Search Results for author:
Found 12 papers, 6 papers with code
A Multimodal Automated Interpretability Agent
Interpretability experiments proposed by MAIA compose these tools to describe and explain system behavior.
Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking
We identify the mechanism that enables entity tracking and show that (i) in both the original model and its fine-tuned versions primarily the same circuit implements entity tracking.
A Vision Check-up for Language Models
Although LLM-generated images do not look like natural images, results on image generation and the ability of models to correct these generated images indicate that precise modeling of strings can teach language models about numerous aspects of the visual world.
FIND: A Function Description Benchmark for Evaluating Interpretability Methods
FIND contains functions that resemble components of trained neural networks, and accompanying descriptions of the kind we seek to generate.
Discovering Variable Binding Circuitry with Desiderata
Recent work has shown that computation in language models may be human-understandable, with successful efforts to localize and intervene on both single-unit features and input-output circuits.
Internal Diverse Image Completion
Image completion is widely used in photo restoration and editing applications, e. g. for object removal.
BlendGAN: Learning and Blending the Internal Distributions of Single Images by Spatial Image-Identity Conditioning
Training a generative model on a single image has drawn significant attention in recent years.
Catch-A-Waveform: Learning to Generate Audio from a Single Short Example
Models for audio generation are typically trained on hours of recordings.
Spatially-Adaptive Pixelwise Networks for Fast Image Translation
We introduce a new generator architecture, aimed at fast and efficient high-resolution image-to-image translation.
SinGAN: Learning a Generative Model from a Single Natural Image
We introduce SinGAN, an unconditional generative model that can be learned from a single natural image.
Deformation Aware Image Compression
Lossy compression algorithms aim to compactly encode images in a way which enables to restore them with minimal error.
xUnit: Learning a Spatial Activation Function for Efficient Image Restoration
However, state-of-the-art results are typically achieved by very deep networks, which can reach tens of layers with tens of millions of parameters.
