Cover Image

Linear-Readout Floors and Threshold Recovery in Computation in Superposition

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Borobia, Hector, Seguí-Mas, Elies, Tormo-Carbó, Guillermina
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Information Theory I.2.6; F.1.1
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models
by: Borobia, Hector, et al.
Published: (2026)

Where Should LoRA Go? Component-Type Placement in Hybrid Language Models
by: Borobia, Hector, et al.
Published: (2026)

Component-Aware Self-Speculative Decoding in Hybrid Language Models
by: Borobia, Hector, et al.
Published: (2026)

Emergence of Quantised Representations Isolated to Anisotropic Functions
by: Bird, George
Published: (2025)

The Affine Divergence: Aligning Activation Updates Beyond Normalisation
by: Bird, George
Published: (2025)

Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion
by: Lee, Dohoon, et al.
Published: (2024)

From Provable Correctness to Probabilistic Generation: A Comparative Review of Program Synthesis Paradigms
by: Kobaladze, Zurabi, et al.
Published: (2025)

The Inclusion Depth of Pattern Languages: An Open Problem in Algorithmic Learning Theory
by: Luo, Wei
Published: (2026)

On the Compatibility of Generative AI and Generative Linguistics
by: Portelance, Eva, et al.
Published: (2024)

Prompting Neural-Guided Equation Discovery Based on Residuals
by: Brugger, Jannis, et al.
Published: (2025)

Binarized Neural Networks Converge Toward Algorithmic Simplicity: Empirical Support for the Learning-as-Compression Hypothesis
by: Sakabe, Eduardo Y., et al.
Published: (2025)

Information-Theoretic Quality Metric of Low-Dimensional Embeddings
by: Gutiérrez-Bernal, Sebastián, et al.
Published: (2025)

Towards a Neural Lambda Calculus: Neurosymbolic AI Applied to the Foundations of Functional Programming
by: Flach, João, et al.
Published: (2023)

Does Editing Provide Evidence for Localization?
by: Wang, Zihao, et al.
Published: (2025)

Maximum Mean Discrepancy on Exponential Windows for Online Change Detection
by: Kalinke, Florian, et al.
Published: (2022)

Tricks and Plug-ins for Gradient Boosting with Transformers
by: Fang, Biyi, et al.
Published: (2025)

Hybrid Quantum-Classical Mixture of Experts: Unlocking Topological Advantage via Interference-Based Routing
by: Heddad, Reda, et al.
Published: (2025)

Functional Component Ablation Reveals Specialization Patterns in Hybrid Language Model Architectures
by: Borobia, Hector, et al.
Published: (2026)

Nyström $M$-Hilbert-Schmidt Independence Criterion
by: Kalinke, Florian, et al.
Published: (2023)

Decoding Rewards in Competitive Games: Inverse Game Theory with Entropy Regularization
by: Liao, Junyi, et al.
Published: (2026)

Privacy-Preserving Explainable AIoT Application via SHAP Entropy Regularization
by: Sharma, Dilli Prasad, et al.
Published: (2025)

Theoretical Analysis of Positional Encodings in Transformer Models: Impact on Expressiveness and Generalization
by: Li, Yin
Published: (2025)

Leibniz's Monadology as Foundation for the Artificial Age Score: A Formal Architecture for Al Memory Evaluation
by: Kayadibi, Seyma Yaman
Published: (2025)

Graded Transformers
by: Shaska Sr, Tony
Published: (2025)

Entropy-Reservoir Bregman Projection: An Information-Geometric Unification of Model Collapse
by: Chen, Jingwei
Published: (2025)

Internalizing Tools as Morphisms in Graded Transformers
by: Shaska, Tony
Published: (2025)

Understanding the Nature of Generative AI as Threshold Logic in High-Dimensional Space
by: Levin, Ilya
Published: (2026)

Information-Theoretic Measures in AI: A Practical Decision Guide
by: Papadopoulos, Nikolaos Al., et al.
Published: (2026)

Neural-Guided Equation Discovery
by: Brugger, Jannis, et al.
Published: (2025)

STACHE: Local Black-Box Explanations for Reinforcement Learning Policies
by: Elashkin, Andrew, et al.
Published: (2025)

NeurOptimisation: The Spiking Way to Evolve
by: Cruz-Duarte, Jorge Mario, et al.
Published: (2025)

SATORIS-N: Spectral Analysis based Traffic Observation Recovery via Informed Subspaces and Nuclear-norm minimization
by: Mohanty, Sampad, et al.
Published: (2026)

Privacy as Commodity: MFG-RegretNet for Large-Scale Privacy Trading in Federated Learning
by: Sun, Kangkang, et al.
Published: (2026)

Probabilistic Digital Twins of Users: Latent Representation Learning with Statistically Validated Semantics
by: David, Daniel
Published: (2025)

One Policy, Infinite NPCs: Persona-Traceable Shared RL Policies for Scalable Game Agents
by: Hong, Yoosung
Published: (2026)

Long-Term Electricity Demand Prediction Using Non-negative Tensor Factorization and Genetic Algorithm-Driven Temporal Modeling
by: Masaki, Toma, et al.
Published: (2025)

The Serial Scaling Hypothesis
by: Liu, Yuxi, et al.
Published: (2025)

HybridVFL: Disentangled Feature Learning for Edge-Enabled Vertical Federated Multimodal Classification
by: Anoosha, Mostafa, et al.
Published: (2025)

The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
by: Pal, Biplab, et al.
Published: (2026)

Taming Polysemanticity in LLMs: Provable Feature Recovery via Sparse Autoencoders
by: Chen, Siyu, et al.
Published: (2025)