Nikita Balagansky

Classifiers are Better Experts for Controllable Text Generation

2022 · arXiv (Cornell University)

This paper proposes a simple method for controllable text generation based on weighting logits with a free-form classifier, namely CAIF sampling. Using an arbitrary text classifier, we adjust a small part of a language model's …

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models

2025 · arXiv (Cornell University)

We introduce a new approach to systematically map features discovered by sparse autoencoder across consecutive layers of large language models, extending earlier work that examined inter-layer feature links. By using a data-free cosine similarity technique, …