Cover Image

Disentangling Polysemantic Neurons with a Null-Calibrated Polysemanticity Index and Causal Patch Interventions

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Gupta, Manan, Kumar, Dhruv
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Disentangling Polysemantic Channels in Convolutional Neural Networks
by: Hesse, Robin, et al.
Published: (2025)

Interpretability Without Tradeoffs: Disentangling Polysemanticity At Equal Predictive Performance
by: Bağcı, Doğukan, et al.
Published: (2026)

PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits
by: Dreyer, Maximilian, et al.
Published: (2024)

Tackling Polysemanticity with Neuron Embeddings
by: Foote, Alex
Published: (2024)

CoE: Chain-of-Explanation via Automatic Visual Concept Circuit Description and Polysemanticity Quantification
by: Yu, Wenlong, et al.
Published: (2025)

NeuronScope: A Multi-Agent Framework for Explaining Polysemantic Neurons in Language Models
by: Liu, Weiqi, et al.
Published: (2026)

Polysemanticity and Capacity in Neural Networks
by: Scherlis, Adam, et al.
Published: (2022)

BB-Patch: BlackBox Adversarial Patch-Attack using Zeroth-Order Optimization
by: Kumar, Satyadwyoom, et al.
Published: (2024)

CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention
by: Tang, Jiacheng, et al.
Published: (2026)

Disentanglement in T-space for Faster and Distributed Training of Diffusion Models with Fewer Latent-states
by: Gupta, Samarth, et al.
Published: (2025)

PaCX-MAE: Physiology-Augmented Chest X-Ray Masked Autoencoder
by: Liu, Yancheng, et al.
Published: (2026)

PatchAlign:Fair and Accurate Skin Disease Image Classification by Alignment with Clinical Labels
by: Aayushman, et al.
Published: (2024)

The Rate-Distortion-Polysemanticity Tradeoff in SAEs
by: Mencattini, Tommaso, et al.
Published: (2026)

PatchFlow: Leveraging a Flow-Based Model with Patch Features
by: Zhang, Boxiang, et al.
Published: (2026)

Mixstyle-Entropy: Domain Generalization with Causal Intervention and Perturbation
by: Tang, Luyao, et al.
Published: (2024)

Patch Ranking: Efficient CLIP by Learning to Rank Local Patches
by: Wu, Cheng-En, et al.
Published: (2024)

Learning in the Null Space: Small Singular Values for Continual Learning
by: Pham, Cuong Anh, et al.
Published: (2026)

VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations
by: Zargarbashi, Fatemeh, et al.
Published: (2026)

C$^{2}$INet: Realizing Incremental Trajectory Prediction with Prior-Aware Continual Causal Intervention
by: Li, Xiaohe, et al.
Published: (2024)

SkeySpot: Automating Service Key Detection for Digital Electrical Layout Plans in the Construction Industry
by: Dosi, Dhruv, et al.
Published: (2025)

Patch2Loc: Learning to Localize Patches for Unsupervised Brain Lesion Detection
by: Baker, Hassan, et al.
Published: (2025)

REOrdering Patches Improves Vision Models
by: Kutscher, Declan, et al.
Published: (2025)

Reproducibility Study of CDUL: CLIP-Driven Unsupervised Learning for Multi-Label Image Classification
by: Shah, Manan, et al.
Published: (2024)

Finding the Trigger: Causal Abductive Reasoning on Video Events
by: Le, Thao Minh, et al.
Published: (2025)

Linear Explanations for Individual Neurons
by: Oikarinen, Tuomas, et al.
Published: (2024)

Disentangle and Regularize: Sign Language Production with Articulator-Based Disentanglement and Channel-Aware Regularization
by: Tasyurek, Sumeyye Meryem, et al.
Published: (2025)

Polysemantic Experts, Monosemantic Paths: Routing as Control in MoEs
by: Ye, Charles, et al.
Published: (2026)

Rényi Attention Entropy for Patch Pruning
by: Aizawa, Hiroaki, et al.
Published: (2026)

MedicalPatchNet: A Patch-Based Self-Explainable AI Architecture for Chest X-ray Classification
by: Wienholt, Patrick, et al.
Published: (2025)

GNSP: Gradient Null Space Projection for Preserving Cross-Modal Alignment in VLMs Continual Learning
by: Peng, Tiantian, et al.
Published: (2025)

Predicting Traffic Flow with Federated Learning and Graph Neural with Asynchronous Computations Network
by: Yaqub, Muhammad, et al.
Published: (2024)

PatchMorph: A Stochastic Deep Learning Approach for Unsupervised 3D Brain Image Registration with Small Patches
by: Skibbe, Henrik, et al.
Published: (2023)

Application of Disentanglement to Map Registration Problem
by: Song, Hae Jin, et al.
Published: (2024)

Disentanglement with Factor Quantized Variational Autoencoders
by: Baykal, Gulcin, et al.
Published: (2024)

APTx Neuron: A Unified Trainable Neuron Architecture Integrating Activation and Computation
by: Kumar, Ravin
Published: (2025)

MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
by: Qiu, Zihuan, et al.
Published: (2025)

Calibrated Decomposition of Aleatoric and Epistemic Uncertainty in Deep Features for Inference-Time Adaptation
by: Kumar, Divake, et al.
Published: (2025)

Disentangled Representation Learning with Transmitted Information Bottleneck
by: Dang, Zhuohang, et al.
Published: (2023)

Disentanglement-Based Equivariant Learning for Compositional VQA
by: Du, Zhou, et al.
Published: (2026)

Disentangled Representation Learning with the Gromov-Monge Gap
by: Uscidda, Théo, et al.
Published: (2024)