I-Con: A Unifying Framework for Representation Learning
Paper • 2504.16929 • Published • 31
Theory and Representation learning
-
-
SONAR-LLM: Autoregressive Transformer that Thinks in Sentence Embeddings and Speaks in Tokens
Paper • 2508.05305 • Published • 49 -
The Strong Lottery Ticket Hypothesis for Multi-Head Attention Mechanisms
Paper • 2511.04217 • Published • 17 -
Large Language Models as Markov Chains
Paper • 2410.02724 • Published • 33 -
Alignment Makes Language Models Normative, Not Descriptive
Paper • 2603.17218 • Published • 46 -
V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning
Paper • 2603.14482 • Published • 37 -
Learn2Fold: Structured Origami Generation with World Model Planning
Paper • 2603.29585 • Published • 18 -
Demystifying When Pruning Works via Representation Hierarchies
Paper • 2603.24652 • Published • 20 -
Hallucinations Undermine Trust; Metacognition is a Way Forward
Paper • 2605.01428 • Published • 24 -
RigidFormer: Learning Rigid Dynamics using Transformers
Paper • 2605.09196 • Published • 14 -
Unified Neural Scaling Laws
Paper • 2605.26248 • Published • 7 -
A Theory of Contrastive Learning with Natural Images
Paper • 2607.07470 • Published • 16 -
Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation
Paper • 2607.27372 • Published • 13