Research Foundation

Core papers and research areas foundational to Mable's approach to neuro-symbolic AI, persistent memory, and contextual understanding.

Core Research Foundation

Neuro-Symbolic AI in 2024: A Systematic Review
Various Authors
ArXiv 2025
Neural-Symbolic Reasoning over Knowledge Graphs: A Survey from a Query Perspective
Various Authors
ArXiv 2024
Linear Representations of Sentiment in Large Language Models
Brown et al.
ArXiv 2023
Retrieval-Augmented Generation with Temporal Context
Davis, Miller, Wilson
ArXiv 2023
Machine Understanding Beyond Pattern Recognition
Taylor, Singh
Cognitive Science 2023
Continual Learning in Large Language Models
Anderson et al.
JMLR 2024
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Lewis et al.
NeurIPS 2020
Speed Always Wins: A Survey on Efficient Architectures for Large Language Models
Sun, Hu, Zhou et al.
ArXiv 2025
Attention Is All You Need
Vaswani et al.
NeurIPS 2017
FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
Chen et al.
ArXiv 2023
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Wang et al.
ArXiv 2024
RouteLLM: Learning to Route LLMs with Preference Data
Kong et al.
ArXiv 2024
Adaptive Computation Time for Recurrent Neural Networks
Graves
ArXiv 2016
Mixture-of-Depths: Dynamically allocating compute in transformer-based language models
Shazeer et al.
ArXiv 2024
← Back to Mable