相关论文: Not All Language Model Features Are One-Dimensiona…

Deep learning models have become fundamental tools in drug design. In particular, large language models trained on biochemical sequences learn feature vectors that guide drug discovery through virtual screening. However, such models do not…

Large Language Models (LLMs) such as Mistral and LLaMA have showcased remarkable performance across various natural language processing (NLP) tasks. Despite their success, these models inherit social biases from the diverse datasets on…

Recent literature shows that large-scale language modeling provides excellent reusable sentence representations with both recurrent and self-attentive architectures. However, there has been less clarity on the commonalities and differences…

Transformers excel in Natural Language Processing (NLP) due to their prowess in capturing long-term dependencies but suffer from exponential resource consumption with increasing sequence lengths. To address these challenges, we propose MCSD…

From extracting features to generating text, the outputs of large language models (LLMs) typically rely on the final layers, following the conventional wisdom that earlier layers capture only low-level cues. However, our analysis shows that…

Large Language Models (LLMs) exhibit highly anisotropic internal representations, often characterized by massive activations, a phenomenon where a small subset of feature dimensions possesses magnitudes significantly larger than the rest.…

Large language models perform text generation through high-dimensional internal dynamics, yet the temporal organisation of these dynamics remains poorly understood. Most interpretability approaches emphasise static representations or causal…

A central question in cognitive science is whether conceptual representations converge onto a shared manifold to support generalization, or diverge into orthogonal subspaces to minimize task interference. While prior work has discovered…

Transformer-based language models have achieved significant success; however, their internal mechanisms remain largely opaque due to the complexity of non-linear interactions and high-dimensional operations. While previous studies have…

Language models can perceive visual properties of text despite receiving only sequences of tokens-we mechanistically investigate how Claude 3.5 Haiku accomplishes one such task: linebreaking in fixed-width text. We find that character…

Current multilingual vision-language models either require a large number of additional parameters for each supported language, or suffer performance degradation as languages are added. In this paper, we propose a Scalable Multilingual…

Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks, these networks appear to extract generally useful linguistic…

We study last-layer outlier dimensions, i.e. dimensions that display extreme activations for the majority of inputs. We show that outlier dimensions arise in many different modern language models, and trace their function back to the…

Large language models (LLMs) achieve state-of-the-art results across many natural language tasks, but their internal mechanisms remain difficult to interpret. In this work, we extract, process, and visualize latent state geometries in…

Transformers have significantly advanced the field of natural language processing, but comprehending their internal mechanisms remains a challenge. In this paper, we introduce a novel geometric perspective that elucidates the inner…

We assess how multilingual language models maintain a shared multilingual representation space while still encoding language-sensitive information in each language. Using XLM-R as a case study, we show that languages occupy similar linear…

We explore the topology of representation manifolds arising in autoregressive neural language models trained on raw text data. In order to study their properties, we introduce tools from computational algebraic topology, which we use as a…

Development of language proficiency models for non-native learners has been an active area of interest in NLP research for the past few years. Although language proficiency is multidimensional in nature, existing research typically…

Self-explaining models are models that reveal decision making parameters in an interpretable manner so that the model reasoning process can be directly understood by human beings. General Linear Models (GLMs) are self-explaining because the…

Words have been represented in a high-dimensional vector space that encodes their semantic similarities, enabling downstream applications such as retrieving synonyms, antonyms, and relevant contexts. However, despite recent advances in…