Related papers: Neurons in Large Language Models: Dead, N-gram, Po…

Many recent studies have found evidence for emergent reasoning capabilities in large language models (LLMs), but debate persists concerning the robustness of these capabilities, and the extent to which they depend on structured reasoning…

In Transformer language models, activation vectors transform from current token embeddings to next token predictions as they pass through the model. To isolate a minimal form of this transformation, we identify language model subnetworks…

While graph neural networks (GNNs) have shown remarkable performance across diverse graph-related tasks, their high-dimensional hidden representations render them black boxes. In this work, we propose Graph Lingual Network (GLN), a GNN…

Large language models (LLMs) often develop learned mechanisms specialized to specific datasets, such as reliance on domain-specific correlations, which yield high-confidence predictions without generalizable reasoning. While beneficial in…

Understanding the latent space geometry of large language models (LLMs) is key to interpreting their behavior and improving alignment. Yet it remains unclear to what extent LLMs linearly organize representations related to semantic…

Text-attributed graphs have recently garnered significant attention due to their wide range of applications in web domains. Existing methodologies employ word embedding models for acquiring text representations as node features, which are…

Projecting visual features into word embedding space has become a significant fusion strategy adopted by Multimodal Large Language Models (MLLMs). However, its internal mechanisms have yet to be explored. Inspired by multilingual research,…

This work demonstrates how increasing the number of neurons in a network without increasing its number of non-zero parameters improves performance. We show that this gain corresponds with a decrease in interference between multiple features…

The drive toward automating cellular network operations has grown with the increasing complexity of these systems. Despite advancements, full autonomy currently remains out of reach due to reliance on human intervention for modeling network…

Modern Deep Neural Networks (DNNs) require significant memory to store weight, activations, and other intermediate tensors during training. Hence, many models do not fit one GPU device or can be trained using only a small per-GPU batch…

Learning to read places a strong challenge on the visual system. Years of expertise lead to a remarkable capacity to separate highly similar letters and encode their relative positions, thus distinguishing words such as FORM and FROM,…

A number of differences have emerged between modern and classic approaches to constituency parsing in recent years, with structural components like grammars and feature-rich lexicons becoming less central while recurrent neural network…

The development of large language models (LLMs) is costly and has significant commercial value. Consequently, preventing unauthorized appropriation of open-source LLMs and protecting developers' intellectual property rights have become…

We introduce a framework for reasoning about what meaning is captured by the neurons in a trained neural network. We provide a strategy for discovering meaning by training a second model (referred to as an observer model) to classify the…

We train neural models for morphological analysis, generation and lemmatization for morphologically rich languages. We present a method for automatically extracting substantially large amount of training data from FSTs for 22 languages, out…

Foundation models are powerful yet often opaque in their decision-making. A topic of continued interest in both neuroscience and artificial intelligence is whether some neurons behave like grandmother cells, i.e., neurons that are…

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…

Large Language Models (LLMs) have fundamentally altered how we approach scaling in machine learning. However, these models pose substantial computational and memory challenges, primarily due to the reliance on matrix multiplication (MatMul)…

Despite the remarkable evolution of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. Previous work largely focused on what these models learn at the representation level. We break this…

Multilingual neural machine translation with a single model has drawn much attention due to its capability to deal with multiple languages. However, the current multilingual translation paradigm often makes the model tend to preserve the…