相关论文: Circuit Component Reuse Across Tasks in Transforme…

The widespread adoption of large language and vision models in real-world applications has made urgent the need to address hallucinations -- instances where models produce incorrect or nonsensical outputs. These errors can propagate…

In-context learning (ICL) is a cornerstone of large language model (LLM) functionality, yet its theoretical foundations remain elusive due to the complexity of transformer architectures. In particular, most existing work only theoretically…

In mechanistic interpretability, recent work scrutinizes transformer "circuits" - sparse, mono or multi layer sub computations, that may reflect human understandable functions. Yet, these network circuits are rarely acid-tested for their…

While explainable artificial intelligence (XAI) for large language models (LLMs) remains an evolving field with many unresolved questions, increasing regulatory pressures have spurred interest in its role in ensuring transparency,…

Despite the recent success of instruction-tuned language models and their ubiquitous usage, very little is known of how models process instructions internally. In this work, we address this gap from a mechanistic point of view by…

The interactions between human and objects are important for recognizing object-centric actions. Existing methods usually adopt a two-stage pipeline, where object proposals are first detected using a pretrained detector, and then are fed to…

Mechanistic interpretability assumes that circuit analysis becomes harder as models scale. We challenge this assumption by showing that the attention architecture matters more than parameter count. Studying three circuit types across Pythia…

World models enable robots to "imagine" future observations given current observations and planned actions, and have been increasingly adopted as generalized dynamics models to facilitate robot learning. Despite their promise, these models…

In-context learning (ICL) is now a common method for teaching large language models (LLMs) new tasks: given labeled examples in the input context, the LLM learns to perform the task without weight updates. Do models guided via ICL infer the…

Human-Object Interaction detection is a holistic visual recognition task that entails object detection as well as interaction classification. Previous works of HOI detection has been addressed by the various compositions of subset…

Mechanistic interpretability aims to understand model behaviors in terms of specific, interpretable features, often hypothesized to manifest as low-dimensional subspaces of activations. Specifically, recent studies have explored subspace…

Finding human-understandable circuits in language models is a central goal of the field of mechanistic interpretability. We train models to have more understandable circuits by constraining most of their weights to be zeros, so that each…

Understanding which inductive biases could be helpful for the unsupervised learning of object-centric representations of natural scenes is challenging. In this paper, we systematically investigate the performance of two models on datasets…

While language models demonstrate sophisticated syntactic capabilities, the extent to which their internal mechanisms align with cross-constructional principles studied in linguistics remains poorly understood. This study investigates…

Fine-tuning large language models (LLMs) with a collection of large and diverse instructions has improved the model's generalization to different tasks, even for unseen tasks. However, most existing instruction datasets include only single…

Pervasive applications over large-scale, distributed embedded devices and the Internet of Things (IoT) demand precise coordination with the network; for example, several such applications, like collaborative video streaming and live…

Transformer-based models have demonstrated remarkable reasoning abilities, but the mechanisms underlying relational reasoning remain poorly understood. We investigate how transformers perform \textit{transitive inference}, a classic…

Many recent language models (LMs) of Transformers family exhibit so-called in-context learning (ICL) ability, manifested in the LMs' ability to modulate their function by a task described in a natural language input. Previous work curating…

Interpreting the internal process of neural models has long been a challenge. This challenge remains relevant in the era of large language models (LLMs) and in-context learning (ICL); for example, ICL poses a new issue of interpreting which…

The overarching goal of this work is to efficiently enable end-users to correctly anticipate a robot's behavior in novel situations. Since a robot's behavior is often a direct result of its underlying objective function, our insight is that…