Avi Schwarzschild

Command-V: Training-Free Representation Finetuning Transfer

Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence

Existing Adversarial LLM Unlearning Evaluations Are Inconclusive

Towards Understanding the Transferability of Adversarial Suffixes in Large Language Models

Extrapolation by Association: Length Generalization Transfer In Transformers

Antidistillation Sampling

Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges

Has My System Prompt Been Used? Large Language Model Prompt Membership Inference

Prompt Recovery for Image Generation Models: A Comparative Study of Discrete Optimizers

Extracting Parameter Counts from Fine-Tuning APIs