arXiv subjects

Emre Neftci

Publications and source records attributed to Emre Neftci.

At least 19 recordsLinked to original sources

arXiv2026

GRINQH: Graded Input-based Quantization Hierarchy for Efficient LLM Generation

Autoregressive decoding with LLMs is primarily bottlenecked by GPU memory bandwidth, especially in edge-computing settings. While quantization is essential for mitigating this bottleneck, most existing methods treat inference as a uniform process and fail to account for the asymmetry between the compute-bound prefill stage and the memory-bound decoding stage. We propose GRINQH (GRaded INput-based Quantization Hierarchy), a weight-only post-training quantization framework that accelerates decoding by unifying quantization and sparsification. GRINQH leverages activation magnitudes as a proxy for computational importance to dynamically assign weight channels to different precision levels, enabling flexible average bit widths during decoding. Evaluated on Llama3 and Qwen3 models, GRINQH outperforms state-of-the-art fixed- and mixed-precision baselines at comparable 3- and 4-bit settings, even enabling effective 2-bit generation. We experimentally verify theoretical speedups by leveraging a hierarchical nested memory layout for multi-precision storage in a custom GPU kernel. Ultimately, GRINQH establishes a new state-of-the-art Pareto frontier for LLM generation, enabling a dynamic trade-off between generation quality and inference speed.

cs.LG↗

arXiv2026

Sparse Axonal and Dendritic Delays Enable Competitive SNNs for Keyword Classification

Training transmission delays in spiking neural networks (SNNs) has been shown to substantially improve their performance on complex temporal tasks. In this work, we show that learning either axonal or dendritic delays enables deep feedforward SNNs composed of leaky integrate-and-fire (LIF) neurons to reach accuracy comparable to existing synaptic delay learning approaches, while significantly reducing memory and computational overhead. SNN models with either axonal or dendritic delays achieve up to $95.58\%$ on the Google Speech Command (GSC) and $80.97\%$ on the Spiking Speech Command (SSC) datasets, matching or exceeding prior methods based on synaptic delays or more complex neuron models. By adjusting the delay parameters, we obtain improved performance for synaptic delay learning baselines, strengthening the comparison. We find that axonal delays offer the most favorable trade-off, combining lower buffering requirements with slightly higher accuracy than dendritic delays. We further show that the performance of axonal and dendritic delay models is largely preserved under strong delay sparsity, with as few as $20\%$ of delays remaining active, further reducing buffering requirements. Overall, our results indicate that learnable axonal and dendritic delays provide a resource-efficient and effective mechanism for temporal representation in SNNs. Code will be made available publicly upon acceptance. Code is available at https://github.com/YounesBouhadjar/AxDenSynDelaySNN

cs.NE↗

arXiv2026

Learning to Remember, Learn, and Forget in Attention-Based Models

In-Context Learning (ICL) in transformers acts as an online associative memory and is believed to underpin their high performance on complex sequence processing tasks. However, in gated linear attention models, this memory has a fixed capacity and is prone to interference, especially for long sequences. We propose Palimpsa, a self-attention model that views ICL as a continual learning problem that must address a stability-plasticity dilemma. Palimpsa uses Bayesian metaplasticity, where the plasticity of each attention state is tied to an importance state grounded by a prior distribution that captures accumulated knowledge. We demonstrate that various gated linear attention models emerge as specific architecture choices and posterior approximations, and that Mamba2 is a special case of Palimpsa where forgetting dominates. This theoretical link enables the transformation of any non-metaplastic model into a metaplastic one, significantly expanding its memory capacity. Our experiments show that Palimpsa consistently outperforms baselines on the Multi-Query Associative Recall (MQAR) benchmark and on Commonsense Reasoning tasks.

cs.LG↗

arXiv2026

Dissecting Linear Recurrent Models: How Different Gating Strategies Drive Selectivity and Generalization

Linear recurrent neural networks have emerged as efficient alternatives to the original Transformer's softmax attention mechanism, thanks to their highly parallelizable training and constant memory and computation requirements at inference. Iterative refinements of these models have introduced an increasing number of architectural mechanisms, leading to increased complexity and computational costs. Nevertheless, systematic direct comparisons among these models remain limited. Existing benchmark tasks are either too simplistic to reveal substantial differences or excessively resource-intensive for experimentation. In this work, we propose a refined taxonomy of linear recurrent models and introduce SelectivBench, a set of lightweight and customizable synthetic benchmark tasks for systematically evaluating sequence models. SelectivBench specifically evaluates selectivity in sequence models at small to medium scale, such as the capacity to focus on relevant inputs while ignoring context-based distractors. It employs rule-based grammars to generate sequences with adjustable complexity, incorporating irregular gaps that intentionally violate transition rules. Evaluations of linear recurrent models on SelectivBench reveal performance patterns consistent with results from large-scale language tasks. Our analysis clarifies the roles of essential architectural features: gating and rapid forgetting mechanisms facilitate recall, in-state channel mixing is unnecessary for selectivity, but critical for generalization, and softmax attention remains dominant due to its memory capacity scaling with sequence length. Our benchmark enables targeted, efficient exploration of linear recurrent models and provides a controlled setting for studying behaviors observed in large-scale evaluations. Code is available at https://github.com/symseqbench/selectivbench

cs.LG↗

arXiv2026

Hybrid guided variational autoencoder for visual place recognition

Autonomous agents such as cars, robots and drones need to precisely localize themselves in diverse environments, including in GPS-denied indoor environments. One approach for precise localization is visual place recognition (VPR), which estimates the place of an image based on previously seen places. State-of-the-art VPR models require high amounts of memory, making them unwieldy for mobile deployment, while more compact models lack robustness and generalization capabilities. This work overcomes these limitations for robotics using a combination of event-based vision sensors and an event-based novel guided variational autoencoder (VAE). The encoder part of our model is based on a spiking neural network model which is compatible with power-efficient low latency neuromorphic hardware. The VAE successfully disentangles the visual features of 16 distinct places in our new indoor VPR dataset with a classification performance comparable to other state-of-the-art approaches while, showing robust performance also under various illumination conditions. When tested with novel visual inputs from unknown scenes, our model can distinguish between these places, which demonstrates a high generalization capability by learning the essential features of location. Our compact and robust guided VAE with generalization capabilities poses a promising model for visual place recognition that can significantly enhance mobile robot navigation in known and unknown indoor environments.

cs.CV↗

arXiv2025

SymSeqBench: a unified framework for the generation and analysis of rule-based symbolic sequences and datasets

Sequential structure is a key feature of multiple domains of natural cognition and behavior, such as language, movement and decision-making. Likewise, it is also a central property of tasks to which we would like to apply artificial intelligence. It is therefore of great importance to develop frameworks that allow us to evaluate sequence learning and processing in a domain agnostic fashion, whilst simultaneously providing a link to formal theories of computation and computability. To address this need, we introduce two complementary software tools: SymSeq, designed to rigorously generate and analyze structured symbolic sequences, and SeqBench, a comprehensive benchmark suite of rule-based sequence processing tasks to evaluate the performance of artificial learning systems in cognitively relevant domains. In combination, SymSeqBench offers versatility in investigating sequential structure across diverse knowledge domains, including experimental psycholinguistics, cognitive psychology, behavioral analysis, neuromorphic computing and artificial intelligence. Due to its basis in Formal Language Theory (FLT), SymSeqBench provides researchers in multiple domains with a convenient and practical way to apply the concepts of FLT to conceptualize and standardize their experiments, thus advancing our understanding of cognition and behavior through shared computational frameworks and formalisms. The tool is modular, openly available and accessible to the research community.

q-bio.NC↗

arXiv2025

QS4D: Quantization-aware training for efficient hardware deployment of structured state-space sequential models

Structured State Space models (SSM) have recently emerged as a new class of deep learning models, particularly well-suited for processing long sequences. Their constant memory footprint, in contrast to the linearly scaling memory demands of Transformers, makes them attractive candidates for deployment on resource-constrained edge-computing devices. While recent works have explored the effect of quantization-aware training (QAT) on SSMs, they typically do not address its implications for specialized edge hardware, for example, analog in-memory computing (AIMC) chips. In this work, we demonstrate that QAT can significantly reduce the complexity of SSMs by up to two orders of magnitude across various performance metrics. We analyze the relation between model size and numerical precision, and show that QAT enhances robustness to analog noise and enables structural pruning. Finally, we integrate these techniques to deploy SSMs on a memristive analog in-memory computing substrate and highlight the resulting benefits in terms of computational efficiency.

cs.LG↗

arXiv2025

SiLIF: Structured State Space Model Dynamics and Parametrization for Spiking Neural Networks

Multi-state spiking neurons combine sparse binary activations with rich second-order nonlinear recurrent dynamics, making them a promising alternative to standard deep learning models. However, gradient propagation through these dynamics often leads to instabilities that hinder scalability and performance. Inspired by the stable training and strong performance of state space models (SSMs) on long sequences, we introduce two SSM-inspired Leaky Integrate-and-Fire (SiLIF) neuron models. The first extends a two-state neuron with a learnable discretization timestep and logarithmic reparametrization, while the second additionally incorporates the initialization scheme and structure of complex-state SSMs, enabling oscillatory regimes. Our two SiLIF models achieve new state-of-the-art performance among spiking neuron models on both event-based and raw-audio speech recognition datasets. We further demonstrate a favorable performance-efficiency trade-off compared to SSMs, even surpassing them while using half the computational cost through the use of synaptic delays. Our code is available at https://github.com/Maxtimer97/SSM-inspired-LIF.

cs.NE↗

arXiv2025

Contrastive Consolidation of Top-Down Modulations Achieves Sparsely Supervised Continual Learning

Biological brains learn continually from a stream of unlabeled data, while integrating specialized information from sparsely labeled examples without compromising their ability to generalize. Meanwhile, machine learning methods are susceptible to catastrophic forgetting in this natural learning setting, as supervised specialist fine-tuning degrades performance on the original task. We introduce task-modulated contrastive learning (TMCL), which takes inspiration from the biophysical machinery in the neocortex, using predictive coding principles to integrate top-down information continually and without supervision. We follow the idea that these principles build a view-invariant representation space, and that this can be implemented using a contrastive loss. Then, whenever labeled samples of a new class occur, new affine modulations are learned that improve separation of the new class from all others, without affecting feedforward weights. By co-opting the view-invariance learning mechanism, we then train feedforward weights to match the unmodulated representation of a data sample to its modulated counterparts. This introduces modulation invariance into the representation space, and, by also using past modulations, stabilizes it. Our experiments show improvements in both class-incremental and transfer learning over state-of-the-art unsupervised approaches, as well as over comparable supervised approaches, using as few as 1% of available labels. Taken together, our work suggests that top-down modulations play a crucial role in balancing stability and plasticity.

cs.LG↗

arXiv2025

A Grid Cell-Inspired Structured Vector Algebra for Cognitive Maps

The entorhinal-hippocampal formation is the mammalian brain's navigation system, encoding both physical and abstract spaces via grid cells. This system is well-studied in neuroscience, and its efficiency and versatility make it attractive for applications in robotics and machine learning. While continuous attractor networks (CANs) successfully model entorhinal grid cells for encoding physical space, integrating both continuous spatial and abstract spatial computations into a unified framework remains challenging. Here, we attempt to bridge this gap by proposing a mechanistic model for versatile information processing in the entorhinal-hippocampal formation inspired by CANs and Vector Symbolic Architectures (VSAs), a neuro-symbolic computing framework. The novel grid-cell VSA (GC-VSA) model employs a spatially structured encoding scheme with 3D neuronal modules mimicking the discrete scales and orientations of grid cell modules, reproducing their characteristic hexagonal receptive fields. In experiments, the model demonstrates versatility in spatial and abstract tasks: (1) accurate path integration for tracking locations, (2) spatio-temporal representation for querying object locations and temporal relations, and (3) symbolic reasoning using family trees as a structured test case for hierarchical relationships.

cs.NE↗

arXiv2025

A Truly Sparse and General Implementation of Gradient-Based Synaptic Plasticity

Online synaptic plasticity rules derived from gradient descent achieve high accuracy on a wide range of practical tasks. However, their software implementation often requires tediously hand-derived gradients or using gradient backpropagation which sacrifices the online capability of the rules. In this work, we present a custom automatic differentiation (AD) pipeline for sparse and online implementation of gradient-based synaptic plasticity rules that generalizes to arbitrary neuron models. Our work combines the programming ease of backpropagation-type methods for forward AD while being memory-efficient. To achieve this, we exploit the advantageous compute and memory scaling of online synaptic plasticity by providing an inherently sparse implementation of AD where expensive tensor contractions are replaced with simple element-wise multiplications if the tensors are diagonal. Gradient-based synaptic plasticity rules such as eligibility propagation (e-prop) have exactly this property and thus profit immensely from this feature. We demonstrate the alignment of our gradients with respect to gradient backpropagation on an synthetic task where e-prop gradients are exact, as well as audio speech classification benchmarks. We demonstrate how memory utilization scales with network size without dependence on the sequence length, as expected from forward AD methods.

cs.NE↗

arXiv2024

Optimal Gradient Checkpointing for Sparse and Recurrent Architectures using Off-Chip Memory

Recurrent neural networks (RNNs) are valued for their computational efficiency and reduced memory requirements on tasks involving long sequence lengths but require high memory-processor bandwidth to train. Checkpointing techniques can reduce the memory requirements by only storing a subset of intermediate states, the checkpoints, but are still rarely used due to the computational overhead of the additional recomputation phase. This work addresses these challenges by introducing memory-efficient gradient checkpointing strategies tailored for the general class of sparse RNNs and Spiking Neural Networks (SNNs). SNNs are energy efficient alternatives to RNNs thanks to their local, event-driven operation and potential neuromorphic implementation. We use the Intelligence Processing Unit (IPU) as an exemplary platform for architectures with distributed local memory. We exploit its suitability for sparse and irregular workloads to scale SNN training on long sequence lengths. We find that Double Checkpointing emerges as the most effective method, optimizing the use of local memory resources while minimizing recomputation overhead. This approach reduces dependency on slower large-scale memory access, enabling training on sequences over 10 times longer or 4 times larger networks than previously feasible, with only marginal time overhead. The presented techniques demonstrate significant potential to enhance scalability and efficiency in training sparse and recurrent networks across diverse hardware platforms, and highlights the benefits of sparse activations for scalable recurrent neural network training.

cs.NE↗

arXiv2024

Zero-Shot Temporal Resolution Domain Adaptation for Spiking Neural Networks

Spiking Neural Networks (SNNs) are biologically-inspired deep neural networks that efficiently extract temporal information while offering promising gains in terms of energy efficiency and latency when deployed on neuromorphic devices. SNN parameters are sensitive to temporal resolution, leading to significant performance drops when the temporal resolution of target data during deployment is not the same as that of the source data used for training, especially when fine-tuning with the target data is not possible during deployment. To address this challenge, we propose three novel domain adaptation methods for adapting neuron parameters to account for the change in time resolution without re-training on target time resolution. The proposed methods are based on a mapping between neuron dynamics in SNNs and State Space Models (SSMs) and are applicable to general neuron models. We evaluate the proposed methods under spatio-temporal data tasks, namely the audio keyword spotting datasets SHD and MSWC, and the neuromorphic image NMINST dataset. Our methods provide an alternative to-and in most cases significantly outperform-the existing reference method that consists of scaling only the time constant. Notably, when the temporal resolution of the target data is double that of the source data, applying one of our proposed methods instead of the benchmark achieves classification accuracy of 89.5% instead of 53.0% on SHD, 93.6% instead of 38.8% on MSWC and 98.5% instead of 97.2% aon NMNIST. Moreover, our results show that high accuracy on high temporal resolution data can be obtained by time-efficient training on lower temporal resolution data.

cs.LG↗

arXiv2024

Gain Cell-Based Analog Content Addressable Memory for Dynamic Associative tasks in AI

Analog Content Addressable Memories (aCAMs) have proven useful for associative in-memory computing applications like Decision Trees, Finite State Machines, and Hyper-dimensional Computing. While non-volatile implementations using FeFETs and ReRAM devices offer speed, power, and area advantages, they suffer from slow write speeds and limited write cycles, making them less suitable for computations involving fully dynamic data patterns. To address these limitations, in this work, we propose a capacitor gain cell-based aCAM designed for dynamic processing, where frequent memory updates are required. Our system compares analog input voltages to boundaries stored in capacitors, enabling efficient dynamic tasks. We demonstrate the application of aCAM within transformer attention mechanisms by replacing the softmax-scaled dot-product similarity with aCAM similarity, achieving competitive results. Circuit simulations on a TSMC 28 nm node show promising performance in terms of energy efficiency, precision, and latency, making it well-suited for fast, dynamic AI applications.

cs.ET↗

arXiv2024

Unsupervised Learning of Spatio-Temporal Patterns in Spiking Neuronal Networks

The ability to predict future events or patterns based on previous experience is crucial for many applications such as traffic control, weather forecasting, or supply chain management. While modern supervised Machine Learning approaches excel at such sequential tasks, they are computationally expensive and require large training data. A previous work presented a biologically plausible sequence learning model, developed through a bottom-up approach, consisting of a spiking neural network and unsupervised local learning rules. The model in its original formulation identifies only a specific type of sequence elements composed of synchronous spikes by activating a subset of neurons with identical stimulus preference. In this work, we extend the model to detect and learn sequences of various spatio-temporal patterns (STPs) by incorporating plastic connections in the input synapses. We showcase that the model is able to learn and predict high-order sequences. We further study the robustness of the model against different input settings and parameters.

q-bio.NC↗

arXiv2024

On-Chip Learning via Transformer In-Context Learning

Autoregressive decoder-only transformers have become key components for scalable sequence processing and generation models. However, the transformer's self-attention mechanism requires transferring prior token projections from the main memory at each time step (token), thus severely limiting their performance on conventional processors. Self-attention can be viewed as a dynamic feed-forward layer, whose matrix is input sequence-dependent similarly to the result of local synaptic plasticity. Using this insight, we present a neuromorphic decoder-only transformer model that utilizes an on-chip plasticity processor to compute self-attention. Interestingly, the training of transformers enables them to ``learn'' the input context during inference. We demonstrate this in-context learning ability of transformers on the Loihi 2 processor by solving a few-shot classification problem. With this we emphasize the importance of pretrained models especially their ability to find simple, local, backpropagation free, learning rules enabling on-chip learning and adaptation in a hardware friendly manner.

cs.NE↗

arXiv2024

Analog In-Memory Computing Attention Mechanism for Fast and Energy-Efficient Large Language Models

Transformer networks, driven by self-attention, are central to Large Language Models. In generative Transformers, self-attention uses cache memory to store token projections, avoiding recomputation at each time step. However, GPU-stored projections must be loaded into SRAM for each new generation step, causing latency and energy bottlenecks. We present a custom self-attention in-memory computing architecture based on emerging charge-based memories called gain cells, which can be efficiently written to store new tokens during sequence generation and enable parallel analog dot-product computation required for self-attention. However, the analog gain cell circuits introduce non-idealities and constraints preventing the direct mapping of pre-trained models. To circumvent this problem, we design an initialization algorithm achieving text processing performance comparable to GPT-2 without training from scratch. Our architecture respectively reduces attention latency and energy consumption by up to two and five orders of magnitude compared to GPUs, marking a significant step toward ultra-fast, low-power generative Transformers.

cs.NE↗

arXiv2024

SNNAX -- Spiking Neural Networks in JAX

Spiking Neural Networks (SNNs) simulators are essential tools to prototype biologically inspired models and neuromorphic hardware architectures and predict their performance. For such a tool, ease of use and flexibility are critical, but so is simulation speed especially given the complexity inherent to simulating SNN. Here, we present SNNAX, a JAX-based framework for simulating and training such models with PyTorch-like intuitiveness and JAX-like execution speed. SNNAX models are easily extended and customized to fit the desired model specifications and target neuromorphic hardware. Additionally, SNNAX offers key features for optimizing the training and deployment of SNNs such as flexible automatic differentiation and just-in-time compilation. We evaluate and compare SNNAX to other commonly used machine learning (ML) frameworks used for programming SNNs. We provide key performance metrics, best practices, documented examples for simulating SNNs in SNNAX, and implement several benchmarks used in the literature.

cs.NE↗