Related papers: CSP-Atlas: Concept-Specific Neural Circuits in a S…

Explicit Chain-of-Thought improves the reasoning performance of large language models but often incurs high inference cost due to verbose token-level traces. While recent approaches reduce this overhead via concise prompting or step…

We introduce the Convolutional Set Transformer (CST), a novel neural architecture designed to process image sets of arbitrary cardinality that are visually heterogeneous yet share high-level semantics - such as a common category, scene, or…

Automated construction is one of the most promising areas that can improve efficiency, reduce costs and minimize errors in the process of building construction. In this paper, a comparative analysis of three neural network models for…

Weakly-supervised semantic segmentation aims to assign each pixel a semantic category under weak supervisions, such as image-level tags. Most of existing weakly-supervised semantic segmentation methods do not use any feedback from…

Sparse coding, which is the decomposition of a vector using only a few basis elements, is widely used in machine learning and image processing. The basis set, also called dictionary, is learned to adapt to specific data. This approach has…

We propose a new unbiased threshold for network analysis named the Cluster-Span Threshold (CST). This is based on the clustering coefficient, C, following logic that a balance of `clustering' to `spanning' triples results in a useful…

A widely used strategy to discover and understand language model mechanisms is circuit analysis. A circuit is a minimal subgraph of a model's computation graph that executes a specific task. We identify a gap in existing circuit discovery…

A fundamental question in interpretability research is to what extent neural networks, particularly language models, implement reusable functions through subnetworks that can be composed to perform more complex tasks. Recent advances in…

We study how reliably sparse autoencoders (SAEs) support claims about reasoning-related internal features in large language models. We first give a stylized analysis showing that sparsity-regularized decoding can preferentially retain…

Recently, large models, such as Vision Transformer and BERT, have garnered significant attention due to their exceptional performance. However, their extensive computational requirements lead to considerable power and hardware resource…

We present the performance of a semantic segmentation network, SparseSSNet, that provides pixel-level classification of MicroBooNE data. The MicroBooNE experiment employs a liquid argon time projection chamber for the study of neutrino…

Boolean circuits form the foundational computational substrate of symmetric cryptography, yet the exploration of their architectural design space has remained largely confined to a handful of canonical paradigms - SPN, Feistel networks, and…

Attention improves representation learning over RNNs, but its discrete nature limits continuous-time (CT) modeling. We introduce Neuronal Attention Circuit (NAC), a novel, biologically inspired CT-Attention mechanism that reformulates…

Sparse convolutional neural networks (CNNs) have gained significant traction over the past few years as sparse CNNs can drastically decrease the model size and computations, if exploited befittingly, as compared to their dense counterparts.…

We propose a novel weakly-supervised semantic segmentation algorithm based on Deep Convolutional Neural Network (DCNN). Contrary to existing weakly-supervised approaches, our algorithm exploits auxiliary segmentation annotations available…

Deep vision models have achieved remarkable classification performance by leveraging a hierarchical architecture in which human-interpretable concepts emerge through the composition of individual neurons across layers. Given the distributed…

The functional and structural representation of the brain as a complex network is marked by the fact that the comparison of noisy and intrinsically correlated high-dimensional structures between experimental conditions or groups shuns…

Large language models handle single-turn generation well, but multi-turn interactions still require the model to reconstruct user intent and task state from an expanding token history because internal representations do not persist across…

To enhance developer productivity, all modern integrated development environments (IDEs) include code suggestion functionality that proposes likely next tokens at the cursor. While current IDEs work well for statically-typed languages,…

Semantic segmentation, like other fields of computer vision, has seen a remarkable performance advance by the use of deep convolution neural networks. However, considering that neighboring pixels are heavily dependent on each other, both…