Related papers: Sparse Concept Anchoring for Interpretable and Con…

CLIP embeddings have demonstrated remarkable performance across a wide range of multimodal applications. However, these high-dimensional, dense vector representations are not easily interpretable, limiting our understanding of the rich…

We introduce a weakly supervised approach for inferring the property of abstractness of words and expressions in the complete absence of labeled data. Exploiting only minimal linguistic clues and the contextual usage of a concept as…

Concept erasure in text-to-image diffusion models seeks to remove undesired concepts while preserving overall generative capability. Localized erasure methods aim to restrict edits to the spatial region occupied by the target concept.…

In this note we present a generative model of natural images consisting of a deep hierarchy of layers of latent random variables, each of which follows a new type of distribution that we call rectified Gaussian. These rectified Gaussian…

Recent advances in explainable machine learning have highlighted the potential of sparse autoencoders in uncovering mono-semantic features in densely encoded embeddings. While most research has focused on Large Language Model (LLM)…

Generative modeling of anatomical structures plays a crucial role in virtual imaging trials, which allow researchers to perform studies without the costs and constraints inherent to in vivo and phantom studies. For clinical relevance,…

Large pre-trained transformers have revolutionized artificial intelligence across various domains, and fine-tuning remains the dominant approach for adapting these models to downstream tasks due to the cost of training from scratch.…

Sparsity is regarded as a desirable property of representations, especially in terms of explanation. However, its usage has been limited due to the gap with dense representations. Most NLP research progresses in recent years are based on…

Interpretability is critical in high-stakes domains such as medical imaging, where understanding model decisions is essential for clinical adoption. In this work, we introduce Sparse Autoencoder (SAE)-based interpretability to breast…

A key barrier to interpreting large language models is polysemanticity, where neurons activate for multiple unrelated concepts. Sparse autoencoders (SAEs) have been proposed to mitigate this issue by transforming dense activations into…

A new line of research for feature selection based on neural networks has recently emerged. Despite its superiority to classical methods, it requires many training iterations to converge and detect informative features. The computational…

Neural network-based language models deal with data sparsity problems by mapping the large discrete space of words into a smaller continuous space of real-valued vectors. By learning distributed vector representations for words, each…

Sparse neural networks are a key factor in developing resource-efficient machine learning applications. We propose the novel and powerful sparse learning method Adaptive Regularized Training (ART) to compress dense into sparse networks.…

Learning embedding spaces of suitable geometry is critical for representation learning. In order for learned representations to be effective and efficient, it is ideal that the geometric inductive bias aligns well with the underlying…

Diffusion models, while powerful, can inadvertently generate harmful or undesirable content, raising significant ethical and safety concerns. Recent machine unlearning approaches offer potential solutions but often lack transparency, making…

Real-world processes often contain intermediate state that can be modeled as an extremely sparse activation tensor. In this work, we analyze the identifiability of such sparse and local latent intermediate variables, which we call motifs.…

This paper introduces a new probabilistic framework for supervised learning in neural systems. It is designed to model complex, uncertain systems whose random outputs are strongly non-Gaussian given deterministic inputs. The architecture…

Following the general theoretical framework of VSA (Vector Symbolic Architecture), a cognitive model with the use of sparse binary hypervectors is proposed. In addition, learning algorithms are introduced to bootstrap the model from…

Decomposing model activations into interpretable components is a key open problem in mechanistic interpretability. Sparse autoencoders (SAEs) are a popular method for decomposing the internal activations of trained transformers into sparse,…

We propose semantic anchoring, a unified account of how large language models turn pretrained capacity into goal-directed behavior: external structure (in-context examples, retrieval, or light tuning) binds the model's latent patterns to…