Neurons

Follow this topic Ask about this topic

Momentum

22 papers in the last four weeks, up 175% on the four weeks before. 0.3% of all new papers.

Jul 6Week of Sep 21

Latest papers 134

Jun 30, 2026cs.CL

Readable but Not Controllable: Neuron-Level Evidence for Medical LLM Hallucination

Hallucination remains one of the central obstacles to deploying medical LLMs. Yet, even when hallucination can be detected, it is still unclear whether the internal representations associated with it can be used for control rather than detection alone. Using four open-source models across a suite of medical question-answering datasets, we show that a simple, carefully conditioned probe can reliably detect hallucination, with AUROC scores between 0.77 and 0.86 in our case. We further show that this signal is distributed and redundant rather than narrowly localized. Systematically selected neurons outperform random neurons only at very small subset sizes, whereas random subsets of a few hundred neurons recover nearly the full signal, and low-dimensional random projections preserve most of the detection performance. Beyond detection, we test whether this representation is causally actionable. Across 16 model--dataset combinations, our results reveal a sharp gap between decodability and controllability. The same internal structure that makes hallucination easy to detect does not translate into reliable neuron-level control. These findings show that medical hallucination seems to be readily visible in internal activations, but not easily corrected by steering the neurons most associated with it. More broadly, our results suggest that hallucination mitigation is not simply a matter of identifying the right neurons, and point to a deeper separation between what representations reveal and what they allow us to change.

Vijay Vankadaru, Asha Matthews, Tanya Roosta +1

School of Information University of California, Berkeley California, USA

Hallucination Mitigation Neurons

Jun 30, 2026cs.CV

Seeing Through Multiple Views: Parameter-Efficient Fine-Tuning via Selective Neurons for Consistent Radiology Report Generation

Recent years have seen substantial advances in radiology report generation (RRG), yet existing approaches predominantly adopt direct feature fusion when handling multi-view X-ray images. Such approaches overlook the potential clinical inconsistencies and inaccuracies arising when a single model processes different views, adversely impacting performance and clinical reliability. To this end, we introduce View-PNDF (View-specific Pattern Neuron Detection and Fine-tuning), a parameter-efficient framework that fosters view-consistent report generation from a neuronal perspective. Specifically, View-PNDF comprises: (i) a view-specific neuron detection module identifying neurons responsive to particular views, (ii) a verification module quantifying the existence of these neurons, and (iii) a selective fine-tuning strategy strengthening detected neurons while preserving view-agnostic representations. By updating only view-specific neurons, View-PNDF achieves consistent diagnoses across different views with reduced computational costs. Subsequently, we employ Large Language Models (LLMs) to consolidate the view-specific reports into a complete radiology report. Furthermore, we use traditional Natural Language Generation (NLG) metrics-based assessment on integrated reports for baseline comparison and employ LLM-based assessment (e.g., GPT-4o) on view-specific reports to capture clinical significance. Extensive experiments on two medical RRG benchmarks demonstrate that View-PNDF substantially improves view-specific chest X-ray report generation quality while maintaining robust general-view performance.

Yucheng Chen, Jinjing Zhu, Yang Yu +7

MedVisAI Lab · Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore · Centre of AI in Medicine, Singapore +3

Radiology Report Generation Chest X-Ray Medical Report Generation Multimodal Clinical Data Neurons

Jun 28, 2026cs.CV

FiRe: Frequency Reparameterization as a Preconditioner for Periodic Implicit Neural Representations

Periodic Implicit Neural Representations (INRs) such as SIREN and FINER assign every neuron, the same global frequency, spending the representational budget inefficiently when local signal content varies. We introduce FiRe (Frequency Reparameterization), that accelerates optimization by reparameterizing per-neuron frequency of periodic INRs without changing their underlying activation function. FiRe gives each neuron a bounded, input-dependent frequency via a separate low-rank gating path and is applicable to any periodic activation function. The gate acts as an implicit preconditioner that improves optimization conditioning at initialization via the Neural Tangent Kernel (NTK). This better-conditioned initialization makes optimization converge faster, and the high-frequency content of the reconstruction tracks the target more closely at a fixed computational budget. On 2D image fitting, FiRe increases PSNR over a parameter-matched baseline (up to +1 dB at short training budgets), with gains that vary with resolution and diminish at full convergence. We characterize how performance depends on resolution, rank, and training budget, and give an NTK account that predicts these trends.

Harinandan Shukla, Rajarshi Verma, Jitin Singla

Department of Computer Science and Engineering, IIT Roorkee · Department of Biosciences and Bioengineering, IIT Roorkee

Neural Representations Reparameterization Spectral Preconditioning Neurons Preconditioner

Jun 26, 2026cs.CL

SHIFT: Gate-Modulated Activation Steering for Knowledge Conflict Mitigation in Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) enhances LLMs by incorporating external knowledge to support response generation. However, conflicts between retrieved context and parametric knowledge have emerged as a critical challenge in RAG systems. To mitigate such conflicts, numerous studies have attempted to identify and edit knowledge-related internal neurons, aiming to improve the ability of LLMs to rely on contextual evidence during generation. However, these neuron-level approaches may introduce unintended cascading effects that compromise the general capabilities of LLMs, as the modified neurons are often entangled with broader model behaviors and functionalities. In this paper, we introduce SHIFT, a novel framework that reformulates neuron-level modification as learnable gate modulation, allowing LLMs to adaptively regulate internal activations for knowledge conflict resolution. Technically, our SHIFT equips LLMs with a lightweight gate module and optimizes fewer than 0.01% trainable parameters while keeping the backbone model frozen. During generation, the gate module adjusts the model's internal representations to adaptively leverage contextual and parametric knowledge. Extensive experiments on six datasets validate the effectiveness of our SHIFT in comparison with various competing baselines. All datasets and code are available at https://github.com/OpenBMB/SHIFT.

Ruochang Li, Pengcheng Huang, Zhenghao Liu +5

♡School of Computer Science and Engineering, Northeastern University, Shenyang, China · ♠Department of Computer Science and Technology, Tsinghua University, Beijing, China

Agentic Retrieval-Augmented Generation Systems Linear Activation Steering Neurons

Jun 24, 2026cs.LG

Stagnant Neuron: Towards Understanding the Plasticity Loss in Multi-Agent Reinforcement Learning Value Factorization Methods

Multi-Agent Reinforcement Learning (MARL) value factorization methods can suffer from a loss of plasticity, gradually failing to adapt when transferring to new task instances. We trace this issue to stagnant neurons, units whose gradient updates become negligibly small relative to their weights, thereby hindering learning. While existing plasticity injection methods exist, they prove ineffective for such neurons. To address this, we propose Knowledge-retentive Neuron-level PlastIcity Focusing InjEction (KNIFE), a novel method that directly targets stagnant neurons. KNIFE replaces each stagnant neuron with a composite unit comprising three specialized components: a frozen knowledge neuron to preserve acquired knowledge, a re-initialized active neuron to restore learning capacity, and a compensation neuron to ensure the combined output matches the original, thus maintaining previous learned cooperation knowledge. Extensive experiments on SMACv2, predator-prey, and matrix games demonstrate that KNIFE significantly outperforms state-of-the-art plasticity injection methods.

Zhengzhu Liu, Zeming Gao, Haoyuan Qin +7

Fujian Key Laboratory of Urban Intelligent Sensing and Computing, School of Informatics, Xiamen University (XMU), China · Key Laboratory of Multimedia Trusted Perception and Efficient Computing, XMU, China

Multi-Agent Reinforcement Learning Plasticity Synaptic Plasticity Neurons

Jun 22, 2026cs.ET

An Open-Source LFSR-Based Stochastic Leaky Integrate-and-Fire Neuron in SkyWater 130 nm: Design, Stochastic Characterisation, and Rate Coding

Stochastic spiking neurons trade exact arithmetic for controlled randomness, lowering area and tolerating input noise, which suits event-driven edge hardware. We present a compact, configurable stochastic leaky integrate-and-fire neuron in standard-cell CMOS on the SkyWater 130 nm process, released openly. A 16-bit configurable-polynomial linear-feedback shift register drives an eight-entry programmable activation table that sets a Bernoulli firing probability, and a saturating 16-bit leaky integrator with a programmable threshold and a refractory period of zero to seven cycles produces the spike train. All parameters are set through a sixteen-register serial interface, and the neuron runs from parallel inputs or entirely from the register file. From a model checked bit-exact against the register-transfer code, the period is 65535 states for a maximal-length polynomial and 63 for the shipped default, the eight-bit comparison value is uniform over the full period, and the per-entry firing probability equals the table value divided by 256. We also characterise a property a system-level model would not expose: the comparator output is serially correlated at short lags, with a negative lobe near lag eight, because the compared byte shifts by one bit each cycle; subsampling every sixteen cycles restores whiteness. Rate-coding sweeps show monotonic control of the output rate by the input weight and the threshold, and the refractory period caps the rate at one spike per refractory-plus-one cycles. The neuron occupies about 10,600 square micrometres at 70 per cent utilisation on a single Tiny Tapeout tile, meets 50 MHz timing with positive margin, and passes eighteen directed cocotb tests at register-transfer and gate level. All results are pre-silicon, from simulation and the open flow. The neuron is an openly released companion to a four-block neuromorphic suite reported separately.

Poornima Kumaresan, Santhosh Sivasubramani

Intrinsic Lab, Centre for Sensors, Instrumentation and Cyber-Physical System Engineering (SeNSE), Indian Institute of Technology Delhi, New Delhi 110016, India

Lfsr-Based Stochastic Leaky Integrate-And-Fire Neuron Neuromorphic Computing Neurons Stochastic Stochastic Dynamics Printed Circuit Board

Jun 19, 2026cs.NE

Soliton-like Waves in a Two-Dimensional Recurrent Spiking Neural Network with Weighted Spike-Timing-Dependent Plasticity

We construct a minimal but biologically plausible spiking neuron model operating in discrete time, combining multiplicative spike-timing-dependent plasticity (WSTDP), divisive normalization of synaptic integration, homeostatic threshold adaptation, and a one-step refractory period. We show that this normalization admits a biologically plausible dendritic implementation in which each binary junction operates using only locally available information. Assembling excitatory-inhibitory pairs of such neurons into a two-dimensional recurrent network and applying periodic localized stimulation, we find that the network spontaneously gives rise to stable, self-propagating wave packets with the properties of dissipative solitons: they maintain a stable spatial profile, propagate at constant speed, and annihilate upon frontal collision. Their emergence requires a geometric asymmetry between excitatory and inhibitory connection radii, and initial inhibitory synapses stronger than excitatory ones. WSTDP engraves the direction of propagation into the synaptic weight profile, so that the network learns by itself to sustain propagation in one direction while suppressing the reverse. When two sources are active simultaneously, the resulting waves annihilate upon collision, defining a semi-persistent boundary whose position encodes the relative phase and frequency of the two sources. These results provide a minimal computational framework for studying the emergence of cortical traveling waves, activity zone delimitation, and spatial memory from local plasticity rules alone.

Ch. Meessen

Independent researcher, computational neuroscience, France

Synaptic Plasticity Spiking Neural Networks Neurons Oscillatory Dynamics Neural Network Chaos

Jun 18, 2026cs.CL

From Texts to Scores: Tracing the Emergence of Essay Quality Representations in Large Language Models

Recent advances in Large Language Models (LLMs) have substantially transformed Automated Essay Scoring (AES), yet the internal mechanisms underlying LLM-based scoring remain poorly understood. In this work, we systematically analyze the hidden representations of eight LLMs across two English essay datasets (ASAP++, CSEE) and one Portuguese dataset (ENEM). Using linear probing, cross-prompt generalization, dimensionality reduction, and neuron-level analyses, we find consistent evidence that essay quality information is encoded in a linearly accessible form within LLM representations. These representations emerge progressively across layers, remain robust across prompting strategies, and partially transfer across essay prompts despite differences in scoring rubrics. In addition, nonlinear probes provide only marginal and inconsistent improvements over linear probes, suggesting that most essay quality information is already linearly decodable. We further identify individual ``essay scoring neurons'' whose activations strongly correlate with essay scores and whose behavior is sensitive to targeted intervention. Moreover, the layer-wise distribution of these neurons systematically shifts with essay length, with longer essays relying more heavily on deeper layers. Overall, our findings provide evidence that LLMs encode structured representations related to essay quality and offer new insights into the interpretability of LLM-based AES systems.

Jiaxu Zuo, Mu You, Kaixin Lan +5

NLP2CT Lab, Department of Computer and Information Science, University of Macau · Institute of International Language Services Studies, Macau Millennium College · School of Data Science and Information Engineering, Guizhou Minzu University

Automated Essay Scoring Large Language Model Evaluation Textual Dataset Text Analysis and Detection Neurons Emergence

Jun 17, 2026cs.CL

Output Vector Editing for Memorization Mitigation in Large Language Models

Large language models memorize and reproduce sequences from their training data, creating privacy, copyright, and security risks. Existing neuron-level mitigation methods equate editing with zeroing out neuron activations, but the activation only controls whether a neuron engages; the output vector is what writes to the residual stream and, through superposition, encodes multiple features. We propose output vector editing, a constrained-optimization weight edit that locates a small set of MLP neurons responsible for a memorized continuation and minimally modifies their output vectors to introduce a distractor in vocabulary space, redirecting their residual-stream contributions while leaving activations unchanged. Evaluating on four models from 360M to 7B parameters (SmolLM-360M, OLMo-1B, OLMo-7B, Llama2-7B), we center on OLMo-7B (whose open weights and pretraining corpus enable systematic mining) and mine 6831 memorized sequences, achieving up to 87.9% suppression. The 2.7

\times

gap over zero ablation on the same located neurons shows the suppression comes from the output-vector edit, not localization alone. Four edit modes span a spectrum from aggressive suppression to minimal redirection; in ensemble they cover 96.5% of memorized sequences, while our recommended single-mode configuration reaches 81.5% with no catastrophic locality failures. We further identify a mechanistic boundary at

{\sim}14%

of sequences unreachable by MLP-only editing; while these failures are not attention-driven overall, ablating the top contributing attention heads recovers 60--64% of them, with stronger recovery on continuations that copy tokens from the prefix, positioning attention as a complementary fallback rather than a primary mechanism. Edit mode ordering and the success-locality trade-off transfer across all four models, with success rates scaling with model size rather than family.

Ahmad Dawar Hakimi, Kaiwei Lei, Isabelle Augenstein +1

Center for Information and Language Processing, LMU Munich, Germany · Department of Computer Science, University of Copenhagen, Denmark · 3Munich Center for Machine Learning, Germany +1

Large Language Model Memory Memorization Residual Stream Neurons

Jun 16, 2026cs.LG

Effects of sparsity and superposition on loss in simple autoencoders

One of the major difficulties in the mechanistic interpretability of neural networks is the occurrence of polysemanticity, which suggests that each neuron is typically responsible for multiple different tasks, impeding a clean interpretation of their function. The seminal paper of Elhage et al. (2022) argues that this occurs due to superposition, a phenomenon where the neural network represents distinct features as non-orthogonal directions in a lower-dimensional space, a strategy that allows much greater compression of the data without sacrificing fidelity due to the feature sparsity of input vectors. Elhage et al. (2022) empirically validates these hypotheses in a rather natural and simple autoencoder with sparse inputs. The contribution of the present work is to analyze the mathematical basis for the occurrence and optimality of superposition, while rigorously corroborating some of their findings. In particular, we provide upper and lower bounds for the L2 reconstruction loss, tight in the very sparse regime, for power activation functions. A short list of interesting open problems are also included at the end.

Mriganka Basu Roy Chowdhury, Eric McLaughlin Weiner

Department of Statistics, UC Berkeley · Department of Materials Science, UC Berkeley

Improving Sparse Autoencoders Autoencoder Architectures Activation Functions Sparsity Neurons Mechanistic Interpretability+1

Jun 16, 2026cs.NE

An Optimization Framework for Automated Assessment of Biological Plausibility of Spiking Neurons

Biological plausibility is a key concept in neuromorphic computing and spiking neural networks, yet it remains inconsistently defined and difficult to quantify. In this work, we present an open-source framework for the automated assessment of biological plausibility in spiking neuron models. Our method builds on the idea of evaluating a model's ability to replicate canonical neuronal firing patterns observed in biological systems, following the classification proposed by Izhikevich. By encoding these patterns into objective functions and optimizing model parameters accordingly, our framework enables empirical assessment without requiring prior analytical modeling. Treating neuron models as black boxes, it provides a practical and flexible means of characterizing their dynamic capabilities. We demonstrate the effectiveness of the framework on several established models and a previously unexplored custom model. Implemented in Python and compatible with PyTorch and the Norse library, the framework is tailored for machine learning contexts. It is intended as a starting point for systematic research into the relationship between biological plausibility and network-level performance metrics such as accuracy, energy efficiency, robustness, and adaptability.

Sven Nitzsche, Alexandru Ionita, Andreas Faust +2

FZI Research Center for Information Technology, Karlsruhe, Germany · AI Multimedia Lab, CAMPUS Research Institute, National University of Science and Technology Politehnica Bucharest, Romania · Karlsruhe Institute of Technology, Karlsruhe, Germany

Spiking Neural Networks Neuromorphic Computing Neurons

Jun 15, 2026q-bio.NC

Learning Hybrid Biophysical Neuron Models with Neural ODEs

Biophysical neuron models link measurements of neural activity to underlying cellular mechanisms. Yet, a central challenge is that the kinetics of many ion channels are poorly characterized, and practical simplifications -- omitting channels or reducing morphological detail -- introduce systematic gaps between model and biology. Bridging these gaps requires approaches that can flexibly discover unmodeled dynamics while preserving mechanistic interpretability. Here, we introduce a hybrid modeling framework that embeds neural ordinary differential equations into conductance-based biophysical models to capture unknown currents or mis-specified channel kinetics. By parameterizing the neural ODE in terms of voltage-dependent steady-state and time-constant functions, we recover interpretable gating dynamics directly from voltage recordings without assuming a functional form. We show that the hybrid model fits the gating kinetics of 2400 ion channel models and recovers unknown gating dynamics from single current-clamp recordings, generalizing to out-of-distribution stimulus regimes under realistic inputs and parameter misspecification. We also use our method to reduce a multicompartment model of a cortical neuron into a single-compartment hybrid model with a learned axial current, yielding up to an order of magnitude lower computational cost. Together, our results establish a plug-and-play framework for selectively replacing unknown components of conductance-based models with neural ODEs while preserving their mechanistic structure.

Jonas Beck, Michael Deistler, Dóra Viktória Molnár +2

Hertie Institute for AI in Brain Health, Tübingen, Germany · Tübingen AI Center, University of Tübingen, Tübingen, Germany · Machine Learning in Science, Excellence Cluster Machine Learning, University of Tübingen, Tübingen, Germany +2

Neural Ordinary Differential Equations Neurons Discontinuous Gating Kinetics Voltage Channel Modeling

Jun 14, 2026cs.CL

Neuron Level Analysis of Large Language Model in Legal Domain Reasoning

We presented a neuron-level analysis of legal-domain reasoning in LLMs, comparing it with other applied domain tasks across seven open-weight models. Using neuron attribution scores to rank and suppress influential neurons, we confirmed that suppressing the identified neurons collapses accuracy on the target task, whereas suppressing the same number of random neurons does not. We further found a small subset of neurons influential across all seven tasks; once these are removed, suppressing the remaining neurons degrades only the task they were identified from, revealing genuinely task-specific neurons in every model studied. Within the legal domain, the three benchmarks exhibit relatively high neuron overlap and tend to be affected jointly, suggesting of legal components neurons that span jurisdictions. The distribution of identified neurons in our experiments suggests that the hypothesis that influential neurons are concentrated in middle MLP layers may depend on the input format and content, rather than being a universal phenomenon.

Eri Onami, Youmi Ma, Shuhei Kurita +1

Institute of Science Tokyo · NII · AIST

LLM Reasoning Strategies Neurons Multilayer Perceptrons

Jun 12, 2026cs.NE

A Programmer's Guide to Cascaded Adaptive Combiners: Online Learning by Biologically Accurate Models of Multilayer Neuron Networks

Learning in biological multilayer neuronal networks offers insights that extend beyond the classical weighted-sum neuron model commonly used in artificial neural networks. This article presents an accessible guide to a mechanistic neuronal network model that more accurately captures aspects of biological computation while enabling a simple yet powerful mechanism for learning in multilayer neural networks. The proposed approach supports efficient online streamed learning and provides a practical alternative to backpropagation. We demonstrate its potential in an image classification task, achieving competitive classification performance. The approach's simplicity, biological grounding, and broad applicability highlight a promising path toward algorithms that unify mechanistic neuron models and machine learning.

Martin Nilsson, Denis Kleyko

RISE Research Institutes of Sweden, Kista, Sweden · Örebro University, Örebro, Sweden

Neurons Backpropagation

Jun 11, 2026cs.LG

Natively Unlearnable Large Language Models

Unlearning aims to remove the influence of specific training data sources, but this has proved challenging because the contributions of different sources are entangled within the model. Isolating source contributions to disjoint parameters makes removal easier, though it obstructs joint learning across sources. We propose NULLs (Natively Unlearnable LLMs), a model class that satisfies the two opposing goals of isolating source-specific contributions and learning jointly across sources, by training a set of shared backbone neurons alongside a pool of sparsely activated sinks. During training, information specific to a source naturally concentrates in its sinks while information shared across sources accumulates in the backbone. A source is then unlearned at deployment by disabling its corresponding sinks, with no gradient updates and no access to the retained data. We show that NULLs scales to Wikipedia's ~6M articles, isolating each as an independent source. Unlearning a single article removes knowledge specific to it while preserving facts shared with semantically related articles, closely matching retraining from scratch. We note that unlearning with NULLs is also robust: in a case study of unlearning the Harry Potter books, NULLs resists both adversarial extraction and relearning that reverses post-hoc unlearning. Finally, NULLs preserves general language capabilities, matching a standard transformer on downstream benchmarks. Together, these results suggest that source-level unlearning need not be an afterthought. It can be built natively into LLM training while retaining the benefits of shared representation learning.

Gaurav R. Ghosal, Pratyush Maini, Aditi Raghunathan

Carnegie Mellon University · DatologyAI

Large Language Model Unlearning Exact Unlearning Neurons

Jun 11, 2026cs.AR

ReSCom: A Reconfigurable Spiking Neural Network Accelerator Using Stochastic Computing

Spiking Neural Networks (SNNs) provide an attractive framework for energy-efficient inference due to their event-driven computation and biologically inspired dynamics. However, efficient hardware realization of SNNs remains challenging because neuronal computations incur significant power and area costs, and uncontrolled approximate arithmetic can destabilize recurrent state updates when precision is not properly managed. To address these challenges, this paper presents ReSCom, a reconfigurable SNN accelerator that leverages stochastic computing to reduce hardware complexity while maintaining stable inference. The proposed architecture employs stochastic arithmetic for multiplication operations in neuron dynamics, while preserving exact fixed-point addition/subtraction operations. This stochastic strategy enables runtime trade-offs between accuracy, latency, and energy consumption. A unified reconfigurable neuron design supports Integrate-and-Fire (IF), Leaky Integrate-and-Fire (LIF), and Synaptic neuron models within a single hardware framework. Experimental results for MNIST inference on a Xilinx Artix-7 FPGA show that ReSCom achieves

92.80\%

classification accuracy while consuming just

0.05~\mathrm{mJ}

of operational energy per image at

100~\mathrm{MHz}

, outperforming the energy efficiency of recent state-of-the-art implementations. Furthermore, managing the stochastic bit-stream length allows explicit, dynamic control over accuracy-latency-energy trade-offs to meet target application constraints.

Ali Alipour Fereidani, Mohammad Rasoul Roshanshah, Saeed Safari

High-Performance Embedded Architecture Laboratory (HiPEAL) School of Electrical and Computer Engineering College of Engineering, University of Tehran Tehran, Iran · High-Performance Embedded Architecture Laboratory (HiPEAL) School of Electrical and Computer Engineering[cs.AR] College of Engineering, University of Tehran Tehran, Iran

Spiking Neural Networks Neuromorphic Computing Neural Network Inference Neurons

Jun 11, 2026cs.NE

Adaptive-Frequency Resonate-and-Fire Neurons for Spectral Estimation of Streaming Radar Signals

Frequency Modulated Continuous Wave (FMCW) radar systems traditionally rely on Fourier-based methods, such as the Fast Fourier Transform (FFT), to estimate target range and velocity. While computationally efficient, these approaches require storing and processing large blocks of data, which can become a bottleneck in memory-constrained or low-latency applications. In this work, we propose a neuromorphic-inspired signal processing method based on adaptive resonate-and-fire (ARF) neurons formulated as a discrete-time dynamical system. Each neuron dynamically adjusts its internal frequency to match dominant frequency components of the input radar signal, enabling direct estimation of target ranges and velocities without computing the full frequency spectrum. The proposed model operates in a sample-by-sample fashion, resulting in memory requirements that scale with the number of tracked targets rather than the signal length. A feedback mechanism is also introduced to enable multiple neurons to lock on distinct frequency components in multi-target cases. Results on simulated and experimental data demonstrate that the method can successfully track multiple targets. Compared to conventional FFT-based approaches, the proposed method offers reduced memory usage proportional only to the number of tracked targets, making it suitable for resource-constrained and edge-based radar applications.

Stefano Chiavazza, Sen Yuan, Marc Geilen +2

Electronic Systems, Eindhoven University of Technology, The Netherlands · MS3 Group, Delft University of Technology, The Netherlands

Radar Neuromorphic Computing Fourier Lfsr-Based Stochastic Leaky Integrate-And-Fire Neuron Neurons

Jun 11, 2026cs.LG

Understanding helpfulness and harmless tension in reward models

Reward models are a key component of reinforcement learning from human feedback (RLHF), aligning language models toward both helpful and harmless behaviour. However, the internal mechanisms underlying these objectives and their conflicts remain poorly understood. We study alignment tension in reward models trained under helpfulness-only, harmlessness-only, and mixed-objective settings. We find that mixed-objective models often underperform single-objective models, indicating interference between objectives. Using activation-based methods, we identify neurons associated with each objective and study their functional roles via targeted ablations. We find that these neurons causally support their corresponding objectives while often negatively affecting the opposing one. We find that a substantial proportion of neurons are shared between helpfulness and harmlessness, and that these shared neurons exert a disproportionate influence on model behaviour, contributing to alignment tension. Additionally, our results provide insights and mechanistic interpretation into how alignment objectives are represented in reward models and why multi-objective alignment remains challenging, motivating future work on disentangled and controllable alignment methods.

Eshaan Tanwar, Pepa Atanasova

University of Copenhagen

Progress Reward Modeling Value Alignment Neurons Assistant

Jun 9, 2026nlin.AO

Towards Critical Branching Mechanism in Recurrent Neural Networks

Criticality has been proposed as a key organizing principle in biological neural systems, yet its origin and relevance in artificial neural networks remain unclear. We analyze hidden-state dynamics in trained long short-term memory (LSTM) networks and show that small networks near their optimal training epochs (steps) exhibit scale-free avalanche statistics and branching parameters close to unity, indicative of near-critical dynamics, while larger models remain subcritical. To explain the coexistence of subcritical branching with robust

1/f^β

noise, we introduce a mixture branching process framework that links heterogeneous branching dynamics to long-range temporal correlations. These results identify critical-like behavior in LSTMs as an emergent, capacity-dependent dynamical regime.

Feixiang Ren, Ling Feng

Department of Physics, National University of Singapore Singapore · Institute of High Performance Computing Agency for Science, Technology and Research (A*STAR) Singapore

Recurrent Neural Networks Neurons Neural Network Hidden States

Jun 7, 2026cs.LG

When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff

Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has become a standard pipeline for Large Language Model (LLM) post-training. SFT is expected to provide a useful behavioral prior for RL to further enhance model capabilities. However, checkpoints with excessive SFT often show limited improvement during RL. We attribute this failure to the loss of model plasticity: the reduced ability of an SFT-initialized policy to be effectively reshaped by subsequent RL. To better understand this phenomenon, we conduct detailed analysis from multiple perspectives, including parameter changes, output spaces, and RL optimization dynamics. Our results show that models from excessive SFT tend to produce over-confident token distributions and exhibit sharp parameter landscapes, which make them harder to optimize in the RL stage. To enable a more robust SFT-to-RL handoff, we propose \texttt{Rejuvenation}, a simple yet effective method that restores plasticity while preserving useful SFT-acquired priors. Rejuvenation leverages base-anchored model fusion to reduce excessive SFT-induced drift with targeted neuron reset to mitigate model rigidity. Experimental results on both math reasoning tasks and agentic tasks demonstrate that our approach consistently improves RL performance on over-trained SFT models, while also enhancing generalization to out-of-distribution tasks.

Runze Liu, Jiashun Liu, Xu Wan +2

Hong Kong University of Science and Technology · Zhejiang University · State Key Laboratory of Multimodal Artificial Intelligence Systems, CASIA

Plasticity Neurons

Jun 5, 2026physics.optics

Beyond the Thin-Layer Limit: Differentiable Volumetric Training for Visible-Range Diffractive Neural Networks

Diffractive deep neural networks (D2NNs) promise miniaturized, power-efficient, light-speed optical front-ends for machine vision, yet the most mature demonstrations remain in the terahertz regime, built from readily fabricated millimeter-scale neurons. Translating D2NNs to the visible range, where nearly all vision pipelines operate, was long blamed on the difficulty of fabricating nanoscale neurons; but even after recent advances removed that barrier, visible-range D2NNs matching their terahertz counterparts remain out of reach. We identify the true obstacle as the thin-layer approximation underlying nearly all D2NN training, which treats each diffractive layer as an infinitely thin mask. It fails not because of the short wavelength, as is commonly assumed, but because the low-refractive-index materials (n approximately 1.3-1.5) used at visible wavelengths require relief structures thick enough that intra-layer diffraction and phase accumulation become significant. To overcome this, we introduce a differentiable beam-propagation (

\partial

BPM) layer that models each element as a finite-thickness volume and propagates light through it during training, keeping the fabrication-compatible height map end-to-end trainable without full-wave simulation in the loop. Across MNIST, Fashion-MNIST, and CIFAR-100 classification and imaging,

\partial

BPM training substantially reduces the design-to-device mismatch, and full-wave FDTD validation raises classification accuracy from 50% to 90% without re-optimization. The

\partial

BPM layer thus offers a scalable, physics-aware bridge between efficient optical neural-network optimization and fabrication-consistent diffractive design.

Dineth Jayakody, Dushan N. Wadduwage

Department of Computer Science, Old Dominion University, Norfolk, VA 23529, USA · School of Data Science, Old Dominion University, Norfolk, VA 23529, USA · Department of Physics, Old Dominion University, Norfolk, VA 23529, USA

Modern Deep Networks Differentiable Optimization Neural Network Optimization Neurons

Jun 3, 2026cs.LG

Beyond Structural Symmetries: Linear Mode Connectivity via Neuron Identifiability

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 neuron can realize on its input support, and the norm cost of realizing them. We then formalize effective symmetry breaking via neuron identifiability across independent training runs. Our analysis shows that neural networks can admit large families of approximately equivalent solutions even in structurally asymmetric models. We further show that neuron identifiability enables representation merging without prior alignment, and characterize when such merging admits a linear low-loss path. These findings highlight the role of effective function classes in affecting the loss landscape.

Vincent Bürgin, Daniel Herbst, Ya-Wei Eileen Lin +1

Technical University of Munich; School of CIT, MCML, MDSI · MIT; Dept. of EECS, CSAIL

Neurons Symmetry Deep Learning Representational Capacity

Jun 3, 2026cs.LG

TENP: Trapezoidal Expert Neuron Pruning For Mixture-of-Experts

Mixture-of-Experts large language models (LLMs) scale efficiently through sparse activation, yet their deployment is fundamentally constrained by the large static parameter footprint of experts. Existing compression approaches either remove entire experts, disrupting routing topology and harming performance, or rely on unstructured weight pruning with limited practical efficiency. To address the limitations, we propose TENP, a structured Trapezoidal ExpertNeuron Pruning framework. Using a few samples, we identify and retain important experts, while applying expert neuron pruning (ENP) to less important experts, reserving model parameters in a trapezoidal pattern from shallow to deep layers. When evaluating expert importance, we jointly consider both the magnitude of the expert output and its ability to change the direction of the input vector. For ENP, we measure each neuron's projected contribution to the expert output to identify and retain important neurons. We conduct extensive experiments on the Qwen and DeepSeek models. Under a routing expert sparsity of 40% and an average of 63.76% activated expert parameters, the DeepSeek model suffers only a 1-point drop in accuracy compared to the full-parameter model. Moreover, it outperforms the full-parameter model by 10% on code generation tasks.

Jiangyang He, Shaolin Zhu, Deyi Xiong

TJUNLP Lab, School of Computer Science and Technology, Tianjin University, China

Mixture-Of-Experts Large Language Models Large Language Model Compression Mixture-Of-Experts Deepseek Neurons

Jun 3, 2026cs.LG

When Both Layers Learn: Training Dynamics of Representing Linear Models via ReLU Networks

In this paper, we study the gradient descent dynamics for jointly training both layers of a one-hidden-layer ReLU network to fit a linear target function. Concretely, we consider a realizable setting where inputs are drawn i.i.d. from a Gaussian distribution and labels follow a planted linear model. This stylized framework captures salient features of end-to-end training in inverse problems and certain auto-encoder models. Despite its apparent simplicity, the dynamics remain poorly understood, in part because the loss landscape contains multiple non-strict saddle points, making it unclear why gradient descent from random initialization reliably escapes bad stationary regions. We provide a detailed characterization of the optimization landscape and prove that gradient descent from a moderately small random initialization-simultaneously training both layers-converges to a global minimizer at a linear rate with order-wise optimal sample complexity. Our analysis tracks the trajectory through three phases: an alignment phase in which hidden weights progressively align with the planted direction while the output weights maintain the correct sign pattern; a growth phase in which the norms of both layers increase while preserving alignment; and a local refinement phase in which the aligned neurons rapidly converge to the planted direction, yielding fast local convergence. To rigorously show that GD avoids non-strict saddles, we develop trajectory-level control arguments for the end-to-end dynamics. In addition, we establish novel uniform concentration results that hold along the entire trajectory, and are essential for obtaining order-wise optimal sample complexity. We corroborate our theory with extensive experiments across a range of configurations.

Berk Tinaz, Changzhi Xie, Mahdi Soltanolkotabi

Department of Electrical and Computer Engineering University of Southern California Los Angeles, CA, USA · Department of Computer Science University of Southern California Los Angeles, CA, USA

Rectified Linear Unit Networks Gradient Descent Stationary Point Neurons

Jun 3, 2026cs.LG

A Geometric Characterization of the Stationary Plateau for Two-Layer Neural Networks

We investigate the geometric structure of stationary plateaus that arise in the loss landscape of two-layer neural networks with smooth activation functions. We focus on the phenomenon of "neuron splitting" where duplicating a hidden neuron yields an affine set of stationary points in a wider network. We provide a comprehensive classification of all stationary points on these plateaus, determining under what conditions they constitute local minima or saddle points. Our characterization hinges on a per-neuron curvature object we term the "inner Hessian" matrix. Our analysis reveals that the definiteness of the inner Hessian and the choice of splitting coefficients jointly dictate the local geometry of the plateau. We show that "splitting" a local minimum can yield either a mixture of local minima and saddles or an all-saddle plateau, with a concrete sure-saddle region identified under mild assumptions. In contrast, splitting a saddle point always produces a plateau of saddle points. Our results unify and extend prior landscape analyses, elucidating when and how model expansion preserves or alters the nature of stationary points. These findings offer new geometric insights into the effects of width expansion and reparameterization in neural networks.

Tian Ding, Dawei Li, Ruoyu Sun

Shenzhen International Center of Industrial and Applied Mathematics, Shenzhen Research Institute of Big Data, China; Shenzhen Loop Area Institute, China; AutoKernel, China. · Shenzhen Loop Area Institute, China · AutoKernel, China +3

Two-Layer Neural Networks Anisotropic Loss Landscapes Hessian Neural Network Flat Minima Neurons

Jun 2, 2026cs.LG

Smart Transportation Without Neurons -- Fair Metro Network Expansion with Tabular Reinforcement Learning

We tackle the Metro Network Expansion Problem (MNEP), a subset of the Transport Network Design Problem (TNDP), which focuses on expanding metro systems to satisfy travel demand. Traditional methods rely on exact and heuristic approaches that require expert-defined constraints to reduce the search space. Recently, deep reinforcement learning (Deep RL) has emerged due to its effectiveness in complex sequential decision-making processes-it remains, however, computationally expensive, environmentally costly, and requires additional engineering to interpret. We show that MNEP problems are small enough to not require Deep RL methods. Reformulating the MNEP as a Non-Markovian Rewards Decision Process (NMRDP), we use tabular RL to achieve similar performance with significantly fewer training episodes, additionally offering greater interpretability. Additionally, we incorporate social equity criteria into the reward functions, focusing on efficiency and fairness, highlighting the versatility of our method. Evaluated in real-world settings-Xi'an and Amsterdam-our method reduces total episodes by a factor of 18 and total carbon emissions by a factor of 12 on average, while remaining competitive with Deep RL. This approach offers a replicable, modular, interpretable, and resource-efficient solution with potential applications to other combinatorial optimization problems.

Dimitris Michailidis, Sennay Ghebreab, Fernando P. Santos

Socially Intelligent Artificial Systems University of Amsterdam

Bringing Network Coding Neural Combinatorial Optimization Sequential Decision Making Algorithmic Fairness Neurons

Jun 2, 2026cs.LG

Neuron Populations Exhibit Divergent Selectivity with Scale

We investigate whether neuron populations within neural networks evolve predictably with scale, extending scaling laws beyond macroscopic observables such as loss. To probe this question, we study Rosetta Neurons, a previously characterized class of neurons whose activation patterns are similar across independently trained models (Dravid et al., 2023). In separate analyses of language models up to 30B parameters and vision models up to 5B parameters, we observe that the population of Rosetta Neurons follows a sublinear power law in model size, growing in absolute number but occupying a shrinking fraction of the total neuron count. We further observe a Neuron Polarization Effect: Rosetta Neurons become more selective and increasingly monosemantic with scale, separating from a growing non-Rosetta population that remains less selective. An analytical model balancing feature utility against limited neuron capacity explains the sublinear power-law scaling and this polarization effect. Finally, we find that Rosetta Neurons become more domain-specialized with scale and illustrate their selectivity through a targeted data-filtering case study for continued pretraining. Our results point to a scaling law for interpretable, shared neuron-level structure, linking model size to systematic changes in neuron universality, selectivity, and specialization.

Amil Dravid, Yasaman Bahri, Alexei A. Efros +1

UC Berkeley · TTIC

Neurons Scaling Laws Model Size Selectivity Two-Layer Neural Networks

Jun 2, 2026cs.NE

Quadratic integrate-and-fire neurons exhibit less fragmented loss landscapes and outperform leaky integrate-and-fire neurons in spike-based gradient descent

The ability to train spiking neural networks is essential for modeling biological neural networks as well as for neuromorphic computing. However, for the extensively used leaky integrate-and-fire (LIF) neurons, arbitrarily small parameter changes can induce spike (dis)appearances that disrupt subsequent activity, leading to unstable neural representations and permanently silent neurons during exact spike-based gradient descent. Recent work shows that a class of neuron models, which includes the quadratic integrate-and-fire (QIF) neuron, avoids these discontinuities and enables continuous and even smooth spike-based gradient descent. However, it remains unclear whether these advantages translate into practice. Here, we demonstrate that they do so via a controlled comparison between networks of LIF and QIF neurons on the popular Spiking Heidelberg Digits dataset. Specifically, in a first step, we perform a thorough hyperparameter search to optimize both models, revealing a clear performance advantage of QIF neurons. In a second step, we visualize the loss and gradient landscapes. Consistent with their inferior performance, we find that the loss landscapes of LIF neurons, which are discontinuous, appear more fragmented and the related gradients more erratic. An analysis of the landscapes of single samples indicates that these features arise from changes in the temporal order of spikes, which often cause disruptive spike (dis)appearances. Overall, our results advocate replacing LIF neurons with neuron models exhibiting continuous spiking dynamics, such as QIF neurons, for gradient descent training.

Carlo Wenig, Raoul-Martin Memmesheimer, Christian Klos

University of Bonn · University of Tübingen · University of Illinois Urbana-Champaign

Lfsr-Based Stochastic Leaky Integrate-And-Fire Neuron Neurons Anisotropic Loss Landscapes Neuromorphic Computing Gradient Descent Neural Representations+1

Jun 2, 2026cs.NE

Signed Spiking Neuron Enabled by an Orthogonal-Easy-Axis Magnetic Tunnel Junction

Signed spiking neurons carry richer information than standard spiking neurons. This work proposes a compact magnetic tunnel junction (MTJ)-based neuron for signed leaky integrate-and-fire (LIF) operation. With orthogonal easy axes in the free and pinned layers, the device enables bipolar spike generation and maps magnetic-moment dynamics to signed LIF membrane-potential evolution. Landau--Lifshitz--Gilbert simulations show that proper free-layer dimensions allow the device response to follow a signed LIF equation. A representative design of 10 nm x 45 nm x 50 nm corresponds to an aspect ratio of about 2:9:10. Network evaluations using the fitted device-neuron model achieve 91.06% on CIFAR-10 and 77.40% on CIFAR10-DVS, retaining most of the accuracy of ideal signed LIF neurons.

Huannan Zheng, Jingli Liu, Kezhou Yang

Function Hub, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, 51000, China

Lfsr-Based Stochastic Leaky Integrate-And-Fire Neuron Time-To-First-Spike Neurons Magnetic Tunnel Junction

May 29, 2026cs.LG

Structured Neuron Pruning in Deep Neural Networks Using Multi-Armed Bandits

Deep neural networks often contain redundant hidden units. Removing individual weights can reduce parameter count, but unstructured sparsity is not always easy to exploit in standard dense implementations. This paper develops a structured pruning framework in which complete neurons are removed using multi-armed bandit (MAB) algorithms. Each candidate neuron is treated as an arm; pulling an arm temporarily masks that neuron, measures the change in loss on a sampled mini-batch, restores the neuron, and updates an estimate of its safe-removal reward. The framework supports stochastic policies, including Epsilon-Greedy, Softmax, UCB1 and Thompson Sampling, and multiplicative-weight policies, including Hedge-style multiplicative weights and EXP3. We evaluate the method on tabular classification, tabular regression and deep neural-network benchmarks covering image, text and reasoning tasks. Statistical comparisons using the Friedman test followed by the Nemenyi post-hoc test show significant differences between methods. On tabular classification tasks, UCB1 obtains the highest mean rank among pruning policies and improves on the unpruned neural network. On regression tasks, UCB1 obtains the highest mean rank and is statistically competitive with, or superior to, several standard regression models according to R^2. On deep-learning tasks, UCB1 and Thompson Sampling obtain the strongest ranks, and several MAB policies significantly outperform the unpruned model, magnitude-based neuron pruning and greedy activation-variation pruning. The results show that MAB-based neuron pruning is an effective and computationally practical approach for structured model reduction.

Salem Ameen, Sunil Vadera

School of Science, Engineering and Environment, University of Salford, Salford, United Kingdom

Modern Deep Networks Unstructured Pruning Neurons Sparsity

May 29, 2026cs.CL

Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models

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. In this work, we study gender-specific neurons in LMs across three categories: feminine, masculine, and gender-neutral. We propose a neuron-level intervention method to identify neurons that are strongly tied to each gender category. We then test these neurons through controlled generation, showing that activating or masking gender-related neurons can steer a sentence toward a target gender form while preserving its original meaning. To evaluate the effectiveness of our gender-intervention approach, we curate two datasets with controlled sentences labeled across all three gender categories and validate the data quality through human evaluation. Experiments on two open-source LMs show that gender-specific neurons are not evenly distributed across model layers; instead, they concentrate heavily in the earliest layers with smaller contributions from later layers. Compared to existing methods, our method achieves more precise gender control, with less leakage into non-target gender categories and stable output quality through two evaluation criteria. Overall, our work examines how gender is encoded in LMs and provides a simple yet effective approach toward controlled gender intervention for both neuron intervention evaluation and gender bias mitigation. Code and datasets are available at: https://github.com/zhiwenyou103/Gender-Neuron-Intervention

Zhiwen You, Nafiseh Nikeghbal, Jana Diesner

University of Illinois Urbana-Champaign · Technical University of Munich · Munich Center for Machine Learning

Gender Bias Gender Neutrality Neurons

May 24, 2026cs.LG

Tiny Brains, Giant Impact: Uncovering the Keystone Neurons of LLM with Just a Few Prompts

Large language models (LLMs) display strong comprehensive abilities, yet the internal mechanisms that support these behaviors remain insufficiently understood. In this work, we show that across a wide range of open-weight Transformers, a subset of neurons remains consistently highly activated during inference across tasks of multiple capability dimensions. By probing along the cross-task activation strength, an extremely sparse subset is isolated, whose removal causes a collapse in model behavior, which we term keystone neurons. Our analysis reveals that keystone neurons are a stable and intrinsic neuron subset of the model that is largely established during pretraining. The parameters associated with these neurons are tightly calibrated during the training process, and their precise values are critical for the capabilities of the model. Building on these insights, we propose a supervised fine-tuning approach that updates only keystone neurons, achieving task gains comparable to or even better than full-parameter fine-tuning while better preserving performance in other capability dimensions, despite modifying a much smaller number of parameters.

Xiangtian Ji, Yuxin Chen, Zhengzhou Cai +3

National University of Singapore · Beijing University of Posts and Telecommunications · University of Science and Technology of China

Neurons Transformer Architectures Cross-Task Interference Prompt Engineering

May 23, 2026quant-ph

Fermi-Dirac machines as quantizations of neurons

Fermi-Dirac machines were proposed recently as an approach to solving semidefinite optimization problems on quantum computers. Here, we reinterpret them as canonical quantizations of classical neurons. By viewing a classical neuron as an activation function applied to a parameterized classical Hamiltonian, we quantize this model by replacing classical variables with operators whose eigenvalues encode their possible values. This follows the standard approach to canonical quantization in quantum mechanics. Crucially, when the Hamiltonian consists of commuting operators, our construction reduces exactly to a classical neuron. More generally, our approach yields an activation observable, defined as an activation function applied to a parameterized quantum Hamiltonian. The output of this quantized neuron is a random variable with expectation value equal to that of the activation observable with respect to an input state. We develop efficient hybrid quantum-classical algorithms for evaluating outputs and gradients of our quantized neurons, enabling evaluation and training. These algorithms rely on basic primitives that include random sampling, Hamiltonian simulation, and the Hadamard test. We also quantize a whole host of other activation functions, including the smooth rectified linear unit (ReLU), sigmoid linear unit, Gaussian-smoothed ReLU, and Gaussian error linear unit (GeLU), which are known to be useful for deep learning applications. Numerical experiments indicate that neurons based on quantum Hamiltonians can learn functions that classical neurons cannot. We further define a computational decision problem based on Fermi-Dirac neurons and prove that it is BQP-complete, providing complexity-theoretic evidence against efficient classical simulation. Finally, we generalize our approach to continuous quantum variables and sketch two different ways of composing these neurons into networks.

Alexander He, Nana Liu, Mark M. Wilde

Department of Physics, Cornell University, Ithaca, New York 14850, USA · Institute of Natural Sciences, Shanghai Jiao Tong University, Shanghai 200240, China · School of Mathematical Sciences, Shanghai Jiao Tong University, Shanghai 200240, China +3

Quantum Neural Networks Lfsr-Based Stochastic Leaky Integrate-And-Fire Neuron Neurons Quantized Activation Functions

May 21, 2026cs.NE

Rare Events, Real Signals: Functional Ensembles as Units of Computation in Deep Spiking Networks

We investigate how internal representations emerge across hierarchical processing systems by introducing a neuroscience-inspired framework for analyzing deep spiking neural networks (SNN) through the lens of functional connectivity. Drawing on concepts from systems neuroscience and information theory, we form the first-order functionally-connected (1FC) group of a neuron based on its statistically significant pairwise correlations with neurons from the previous layer of a trained SNN architecture. We then track its response properties during inference under various conditions. Our analysis shows that several principles of functional connectivity previously observed in biological cortex are preserved in spiking ResNet architectures. These 1FC ensembles display interesting properties: their aggregate cofiring reliably predicts downstream neuronal responses through a robust, ReLU-like input-output relationship, whose gain scales systematically with ensemble size. Reliable encoding of the presented class emerges only during high 1FC cofiring events, which themselves occur infrequently, indicating that informative representations are concentrated in rare but highly coordinated activity patterns. Under uniform random noise or adversarial perturbations, these response profiles are disrupted, particularly in early and intermediate layers. This enables a targeted high-resolution interrogation at specific nodes and pathways. We showed that the functional connectivity structure is shaped by learning and this structure breaks under weight permutation. These establish 1FC ensembles as a functionally meaningful substrate for input encoding and information transfer, with potential implications in designing targeted fine-grained diagnostics on the information flow.

Aditi Aravind, Konstantinos Ladakis, Mario Alexios Savaglio +2

University of Crete, Greece · 2Foundation of Research & Technology - Hellas (FORTH), Greece · 4Harvard Medical School, USA +2

Spiking Neural Networks Functional Connectivity Neural Circuits Neurons Ensemble Rare Events

May 20, 2026cs.AI

NeuroNL2LTL: A Neurosymbolic Framework for Natural Language Translation of Linear Temporal Logic

Effectively translating between natural language (NL) and formal logics like Linear Temporal Logic (LTL) requires expertise that limits formal verification's reach in safety-critical development. Template-based approaches sacrifice expressiveness for reliability; neural methods achieve fluency but provide no correctness guarantees. We present NeuroNL2LTL, a neurosymbolic architecture unifying learned translation with formal verification. NeuroNL2LTL routes translation through an intermediate representation whose mapping to LTL is structure-preserving by construction. Generated specifications undergo satisfiability and non-triviality checking; a minimal-edit repair mechanism corrects near-miss outputs before they reach downstream tools. The central innovation is verifier-in-the-loop training: verification outcomes serve as reward signals for reinforcement learning, producing neural components that optimize directly for formal correctness. On 200,000+ requirements spanning aerospace, robotics, autonomous vehicles, and ten additional domains, NeuroNL2LTL achieves 28% semantic equivalence with reference specifications while ensuring 86% of outputs are verified satisfiable. The system also generates contextually grounded explanations from LTL, enabling domain experts to validate specifications without specialized training. This work demonstrates that formal verification can function as both training objective and runtime filter for neural specification systems, allowing us to build neural-based tools whose reliability derives from logical guarantees rather than statistical confidence.

Paapa Kwesi Quansah, Ernest Bonnah

Baylor University, Waco TX 76706, USA

Linear Temporal Logics Formal Verification Neuro-Symbolic Framework Natural Language Neural Network Verification Neurons

May 19, 2026cs.CL

When Irregularity Helps: A Subclass Analysis of Inductive Bias in Neural Morphology

Neural morphological generation systems often achieve high aggregate accuracy on benchmark datasets, yet such performance can conceal systematic errors concentrated in rare morphological subclasses. We examine Japanese past-tense verb inflection and show that a very small, structurally specific irregular subtype (<1% of data) accounts for a disproportionate share of model errors. Controlled ablation experiments demonstrate that removing this subtype yields larger improvements in generalization than removing all irregular verbs, indicating that not all irregularity contributes equally to model instability. These findings suggest that error concentration is driven by the interaction between extreme low-frequency morphological patterns and specific morphophonological processes, particularly gemination. We argue that morphological evaluation should incorporate finer-grained subclass analysis beyond standard conjugation categories.

Wen Zhang

Morphology Stemming Mismatch Inductive Bias Neurons

May 19, 2026cs.CV

CPC-VAR:Continual Personalized and Compositional Generation in Visual Autoregressive Models

Visual autoregressive (VAR) models have recently emerged as an efficient paradigm for text-to-image generation. Despite their strong generative capability, existing VAR-based personalization methods remain limited to static settings, failing to accommodate evolving user demands. In particular, sequential concept learning leads to severe catastrophic forgetting, while multi-concept synthesis often suffers from feature entanglement and attribute inconsistency. In this work, we present the first systematic study of continual personalized generation in VAR models. We identify two key challenges: (i) preserving previously learned concepts during sequential customization, and (ii) composing multiple personalized concepts in a controllable manner. To address these issues, we propose a unified framework with two core components. For continual single-concept learning, we introduce Gradient-based Concept Neuron Selection (GCNS), which identifies concept-relevant neurons and constrains only conflicting parameters across tasks, effectively mitigating forgetting without additional model expansion. For multi-concept synthesis, we propose a context-aware composition strategy that performs multi-branch feature modeling and localized cross-attention fusion guided by spatial conditions, enabling precise and disentangled concept composition. Extensive experiments demonstrate that our method significantly improves performance in long-sequence continual personalization while achieving superior results in multi-concept image synthesis compared to existing baselines. These findings highlight the potential of VAR models for scalable and controllable personalized generation.

Junhao Li, Xinhao Zhong, Yi sun +4

Harbin Institute of Technology, Shenzhen · South China University of Technology · Peng Cheng Laboratory +1

Visual Autoregressive Models Personalization Ads Neurons

May 19, 2026cs.CV

Neuron Incidence Redistribution for Fairness in Medical Image Classification

Deep learning models for medical image classification are susceptible to subgroup performance disparities across demographic attributes such as age, gender, and race. We identify a latent representational mechanism underlying these disparities: in transfer-learned models, the dominant penultimate-layer activation channel under positive predictions is co-activated by both disease-positive samples and privileged demographic groups (male, older patients), producing over-diagnosis; conversely, the dominant channel under negative predictions is co-activated by disadvantaged groups (female, younger patients), producing systematic under-diagnosis. To address this, we propose Neuron Incidence Redistribution (NIR), a lightweight regularization method that penalizes the variance of predicted-probability-weighted mean activations across penultimate-layer neurons, requiring no demographic labels at training time. On HAM10000, TPR disparity drops from 10.81% to 0.93% across age groups and from 12.04% to 0.74% across gender, with a marginal AUC improvement of 0.51 points. On Harvard OCT-RNFL, NIR reduces FPR disparity for race (from 15.68% to 10.66%) and age (from 12.69% to 1.80%), demonstrating that distributing latent disease evidence across the full penultimate layer is a principled and effective strategy for improving demographic fairness in medical AI.

Abin Shoby, Lyle John Palmer, Nikhil Cherian Kurian

Australian Institute for Machine Learning · Adelaide University · Adelaide, SA 5000, Australia

Medical Image Classification Imbalanced Classification Algorithmic Fairness Subgroups Neurons Model Activations

May 19, 2026cs.NE

Updating the standard neuron model in artificial neural networks

From their inception in the 1950s, artificial neural networks (ANNs) started using the so-called point neuron model then prevalent in neuroscience, hoping that this analogy would allow for a better emulation of brain function. Over the years the neuroscience literature has shown that the point neuron model is too simplistic to properly represent many fundamental neural processes; however, the standard neuron model in ANNs still remains the same. Here we substitute it by a very recent model of cortical cells and demonstrate through theoretical analyses and experimental results how, simply by using a more realistic neural unit element without augmenting the number of parameters, the resulting ANNs offer a number of important advantages that include increases in expressivity, robustness and learning speed, and a reduction in memorization and the amount of training data needed.

Raul Mohedano, Thomas Batard, Erik Velasco-Salido +4

1Spanish National Research Council (CSIC), Madrid, Spain. · Center for Research in Mathematics (CIMAT), Guanajuato, Mexico. · 3Universidad Aut´onoma de Madrid (UAM), Madrid, Spain. +1

Neurons Neural Network

May 18, 2026cond-mat.dis-nn

Activation Functions, Statistics and Learning of Higher-Order Interactions in Restricted Boltzmann Machines

The great success of neural networks primarily arises from the presence of the large number of weight parameters combined with nonlinearities in the input-output relationship of single neurons. In this work, we study the relationship between the statistical properties of the weights and the nonlinearity of the hidden unit in Restricted Boltzmann Machines (RBMs) on the one side, and the distribution induced on binary visible units. We do this for four commonly used activation functions: Linear, Step, ReLU, and Exponential, and make qualitative predictions about the ability of these models to learn distributions with strong higher order interactions over the visible nodes. We show that in general, in an ensemble of RBMs with Gaussian weights, these distributions are rare and hard to learn, except when the hidden unit activation function is an Exponential.

Giovanni di Sarra, Yasser Roudi

Kavli Institute for Systems Neuroscience, Norwegian University of Science and Technology, 7491 Trondheim, Norway · Department of Mathematics, King’s College London, London, Strand, WC2R 2LS, UK

Activation Functions Neurons Nonlinearity Gaussian Primitives

May 17, 2026q-bio.NC

Von Economo neurons enable reliable social skill acquisition in recurrent spiking neural networks: a computational account with clinical predictions

Von Economo neurons (VENs) are selectively lost in behavioural-variant frontotemporal dementia (bvFTD) and reduced in autism spectrum conditions (ASC), yet their computational role in social learning remains unexplained. We train a spiking neural network (the VENCircuit) embedding VEN-like projection neurons (K=40, 2% of total) in a recurrent pyramidal circuit across 50 matched random initialisations with and without VENs. The network is trained on a controlled binary classification task; we make no claim to model social cognition directly. VEN-intact networks converged in 49/50 cases (98%) versus 35/50 (70%) for VEN-ablated networks (Fisher's exact OR=21.0, 95% CI 2.7-167, p=8.7e-5). Failed ablated networks showed complete absence of learning, inconsistent with a speed-of-learning account. Phase-ablation experiments show VEN removal is most disruptive during mid-training (epochs 5-25), when a co-adaptive dependency forms in the pyramidal circuit. We derive a formal account showing VENs provide a direct gradient pathway immune to Jacobian instabilities affecting the recurrent circuit. Inference-time VEN ablation caused a significant performance drop (Wilcoxon p=0.022), ranging from no change (16/20 networks) to catastrophic collapse (0.989 to 0.620). VENs function as acquisition scaffolds whose developmental absence produces stochastic learning failure - a computational analogue of variable social skill acquisition in ASC - with falsifiable predictions for organoid and electrophysiology studies.

Esila Keskin

School of Computing and Creative Technologies, University of the West of England, Bristol, UK

Neurons Neural Network Alzheimer Cognitive Science

May 15, 2026cs.LG

Shapley Neuron Values for Continual Learning: Which Neurons Matter Most?

Continual learning enables neural networks to learn tasks sequentially without forgetting previously acquired knowledge. However, neural networks suffer from catastrophic forgetting, where learning new tasks degrades performance on earlier ones. We address this problem with Shapley Neuron Valuation (SNV), a principled framework that quantifies Neuron importance in continual learning, grounded in cooperative game theory. SNV selectively freezes important Neurons while keeping others plastic, enabling buffer-free continual learning without expanding architecture. Experiments on ImageNet-1k show that SNV consistently outperforms existing buffer-free methods. In particular, SNV improves accuracy by +2.88% in the class incremental learning and +6.46% in the task incremental learning scenarios compared to the second baseline.

Mohammad Ali Vahedifar, Abhisek Ray, Qi Zhang

DIGIT and Department of Electrical and Computer Engineering, Aarhus University, Denmark

Replay-Based Continual Learning Continual Learning Catastrophic Forgetting Singular Learning Theory Neurons Shapley Value

May 14, 2026cs.NE

Advancing Direct Training for Spiking Neural Networks with Circulate-Firing Neurons and Learnable Gradients

Spiking Neural Networks (SNNs) have emerged with promising energy-efficient property, yet a substantial performance gap persists compared to Artificial Neural Networks (ANNs). This gap stems from at least two key limitations: first, conventional spiking neurons offer limited information representation capacity, underutilizing the rich dynamics of membrane potentials; second, fixed surrogate gradient (SG) functions across time steps leads to imprecise gradient propagation, impeding effective direct training. To address these two challenges, we propose a new direct training algorithm with three core innovations: first, a circulate-firing spiking neuron model that enhances information representation capacity by leveraging membrane potentials more effectively; second, a time-step-wise learnable surrogate gradient function, enabling accurate gradient estimation during backpropagation; third, a positive-negative balanced loss function to achieve equilibrium between positive and negative membrane potentials and further boost SNN performance. Extensive experiments demonstrate that our methods achieve competitive performance across multiple datasets. Our methods can generalize seamlessly to advanced architectures of Transformer, consistently outperforming existing methods. Our work highlights the effectiveness of further harnessing intrinsic membrane dynamics of SNNs for performance improvement, and thus open a new avenue for advancing high-performance spiking neural architectures.

Feifan Zhou, Xiang Wei, Yang Liu +1

School of Artificial and Intelligence, Tianjin University, Tianjin 300072, China

Spiking Neural Networks Neural Network Surrogate Gradient Neurons

May 14, 2026cs.LG

Spectral structural distortion reveals redundant neurons in neural networks

Overparameterized neural networks often contain many removable neurons, yet what makes a neuron redundant remains poorly understood. Existing pruning criteria commonly rely on local quantities such as weight magnitude, activation strength, or gradient sensitivity, but these measures provide limited insight into the structural role of a neuron in the transformation performed by a layer. Here we show that neuronal redundancy can be characterized by weak participation in the spectral structural distortion induced by layer-wise representation transformations. For each hidden layer of a trained network, we record pre-activation and post-activation hidden states, model neurons as graph nodes, and construct input-side and output-side graphs that describe neuron-level relational structure before and after the layer transformation. We then define a spectral structural importance score that measures the contribution of each neuron to the dominant graph-spectral distortion between these two relational structures. Low-participation neurons are treated as structurally redundant and removed through an iterative pruning process in which scores are recomputed after each structural change. No parameter updates are performed during intermediate pruning rounds; after the target parameter reduction is reached, a single recovery fine-tuning stage is applied to the compact model. Direct ablation analysis and experiments across conventional neural networks, encoder-only Transformers, and decoder-only language models show that this graph-spectral criterion identifies removable neurons and Transformer units while preserving task performance after compression. These results suggest that neural redundancy is not merely a consequence of small weights or weak activations, but can be understood through weak participation in the spectral distortion of layer-wise relational structure.

Yongyu Wang

Neurons Unstructured Pruning Neural Network Graph Edge Sparsification

May 13, 2026cs.LG

Shortcut Mitigation via Spurious-Positive Samples

Shortcut mitigation strategies commonly rely on training data annotations, group-balanced held-out data or the presence of all groups, i.e., all combinations of (spurious) attributes and classes, in the training data. However, these requirements are rarely met in practice. We instead propose a method for targeted model analysis to identify a small set of instances in which the model relies on spurious attributes. Using that set and following ``this feature should not be used for prediction'' reasoning, we identify highly relevant neurons in an intermediate layer and regularize their impact. This ensures that models learn to depend on informative features rather than being right for the wrong reasons, thereby improving robustness without requiring additional balanced held-out data or annotations.

Phuong Quynh Le, Jörg Schlötterer, Sari Sadiya +2

University of Marburg · Goethe University Frankfurt

Mitigation Neurons

May 13, 2026cs.NE

FiTS: Interpretable Spiking Neurons via Frequency Selectivity and Temporal Shaping

Spiking Neural Networks (SNNs) are a promising framework for event-driven temporal processing. Prior work has improved temporal modeling through richer neuron dynamics and network-level mechanisms such as recurrence and delays, but it remains unclear how individual spiking neurons should specialize within a network. In this work, we introduce FiTS, a spiking neuron that factorizes temporal computation within each neuron into Frequency Selectivity (FS) and Temporal Shaping (TS). The FS module parameterizes each neuron's target frequency as the maximizer of its subthreshold magnitude response, while the TS module reshapes when frequency components contribute to membrane voltage accumulation through group-delay modulation. On auditory benchmarks where frequency selectivity and timing are central to the input structure, FiTS consistently improves over a plain Leaky Integrate-and-Fire (LIF) baseline in simple feedforward SNNs without recurrence or network-level delays, while remaining competitive with strong temporal SNN baselines. Beyond accuracy, the learned target frequencies and group-delay shifts provide interpretable neuron-level summaries of the frequency and timing organization learned within the network.

Jongmin Choi, Joon Son Chung

Korea Advanced Institute of Science and Technology (KAIST)

Spiking Neural Networks Neurons Spatio-Temporal Network Frequency

May 12, 2026cs.CL

GKnow: Measuring the Entanglement of Gender Bias and Factual Gender

Recent works have analyzed the impact of individual components of neural networks on gendered predictions, often with a focus on mitigating gender bias. However, mechanistic interpretations of gender tend to (i) focus on a very specific gender-related task, such as gendered pronoun prediction, or (ii) fail to distinguish between the production of factually gendered outputs (the correct assumption of gender given a word that carries gender as a semantic property) and gender biased outputs (based on a stereotype). To address these issues, we curate \gknow, a benchmark to assess gender knowledge and gender bias in language models across different types of gender-related predictions. \gknow allows us to identify and analyze circuits and individual neurons responsible for gendered predictions. We test the impact of neuron ablation on benchmarks for disentangling stereotypical and factual gender (DiFair and the test set of GKnow), as well as StereoSet. Results show that gender bias and factual gender are severely entangled on the level of both circuits and neurons, entailing that ablation is an unreliable debiasing method. Furthermore, we show that benchmarks for evaluating gender bias can hide the decrease in factual gender knowledge that accompanies neuron ablation. We curate GKnow as a contribution to the continuous development of robust gender bias benchmarks.

Leonor Veloso, Hinrich Schütze

Center for Information and Language Processing, LMU Munich · Munich Center for Machine Learning (MCML)

Gender Bias Gender Neurons Entanglement

May 12, 2026cs.LG

Scaling Laws and Tradeoffs in Recurrent Networks of Expressive Neurons

Cortical neurons are complex, multi-timescale processors wired into recurrent circuits, shaped by long evolutionary pressure under stringent biological constraints. Mainstream machine learning, by contrast, predominantly builds models from extremely simple units, a default inherited from early neural-network theory. We treat this as a normative architectural question. How should one split a fixed parameter budget

P

between the number of units

N

, per-unit effective complexity

k_e

, and per-unit connectivity

k_c

? What controls the optimal allocation? This calls for a model in which per-unit complexity can be tuned independently of width and connectivity. Accordingly, we introduce the ELM Network, whose recurrent layer is built from Expressive Leaky Memory (ELM) neurons, chosen to mirror functional components of cortical neurons. The architecture allows for individually adjusting

N

,

k_e

, and

k_c

and trains stably across orders of magnitude in scale. We evaluate the model on two qualitatively different sequence benchmarks: the neuromorphic SHD-Adding task and Enwik8 character-level language modeling. Performance improves monotonically along each of the three axes individually. Under a fixed budget, a clear non-trivial optimum emerges in their tradeoff, and larger budgets favor both more and more complex neurons. A closed-form information-theoretic model captures these tradeoffs and attributes the diminishing returns at two ends to: per-neuron signal-to-noise saturation and across-neuron redundancy. A hyperparameter sweep spanning three orders of magnitude in trainable parameters traces a near-Pareto-frontier scaling law consistent with the framework. This suggests that the simple-unit default in ML is not obviously optimal once this tradeoff surface is probed, and offers a normative lens on cortex's reliance on complex spatio-temporal integrators.

Aaron Spieler, Georg Martius, Anna Levina

University of Tübingen, Germany · Max Planck Institute for Intelligent Systems, Tübingen, Germany · Max Planck Institute for Biological Cybernetics, Tübingen, Germany

Neural Circuits Recurrent Neural Networks Neurons Scaling Laws Trade-Off

May 12, 2026cs.CL

YFPO: Yoked Feature Preference Optimization with Neuron-Guided Rewards

Preference optimization has become a widely used post-training paradigm for improving the reasoning abilities of large language models. Existing methods typically learn from preferred and dispreferred responses as external behavioral supervision, while largely ignoring capability-related signals encoded in the model's internal representations. In this work, we study whether such internal signals can provide useful auxiliary supervision for mathematical reasoning. We introduce YFPO (Yoked Feature Preference Optimization), a neuron-guided preference optimization framework that couples response-level preference learning with neuron-level rewards. YFPO first uses AttnLRP to identify math-associated internal features, and then derives an auxiliary reward from the activation margin of these neurons between preferred and dispreferred responses. This reward is combined with the standard preference optimization objective, encouraging the model to align external preferences with internal math-related features. We conduct small-scale experiments on GSM8K with a compact language model. Results show that neuron-guided rewards can influence preference optimization dynamics and yield measurable improvements in several settings, suggesting that internal representations can serve as lightweight and interpretable signals for reasoning-oriented post-training.

Yifan Le

Independent Researcher

Mathematical Reasoning Post-Training Neurons Small Large Language Models

May 12, 2026cs.NE

Multi-Timescale Conductance Spiking Networks: A Sparse, Gradient-Trainable Framework with Rich Firing Dynamics for Enhanced Temporal Processing

Spiking neural networks (SNNs) promise low-power event-driven computation for temporally rich tasks, but commonly used neuron models often trade off gradient-based trainability, dynamical richness, and high activity sparsity. These limitations are acute in regression, where approximation error, noise and spike discretization can severely degrade continuous-valued outputs. Indeed, many state-of-the-art (SOTA) SNNs rely on simple phenomenological dynamics trained with surrogate gradients and offer limited control over spiking diversity and sparsity. To overcome such limitations, we introduce multi-timescale conductance spiking networks, a gradient-trainable framework in which neural dynamics emerge from shaping the current-voltage (I-V) curve by tuning fast, slow and ultra-slow conductances. This parametrization allows systematic control over excitability, can be implemented efficiently in analog circuits, and yields rich firing regimes including tonic, phasic and bursting responses within a single model. We derive a discrete-time formulation of these differentiable dynamics, enabling direct backpropagation through time without surrogate-gradient approximations. To probe both trainability and accuracy, we evaluate feedforward networks of these neurons at the predictability limit of Mackey-Glass time-series regression and compare them to baseline LIF and SOTA AdLIF networks. Our model outperforms LIF and AdLIF networks, while exhibiting substantially sparser activity from both communication and computational perspectives. These results highlight multi-timescale conductance spiking neurons as a promising building block for energy-aware temporal processing and neuromorphic implementation.

Alex Fulleda-Garcia, Saray Soldado-Magraner, Josep Maria Margarit-Taulé

Instituto de Microelectrónica de Barcelona (IMB-CNM), Consejo Superior de Investigaciones Cientíﬁcas (CSIC) Cerdanyola del Vallès, Spain · Department of Neurobiology, University of California Los Angeles (UCLA) Los Angeles, USA · Joint Senior authors

Spiking Neural Networks Time-To-First-Spike Neural Dynamics Neuromorphic Computing Neurons Spatio-Temporal Network+1

May 9, 2026cs.CV

S2FT: Parameter-Efficient Fine-Tuning in Sparse Spectrum Domain

Parameter Efficient Fine-Tuning (PEFT) is a key technique for adapting a large pretrained model to downstream tasks by fine-tuning only a small number of parameters. Recent methods based on Fourier transforms have further reduced the fine-tuned parameters scale by only fine-tuning a few spectral coefficients. Its basic assumption is that the weight change δW is a spatial-domain matrix with a sparse spectrum. However, in this paper, we observe that the spectrum of weight change is not sparse, but instead distributed like power-uniform. This fact implies that fine-tuning only a few spectral coefficients is insufficient to accurately model the weight change with uniform spectrum. To address this issue, we propose to seek an invertible transformation that can transform a latent spatial-domain matrix with sparse spectrum to the weight change, and then perform PEFT on such sparse spectrum domain with few spectral coefficients, called S2FT. To seek such transformation, we first pre-estimate a coarse weight change as a prior. Then, inspired by that sparse spectrum often correspond to locally smooth spatial structures, we regard this transformation as a row and column rearrangement operation on the pre-estimated weight change that smooth spatial structures while keep the structure information of neurons. Finally, we propose to solve the rearrangement search problem in a simple nearest neighbor search manner, thereby obtaining the invertible transformation. Extensive results show our S2FT achieves superior performance by only using 0.08% training parameters.

Baoquan Zhang, Zhehao Yu, Lisai Zhang +5

Harbin Institute of Technology, Shenzhen · Bilibili. Inc · Shenzhen University +1

Parameter-Efficient Fine-Tuning Methods Model Fine-Tuning Neurons

May 8, 2026cs.CL

A Single Neuron Is Sufficient to Bypass Safety Alignment in Large Language Models

Safety alignment in language models operates through two mechanistically distinct systems: refusal neurons that gate whether harmful knowledge is expressed, and concept neurons that encode the harmful knowledge itself. By targeting a single neuron in each system, we demonstrate both directions of failure -- bypassing safety on explicit harmful requests via suppression, and inducing harmful content from innocent prompts via amplification -- across seven models spanning two families and 1.7B to 70B parameters, without any training or prompt engineering. Our findings suggest that safety alignment is not robustly distributed across model weights but is mediated by individual neurons that are each causally sufficient to gate refusal behavior -- suppressing any one of the identified refusal neurons bypasses safety alignment across diverse harmful requests.

Hamid Kazemi, Atoosa Chegini, Maria Safi

Apple · Work done during an internship at Apple. Present affiliation: University of Maryland, College Park.

Safety Alignment Large Language Model Alignment Harmful Content Neurons

May 7, 2026cs.LG

The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity

Despite the prevalence of the attention sink phenomenon in Large Language Models (LLMs), where initial tokens disproportionately monopolize attention scores, its structural origins remain elusive. This work provides a \textit{mechanistic explanation} for this phenomenon. First, we trace its root to the value aggregation process inherent in self-attention, which induces a systematic variance discrepancy. We further demonstrate that this discrepancy is drastically amplified by the activation of super neurons within Feed-Forward Network (FFN) layers. Specifically, the channel-sparse down-projections trigger a dimension disparity of the first-token representation, necessitating the formation of attention sinks as a structural anchor. Then, we validate this causal chain through two controlled interventions: (i) isolating the aggregation effect via attention mask modifications and (ii) amplifying the variance of targeted token representations. Both interventions can replicate attention sinks at arbitrary positions. Our mechanistic understanding offers a foundation for the systematic control of sink formation. Finally, as a proof of concept, we propose \textit{head-wise RMSNorm}, an architectural modification that stabilizes value aggregation outputs during pre-training. Our experiments demonstrate that restoring statistical parity across positions significantly accelerates convergence.

Siquan Li, Kaiqi Jiang, Jiacheng Sun +1

1The Chinese University of Hong Kong, Shenzhen · 2Huawei

Feed-Forward Neurons Variance

May 5, 2026q-bio.NC

Bio-plausible Neuromorphic Disturbance Observer Based on Emulation Theory: Extended Version

Biological neural systems achieve remarkable robustness and adaptability in uncertain environments through sparse, event-driven spike-based information processing and adaptive regulation. Inspired by this paradigm, this paper develops a neuromorhpic disturbance observer (NDO) and control framework that replaces conventional continuous-time signal representations with spike-timing encoding. Both disturbance estimates and control inputs are constructed via integrate-and-fire (IF) neuron dynamics from discrete spike events, yielding intrinsically event-driven updates. An adaptive-threshold triggering mechanism is inspired by spike-frequency adaptation (SFA), enabling history-dependent regulation of spike generation. Simulation results demonstrate that the proposed framework achieves neurally inspired robustness and adaptability, while the adaptive-threshold spiking scheme reduces spike events to 42.6% of the fixed-threshold case under noisy conditions.

Hongfu Xu, Xiaoyu Guo, Shengbo Wang +1

School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China · Department of Mechanical Engineering, City University of Hong Kong, Hong Kong · Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong +1

Neuromorphic Computing Time-To-First-Spike Neurons Emulators Theory Extended Version

May 3, 2026cs.NE

ShiftLIF: Efficient Multi-Level Spiking Neurons with Power-of-Two Quantization

Spiking neural networks (SNNs) are promising for edge sensing due to their event-driven computation and temporal filtering capability. However, standard leaky integrate-and-fire (LIF) neurons communicate only through binary spikes, which severely limit representational capacity. Existing multi-level spiking neurons improve information transmission, but often rely on uniform quantization that mismatches membrane-potential distributions or introduces costly synaptic multiplications. In this paper, we propose ShiftLIF, a multi-level spiking neuron that maps membrane potentials to a logarithmically spaced power-of-two spike set. This design provides finer representation in the small-amplitude regime, where membrane potentials are densely concentrated, while enabling multiplier-free synaptic computation through bit-shift and accumulation operations. As a result, ShiftLIF improves spike-level expressiveness without sacrificing the hardware-friendly nature of standard SNN computation. We evaluate ShiftLIF on 10 datasets spanning wireless, acoustic, motion, and visual sensing tasks. Results show that ShiftLIF consistently matches or exceeds the accuracy of existing multi-level spiking neurons while maintaining synaptic energy consumption close to standard binary LIF. These results indicate that ShiftLIF provides a favorable accuracy-efficiency trade-off for cross-modal edge sensing.

Kaiwen Tang, Di Yu, Jiaqi Zheng +4

Zhejiang University · Shandong University

Spiking Neural Networks Time-To-First-Spike Neurons Synaptic Plasticity

May 2, 2026cs.AI

NEURON: A Neuro-symbolic System for Grounded Clinical Explainability

Clinical AI adoption is hindered by the black-box/grey-box nature of high-performing models, which lack the ontological grounding and narrative transparency required for professional-level explainability. We present NEURON, a neuro-symbolic system designed to enhance both predictive reliability and clinical interpretability. NEURON integrates SNOMED CT ontology-informed structural representations with machine learning models to bridge the gap between raw data and medical nomenclature. To facilitate human-aligned interaction, the system utilizes a Retrieval-Augmented Generation (RAG) grounded LLM layer to synthesize SHAP feature attributions and patient-specific clinical notes into coherent, natural-language explanations. Validated on the MIMIC-IV dataset for Acute Heart Failure mortality prediction, NEURON improved the AUC from 0.74-0.77 to 0.84-0.88 and significantly outperformed raw SHAP visualizations in human-aligned metrics (0.85 vs. 0.50). Our results demonstrate that NEURON offers a robust, scalable engineering solution for deploying trustworthy, human-centered connected health applications.

Anuradha Chandrasekaran, Dimitrios Zikos, Mutlu Mete +3

University of North Texas, USA and Texas Tech University Health Sciences Center, USA · Texas Tech University Health Sciences Center, USA · University of North Texas, USA

Explainability Interpretability Neuro-Symbolic Framework Mimic-Iv Datasets Artificial Intelligence Systems Neurons

Apr 30, 2026cs.CL

DPN-LE: Dual Personality Neuron Localization and Editing for Large Language Models

With the widespread adoption of large language models (LLMs), understanding their personality representation mechanisms has become critical. As a novel paradigm in Personality Editing, most existing methods employ neuron-editing to locate and modify LLM neurons, requiring changes to numerous neurons and leading to significant performance degradation. This raises a fundamental question: Are all modified neurons directly related to personality representation? In this work, we investigate and quantify this specificity through assessments of general capability impact and representation-level patterns. We find that: 1) Current methods can change personalities but reduce overall performance. 2) Neurons are multifunctional, connecting personality traits and general knowledge. 3) Opposing personality traits demonstrate distinctly mutually exclusive representation patterns. Motivated by these findings, we propose DPN-LE (Dual Personality Neuron Localization and Editing), which identifies personality-specific neurons by contrasting MLP activations between high-trait and low-trait samples. DPN-LE constructs layer-wise steering vectors and applies dual-criterion filtering based on Cohen's

d

effect size and activation magnitude to isolate mutually exclusive neuron subsets. Sparse linear intervention on these neurons enables precise personality control at inference time. Using only 1,000 contrastive sample pairs per trait, DPN-LE intervenes on

\sim

0.5% of neurons while achieving competitive personality control and substantially better capability preservation across reasoning tasks. Experiments on LLaMA-3-8B-Instruct and Qwen2.5-7B-Instruct demonstrate the effectiveness and generalizability of our approach.

Lifan Zheng, Xue Yang, Jiawei Chen +6

Southeast University · Shanghai Jiao Tong University · East China Normal University +6

Personality Neurons

Apr 29, 2026cs.CL

Exploring the Limits of Pruning: Task-Specific Neurons, Model Collapse, and Recovery in Task-Specific Large Language Models

Neuron pruning is widely used to reduce the computational cost and parameter footprint of large language models, yet it remains unclear whether neurons in task-specific models contribute uniformly to task performance. In this work, we provide empirical evidence for the existence and importance of task-specific neurons through a systematic pruning study on language models specialized for mathematical reasoning and code generation. We introduce an activation-based selectivity metric to identify neurons with low contribution to the target task and prune them while preserving target-task accuracy, and compare selective pruning with random pruning. Selective pruning consistently outperforms random pruning, indicating that activation-based selectivity provides a systematic advantage over random pruning. Reverse pruning experiments further show that removing a small subset of highly task-specific neurons (~10%) causes complete performance collapse, suggesting that there exist task specific neurons and critical task information is concentrated in a small portion of the network. In contrast, selective pruning of less critical neurons (~30% - ~35%) reduces accuracy but still preserves significant performance. We also observed consistent reductions in parameters and runtime VRAM usage, along with improved inference throughput as pruning increases. Experiments on both 1.5B and 7B models reveal a robustness threshold around 15-20% pruning, beyond which accuracy loss and generation failures increase sharply. Fine-tuning substantially recovers performance across pruning levels, particularly for aggressively pruned models. These findings provide empirical evidence of neuron specialization in task-specific language models and offer insights into pruning robustness, model redundancy, and post-pruning recoverability.

M. K. Khalidi Siam, Md. Tausif-Ul-Islam, Md. Reshad Romim Khan +5

1BRAC University

Neurons Collapse

Apr 28, 2026physics.data-an

Emergent Self-Attention from Astrocyte-Gated Associative Memory Dynamics

We introduce a Hopfield-type associative memory in which effective connectivity is multiplicatively modulated by astrocytic gains evolving under an entropy-regularized replicator equation. The coupled neuron-astrocyte dynamics admit a Lyapunov function, ensuring global convergence. At fixed points, astrocytic gains implement a softmax-normalized allocation over pattern similarity scores, yielding a mechanistic realization of self-attention as emergent routing on the gain simplex. In regimes of high memory load and interference, the model significantly improves retrieval accuracy relative to classical Hopfield dynamics and recent neuron-astrocyte baselines. These results establish a dynamical systems framework linking glial modulation, competitive resource allocation, and attention-like computation.

Arnau Vivet, Alex Arenas

1Departament d’Enginyeria Inform`atica i Matem`atiques, Universitat Rovira i Virgili, 43007 Tarragona, Spain · 2Complexity Science Hub Vienna, Metternichgasse 8, 1030 Vienna, Austria

Hopfield Networks Neurons Linear Attention

Apr 27, 2026cs.CV

On the explainability of max-plus neural networks

We investigate the explanability properties of the recently proposed linear-min-max neural networks. At initialization, they can be interpreted as k-medoids with the infinity norm as a distance. Then, they are trained using subgradient descent to better fit the data. The model has been shown to be a universal approximator. Yet, we can trace the decision process because a single most activated neuron is responsible for the value of the output. Using this property, we designed a pixel fragility measure that determines whether changes to a single pixel may be responsible to a change in the classification output. Experiments on the PneumoniaMnist dataset show that this explanation for the output of the neural network compares favorably to SHAP and Integrated Gradient.

Ikhlas Enaieh, Olivier Fercoq, García Ángel

S2A, LTCI · LTCI, T´el´ecom Paris · Institut Polytechnique de Paris +5

Explainability Neural Network Neurons Fragility