Neuron Identifiability
No mentions found
This entity hasn't been tracked yet, or Iris is still building its knowledge base.
Related Articles from SNS
Beyond Structural Symmetries: Linear Mode Connectivity via Neuron Identifiability
new Abstract: Many striking phenomena in deep learning, such as linear mode connectivity and the structured behavior of training dynamics, are closely tied to parameter symmetries: transformations that leave the realized function unchanged. Despite growing attention to parameter symmetries, the exact interplay between parameters, data, and representations remains underexplored. To investigate this, we develop a theoretical framework of effective function classes, i.e., the set of functions a...
Scientists reverse anxiety by fixing a tiny brain circuit
Scientists reverse anxiety by fixing a tiny brain circuit - Date: - June 3, 2026 - Source: - Universidad Miguel Hernandez de Elche - Summary: - A newly identified group of amygdala neurons appears to play a central role in anxiety and social behavior. Restoring normal activity in this tiny brain circuit reversed anxiety and social deficits in mice, revealing a promising new target for future treatments. - Share: Scientists have identified a specific brain circuit that appears to play a major...
A Self-Priming Neural Chain Links Sequential Behaviors Across Timescales
Behavioral sequences are essential for survival, yet the neural mechanisms that link one action to the next remain incompletely understood. In classical chain models, sequential behaviors arise through feedforward propagation of activity across distinct neuronal populations or network modules. Here, we identify a distinct form of neural chain mechanism in which neurons active during a first behavior modulate themselves into a persistent state of elevated tonic firing that subsequently drives...
Neural Network Compression by Approximate Differential Equivalence
Announce Type: new Abstract: Neural network compression is commonly achieved by pruning parameters based on local importance scores, e.g., magnitude-based pruning. We propose a complementary approach that compresses models by aggregating neurons with similar functional behavior rather than removing weights independently. Our method encodes a trained network as a polynomial ODE system and applies a lumping method called Approximate Forward Differential Equivalence to identify neurons with...
Robust Learning of a Group DRO Neuron
arXiv:2601.18115v2 Announce Type: replace Abstract: We study the problem of learning a single neuron under standard squared loss in the presence of arbitrary label noise and group-level distributional shifts, for a broad family of covariate distributions. Our goal is to identify a ''best-fit'' neuron parameterized by $\mathbf{w}_*$ that performs well under the most challenging reweighting of the groups. Specifically, we address a Group Distributionally Robust Optimization problem: given...
What is Missing? Explaining Neurons Activated by Absent Concepts
arXiv:2603.09787v2 Announce Type: replace Abstract: Explainable artificial intelligence (XAI) aims to provide human-interpretable insights into the behavior of deep neural networks (DNNs), typically by estimating a simplified causal structure of the model. In existing work, this causal structure often includes relationships where the presence of a concept is associated with a strong activation of a neuron. For example, attribution methods primarily identify input pixels that contribute most...
Social Novelty Recruits a Dysfunctional Nucleus Accumbens Ensemble That Drives Social Avoidance in a Shank3-/- Autism Model
Social behavior deficits are a common symptom of neuropsychiatric disorders, including autism spectrum disorder (ASD), but there are limited pharmacological treatments for these symptoms. Understanding how neurons encode social information will give insight into identifying novel pharmacological targets to address this unmet need. SHANK3 encodes a postsynaptic scaffold protein and is a common risk gene for several neuropsychiatric disorders characterized by social deficits, including ASD.
GoodVibe: Security-by-Vibe for LLM-Based Code Generation
arXiv:2602.10778v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used for code generation in fast, informal development workflows, often referred to as vibe coding, where speed and convenience are prioritized, and security requirements are rarely made explicit. In this setting, models frequently produce functionally correct but insecure code, creating a growing security risk. Existing approaches to improving code security rely on full-parameter fine-tuning or...
Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models
Announce Type: new Abstract: Language models (LMs) can produce gendered language and stereotypes even when given neutral prompts. Most prior work on gender bias in LMs primarily examines gender through a binary lens (feminine vs. masculine), with limited attention to gender-neutral forms, such as they/them pronouns or neutrally phrased job titles. How gender-related signals are encoded in the internal representations of LMs remains an open question.
RAIGen: Rare Attribute Identification in Text-to-Image Generative Models
arXiv:2602.06806v2 Announce Type: replace Abstract: Text-to-image diffusion models achieve impressive generation quality but inherit and amplify training-data biases, skewing coverage of semantic attributes. Prior work addresses this in two ways. Closed-set approaches mitigate biases in predefined fairness categories (e.g., gender, race), assuming socially salient minority attributes are known a priori.