Xuming Ran

Haihua Luo, Xuming Ran, Tommi Kärkkäinen et al.

The dynamic evolution of real-world necessitates model editing within Large Language Models. While existing methods explore modular isolation or parameter-efficient strategies, they still suffer from semantic drift or knowledge forgetting due to continual updating. To address these challenges, we propose SoLA, a Semantic routing-based LoRA framework for lifelong model editing. In SoLA, each edit is encapsulated as an independent LoRA module, which is frozen after training and mapped to input by semantic routing, allowing dynamic activation of LoRA modules via semantic matching. This mechanism avoids semantic drift caused by cluster updating and mitigates catastrophic forgetting from parameter sharing. More importantly, SoLA supports precise revocation of specific edits by removing key from semantic routing, which restores model's original behavior. To our knowledge, this reversible rollback editing capability is the first to be achieved in existing literature. Furthermore, SoLA integrates decision-making process into edited layer, eliminating the need for auxiliary routing networks and enabling end-to-end decision-making process. Extensive experiments demonstrate that SoLA effectively learns and retains edited knowledge, achieving accurate, efficient, and reversible lifelong model editing.

Yuan Fang, Yi Xie, Xuming Ran

Large language models encode vast factual knowledge that inevitably becomes outdated or incorrect after deployment, yet retraining is costly prohibitive, motivating model editing in lifelong settings that updates targeted behavior without harming the rest of the model. One line of work installs new facts by directly modifying base weights through locate-then-edit procedures, but accumulated edits progressively disrupt originally preserved knowledge, even with constraint-based projections. A complementary line leaves base weights intact and routes edits through external memory, but it faces routing challenges and its performance degrades at scale. We propose HoReN, a codebook-based parameter-preserving editor with enhanced routing built on three ideas. First, HoReN wraps a single MLP layer with a discrete key-value codebook, where each entry is interpreted simultaneously as a knowledge-memory key and a modern Hopfield stored pattern. Second, both keys and queries are projected onto the unit hypersphere so retrieval is governed by angular similarity, removing magnitude-driven mismatches between an edit prompt and its rephrasings. Third, the query is refined through damped Hopfield attractor dynamics, so paraphrases relax into the correct stored pattern's basin of attraction while unrelated queries remain undisturbed. HoReN achieves well-edited performance with consistent gains across diverse benchmarks spanning standard ZsRE, structured WikiBigEdit, and unstructured UnKE evaluations. Moreover, HoReN scales to 50K sequential edits on ZsRE with stable overall performance above 0.9, while prior editors collapse or degrade severely before reaching 10K. Our code is available at https://github.com/ha11ucin8/HoReN.

Haihua Luo, Xuming Ran, Jiangrong Shen et al.

Incremental Learning (IL) aims to learn new tasks while preserving previously acquired knowledge. Integrating the zero-shot learning capabilities of pre-trained vision-language models into IL methods has marked a significant advancement. However, these methods face three primary challenges: (1) the need for improved training efficiency; (2) reliance on a memory bank to store previous data; and (3) the necessity of a strong backbone to augment the model's capabilities. In this paper, we propose SimE, a Simple and Efficient framework that employs a vision-language model with adapters designed specifically for the IL task. We report a remarkable phenomenon: there is a nonlinear correlation between the number of adaptive adapter connections and the model's IL capabilities. While increasing adapter connections between transformer blocks improves model performance, adding more adaptive connections within transformer blocks during smaller incremental steps does not enhance, and may even degrade the model's IL ability. Extensive experimental results show that SimE surpasses traditional methods by 9.6% on TinyImageNet and outperforms other CLIP-based methods by 5.3% on CIFAR-100. Furthermore, we conduct a systematic study to enhance the utilization of the zero-shot capabilities of CLIP. We suggest replacing SimE's encoder with a CLIP model trained on larger datasets (e.g., LAION2B) and stronger architectures (e.g., ViT-L/14).

Haihua Luo, Xuming Ran, Zhengji Li et al.

Continual learning aims to enable models to acquire new knowledge while retaining previously learned information. Prompt-based methods have shown remarkable performance in this domain; however, they typically rely on key-value pairing, which can introduce inter-task interference and hinder scalability. To overcome these limitations, we propose a novel approach employing task-specific Prompt-Prototype (ProP), thereby eliminating the need for key-value pairs. In our method, task-specific prompts facilitate more effective feature learning for the current task, while corresponding prototypes capture the representative features of the input. During inference, predictions are generated by binding each task-specific prompt with its associated prototype. Additionally, we introduce regularization constraints during prompt initialization to penalize excessively large values, thereby enhancing stability. Experiments on several widely used datasets demonstrate the effectiveness of the proposed method. In contrast to mainstream prompt-based approaches, our framework removes the dependency on key-value pairs, offering a fresh perspective for future continual learning research.

Huiyan Xue, Xuming Ran, Yaxin Li et al.

Sparse neural systems are gaining traction for efficient continual learning due to their modularity and low interference. Architectures such as Sparse Distributed Memory Multi-Layer Perceptrons (SDMLP) construct task-specific subnetworks via Top-K activation and have shown resilience against catastrophic forgetting. However, their rigid modularity limits cross-task knowledge reuse and leads to performance degradation under high sparsity. We propose Selective Subnetwork Distillation (SSD), a structurally guided continual learning framework that treats distillation not as a regularizer but as a topology-aligned information conduit. SSD identifies neurons with high activation frequency and selectively distills knowledge within previous Top-K subnetworks and output logits, without requiring replay or task labels. This enables structural realignment while preserving sparse modularity. Experiments on Split CIFAR-10, CIFAR-100, and MNIST demonstrate that SSD improves accuracy, retention, and representation coverage, offering a structurally grounded solution for sparse continual learning.

Haihua Luo, Xuming Ran, Tommi Kärkkäinen et al.

The world is inherently dynamic, and continual learning aims to enable models to adapt to ever-evolving data streams. While pre-trained models have shown powerful performance in continual learning, they still require finetuning to adapt effectively to downstream tasks. However, prevailing Parameter-Efficient Fine-Tuning (PEFT) methods operate through empirical, black-box optimization at the weight level. These approaches lack explicit control over representation drift, leading to sensitivity to domain shifts and catastrophic forgetting in continual learning scenarios. In this work, we introduce Continual Representation Learning (CoRe), a novel framework that for the first time shifts the finetuning paradigm from weight space to representation space. Unlike conventional methods, CoRe performs task-specific interventions within a low-rank linear subspace of hidden representations, adopting a learning process with explicit objectives, which ensures stability for past tasks while maintaining plasticity for new ones. By constraining updates to a low-rank subspace, CoRe achieves exceptional parameter efficiency. Extensive experiments across multiple continual learning benchmarks demonstrate that CoRe not only preserves parameter efficiency but also significantly outperforms existing state-of-the-art methods. Our work introduces representation finetuning as a new, more effective and interpretable paradigm for continual learning.

Chang Liu, Jiangrong Shen, Xuming Ran et al.

Artificial neural networks (ANNs) have demonstrated outstanding performance in numerous tasks, but deployment in resource-constrained environments remains a challenge due to their high computational and memory requirements. Spiking neural networks (SNNs) operate through discrete spike events and offer superior energy efficiency, providing a bio-inspired alternative. However, current ANN-to-SNN conversion often results in significant accuracy loss and increased inference time due to conversion errors such as clipping, quantization, and uneven activation. This paper proposes a novel ANN-to-SNN conversion framework based on error compensation learning. We introduce a learnable threshold clipping function, dual-threshold neurons, and an optimized membrane potential initialization strategy to mitigate the conversion error. Together, these techniques address the clipping error through adaptive thresholds, dynamically reduce the quantization error through dual-threshold neurons, and minimize the non-uniformity error by effectively managing the membrane potential. Experimental results on CIFAR-10, CIFAR-100, ImageNet datasets show that our method achieves high-precision and ultra-low latency among existing conversion methods. Using only two time steps, our method significantly reduces the inference time while maintains competitive accuracy of 94.75% on CIFAR-10 dataset under ResNet-18 structure. This research promotes the practical application of SNNs on low-power hardware, making efficient real-time processing possible.

Mengyu Yang, Ye Tian, Lanshan Zhang et al.

Recently, prompt-based methods have emerged as a new alternative `parameter-efficient fine-tuning' paradigm, which only fine-tunes a small number of additional parameters while keeping the original model frozen. However, despite achieving notable results, existing prompt methods mainly focus on `what to add', while overlooking the equally important aspect of `where to add', typically relying on the manually crafted placement. To this end, we propose a region-based Adaptive Visual Prompt, named AdaViPro, which integrates the `where to add' optimization of the prompt into the learning process. Specifically, we reconceptualize the `where to add' optimization as a problem of regional decision-making. During inference, AdaViPro generates a regionalized mask map for the whole image, which is composed of 0 and 1, to designate whether to apply or discard the prompt in each specific area. Therefore, we employ Gumbel-Softmax sampling to enable AdaViPro's end-to-end learning through standard back-propagation. Extensive experiments demonstrate that our AdaViPro yields new efficiency and accuracy trade-offs for adapting pre-trained models.

Qi Xu, Shuai Gong, Xuming Ran et al.

While deep learning models have shown strong performance in simulating neural responses, they often fail to clearly separate stable visual encoding from condition-specific adaptation, which limits their ability to generalize across stimuli and individuals. We introduce the Adaptive Visual Model (AVM), a structure-preserving framework that enables condition-aware adaptation through modular subnetworks, without modifying the core representation. AVM keeps a Vision Transformer-based encoder frozen to capture consistent visual features, while independently trained modulation paths account for neural response variations driven by stimulus content and subject identity. We evaluate AVM in three experimental settings, including stimulus-level variation, cross-subject generalization, and cross-dataset adaptation, all of which involve structured changes in inputs and individuals. Across two large-scale mouse V1 datasets, AVM outperforms the state-of-the-art V1T model by approximately 2% in predictive correlation, demonstrating robust generalization, interpretable condition-wise modulation, and high architectural efficiency. Specifically, AVM achieves a 9.1% improvement in explained variance (FEVE) under the cross-dataset adaptation setting. These results suggest that AVM provides a unified framework for adaptive neural modeling across biological and experimental conditions, offering a scalable solution under structural constraints. Its design may inform future approaches to cortical modeling in both neuroscience and biologically inspired AI systems.

Yisu Wang, Ming Wang, Haoyuan Song et al.

Post-training for large language models (LLMs) is constrained by the high cost of acquiring new knowledge or correcting errors and by the unintended side effects that frequently arise from retraining. To address these issues, we introduce REPAIR (Robust Editing via Progressive Adaptive Intervention and Reintegration), a lifelong editing framework designed to support precise and low-cost model updates while preserving non-target knowledge. REPAIR mitigates the instability and conflicts of large-scale sequential edits through a closed-loop feedback mechanism coupled with dynamic memory management. Furthermore, by incorporating frequent knowledge fusion and enforcing strong locality guards, REPAIR effectively addresses the shortcomings of traditional distribution-agnostic approaches that often overlook unintended ripple effects. Our experiments demonstrate that REPAIR boosts editing accuracy by 10%-30% across multiple model families and significantly reduces knowledge forgetting. This work introduces a robust framework for developing reliable, scalable, and continually evolving LLMs.

Xuming Ran, Jie Zhang, Ziyuan Ye et al.

Artificial neural network (ANN) is a versatile tool to study the neural representation in the ventral visual stream, and the knowledge in neuroscience in return inspires ANN models to improve performance in the task. However, it is still unclear how to merge these two directions into a unified framework. In this study, we propose an integrated framework called Deep Autoencoder with Neural Response (DAE-NR), which incorporates information from ANN and the visual cortex to achieve better image reconstruction performance and higher neural representation similarity between biological and artificial neurons. The same visual stimuli (i.e., natural images) are input to both the mice brain and DAE-NR. The encoder of DAE-NR jointly learns the dependencies from neural spike encoding and image reconstruction. For the neural spike encoding task, the features derived from a specific hidden layer of the encoder are transformed by a mapping function to predict the ground-truth neural response under the constraint of image reconstruction. Simultaneously, for the image reconstruction task, the latent representation obtained by the encoder is assigned to a decoder to restore the original image under the guidance of neural information. In DAE-NR, the learning process of encoder, mapping function and decoder are all implicitly constrained by these two tasks. Our experiments demonstrate that if and only if with the joint learning, DAE-NRs can improve the performance of visual image reconstruction and increase the representation similarity between biological neurons and artificial neurons. The DAE-NR offers a new perspective on the integration of computer vision and neuroscience.

Jie Yuan, Xuming Ran, Keyin Liu et al.

Machine learning is playing an increasingly important role in medical image analysis, spawning new advances in the clinical application of neuroimaging. There have been some reviews on machine learning and epilepsy before, and they mainly focused on electrophysiological signals such as electroencephalography (EEG) and stereo electroencephalography (SEEG), while neglecting the potential of neuroimaging in epilepsy research. Neuroimaging has its important advantages in confirming the range of the epileptic region, which is essential in presurgical evaluation and assessment after surgery. However, it is difficult for EEG to locate the accurate epilepsy lesion region in the brain. In this review, we emphasize the interaction between neuroimaging and machine learning in the context of epilepsy diagnosis and prognosis. We start with an overview of epilepsy and typical neuroimaging modalities used in epilepsy clinics, MRI, DWI, fMRI, and PET. Then, we elaborate two approaches in applying machine learning methods to neuroimaging data: i) the conventional machine learning approach combining manual feature engineering and classifiers, ii) the deep learning approach, such as the convolutional neural networks and autoencoders. Subsequently, the application of machine learning on epilepsy neuroimaging, such as segmentation, localization, and lateralization tasks, as well as tasks directly related to diagnosis and prognosis are looked into in detail. Finally, we discuss the current achievements, challenges, and potential future directions in this field, hoping to pave the way for computer-aided diagnosis and prognosis of epilepsy.

Xuming Ran, Mingkun Xu, Qi Xu et al.

Variational auto-encoders (VAEs) are an influential and generally-used class of likelihood-based generative models in unsupervised learning. The likelihood-based generative models have been reported to be highly robust to the out-of-distribution (OOD) inputs and can be a detector by assuming that the model assigns higher likelihoods to the samples from the in-distribution (ID) dataset than an OOD dataset. However, recent works reported a phenomenon that VAE recognizes some OOD samples as ID by assigning a higher likelihood to the OOD inputs compared to the one from ID. In this work, we introduce a new model, namely Bigeminal Priors Variational auto-encoder (BPVAE), to address this phenomenon. The BPVAE aims to enhance the robustness of the VAEs by combing the power of VAE with the two independent priors that belong to the training dataset and simple dataset, which complexity is lower than the training dataset, respectively. BPVAE learns two datasets'features, assigning a higher likelihood for the training dataset than the simple dataset. In this way, we can use BPVAE's density estimate for detecting the OOD samples. Quantitative experimental results suggest that our model has better generalization capability and stronger robustness than the standard VAEs, proving the effectiveness of the proposed approach of hybrid learning by collaborative priors. Overall, this work paves a new avenue to potentially overcome the OOD problem via multiple latent priors modeling.

Xuming Ran, Mingkun Xu, Lingrui Mei et al.

Variational autoencoders (VAEs) are influential generative models with rich representation capabilities from the deep neural network architecture and Bayesian method. However, VAE models have a weakness that assign a higher likelihood to out-of-distribution (OOD) inputs than in-distribution (ID) inputs. To address this problem, a reliable uncertainty estimation is considered to be critical for in-depth understanding of OOD inputs. In this study, we propose an improved noise contrastive prior (INCP) to be able to integrate into the encoder of VAEs, called INCPVAE. INCP is scalable, trainable and compatible with VAEs, and it also adopts the merits from the INCP for uncertainty estimation. Experiments on various datasets demonstrate that compared to the standard VAEs, our model is superior in uncertainty estimation for the OOD data and is robust in anomaly detection tasks. The INCPVAE model obtains reliable uncertainty estimation for OOD inputs and solves the OOD problem in VAE models.