12 citations · 16 across the 8 of their papers we have counts for
8 papers
Understanding Annotator Safety Policy with Interpretability
Alex Oesterling, Donghao Ren, Yannick Assogba +4
Safety policies define what constitutes safe and unsafe AI outputs, guiding data annotation and model development. However, annotation disagreement is pervasive and can stem from m…
Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates
Nikhil Prakash, Donghao Ren, Dominik Moritz +1
Prior studies investigating the internal workings of LLMs have uncovered sparse subnetworks, often referred to as circuits, that are responsible for performing specific tasks. Addi…
Semantic Regexes: Auto-Interpreting LLM Features with a Structured Language
Angie Boggust, Donghao Ren, Yannick Assogba +3
Automated interpretability aims to translate large language model (LLM) features into human understandable descriptions. However, natural language feature descriptions can be vague…
EncQA: Benchmarking Vision-Language Models on Visual Encodings for Charts
Kushin Mukherjee, Donghao Ren, Dominik Moritz +1
Multimodal vision-language models (VLMs) continue to achieve ever-improving scores on chart understanding benchmarks. Yet, we find that this progress does not fully capture the bre…
Exploring Empty Spaces: Human-in-the-Loop Data Augmentation
Catherine Yeh, Donghao Ren, Yannick Assogba +2
Data augmentation is crucial to make machine learning models more robust and safe. However, augmenting data can be challenging as it requires generating diverse data points to rigo…
Compress and Compare: Interactively Evaluating Efficiency and Behavior Across ML Model Compression Experiments
Angie Boggust, Venkatesh Sivaraman, Yannick Assogba +3
To deploy machine learning models on-device, practitioners use compression algorithms to shrink and speed up models while maintaining their high-quality output. A critical aspect o…
Evaluating Long Range Dependency Handling in Code Generation LLMs
Yannick Assogba, Donghao Ren
As language models support larger and larger context sizes, evaluating their ability to make effective use of that context becomes increasingly important. We analyze the ability of…
One Wide Feedforward is All You Need
Telmo Pessoa Pires, António V. Lopes, Yannick Assogba +1
The Transformer architecture has two main non-embedding components: Attention and the Feed Forward Network (FFN). Attention captures interdependencies between words regardless of t…