and institutions
- Non-archival submission policy
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026