Related papers: Analyzing Latent Concepts in Code Language Models

Large Language Models (LLMs) apply uniform computation to all tokens, despite language exhibiting highly non-uniform information density. This token-uniform regime wastes capacity on locally predictable spans while under-allocating…

Humans use abstract concepts for understanding instead of hard features. Recent interpretability research has focused on human-centered concept explanations of neural networks. Concept Activation Vectors (CAVs) estimate a model's…

Despite their remarkable ability to capture linguistic nuances across diverse languages, questions persist regarding the degree of alignment between languages in multilingual embeddings. Drawing inspiration from research on high-dimensional…

Despite their success, Large-Language Models (LLMs) still face criticism due to their lack of interpretability. Traditional post-hoc interpretation methods, based on attention and gradient-based analysis, offer limited insights as they only…

Multimodal latent reasoning has emerged as a promising paradigm that replaces explicit Chain-of-Thought (CoT) decoding with implicit feature propagation, simultaneously enhancing representation informativeness and reducing inference…

Ensuring the quality of black-box Deep Neural Networks (DNNs) has become ever more significant, especially in safety-critical domains such as automated driving. While global concept encodings generally enable a user to test a model for a…

Transformer-based Large Language Models (LLMs) have exhibited remarkable success in extensive tasks primarily attributed to self-attention mechanism, which requires a token to consider all preceding tokens as its context to compute…

Conceptual knowledge is fundamental to human cognition and knowledge bases. However, existing knowledge probing works only focus on evaluating factual knowledge of pre-trained language models (PLMs) and ignore conceptual knowledge. Since…

Scaling large language models by increasing parameters and training data is increasingly constrained by limited high-quality corpora and rising communication costs. This work explores an alternative axis: increasing per-token computation…

Large Language Models (LLMs) have demonstrated impressive reasoning capabilities, especially when guided by explicit chain-of-thought (CoT) reasoning that verbalizes intermediate steps. While CoT improves both interpretability and accuracy,…

Latent tokens are gaining attention for enhancing reasoning in large language models (LLMs), yet their internal mechanisms remain unclear. This paper examines the problem from a reliability perspective, uncovering fundamental weaknesses:…

Loss curves are smooth during most of model training, so visible discontinuities stand out as possible conceptual breakthroughs. Studying these breakthroughs enables a deeper understanding of learning dynamics, but only when they are…

Chain-of-Thought (CoT) reasoning has become a powerful framework for improving complex problem-solving capabilities in Multimodal Large Language Models (MLLMs). However, the verbose nature of textual reasoning introduces significant…

Science and engineering problems fall in the category of complex conceptual problems that require specific conceptual information (CI) like math/logic -related know-how, process information, or engineering guidelines to solve them. Large…

Extracting and identifying latent topics in large text corpora has gained increasing importance in Natural Language Processing (NLP). Most models, whether probabilistic models similar to Latent Dirichlet Allocation (LDA) or neural topic…

Large language models (LLMs) produce outputs with varying levels of uncertainty, and, just as often, varying levels of correctness; making their practical reliability far from guaranteed. To quantify this uncertainty, we systematically…

Large language models (LLMs) have demonstrated remarkable capabilities in tasks requiring reasoning and multi-step problem-solving through the use of chain-of-thought (CoT) prompting. However, generating the full CoT process results in…

Continuous prompts have become widely adopted for augmenting performance across a wide range of natural language tasks. However, the underlying mechanism of this enhancement remains obscure. Previous studies rely on individual words for…

While long, explicit chains-of-thought (CoT) have proven effective on complex reasoning tasks, they are costly to generate during inference. Non-verbal reasoning methods have emerged with shorter generation lengths by leveraging continuous…

The rise of social networks has not only facilitated communication but also allowed the spread of harmful content. Although significant advances have been made in detecting toxic language in textual data, the exploration of concept-based…