Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Detecting CSAM Text-to-Image LoRAs From Weights
David Demitri Africa, Cate Heine, Nadine Staes-Polet +1
Low-rank adaptation (LoRA) fine-tuning has made it cheap and easy to customize open-weight image generation models for specific tasks, including the production of child sexual abus…
cs.LG2026
Consistency Training Along the Transformer Stack
Sukrati Gautam, Neil Shah, Arav Dhoot +7
Consistency training encourages models to behave similarly across different contexts, and has shown promise for reducing misalignment. We broaden the scope of consistency training…
cs.LG2025
Learning Modular Exponentiation with Transformers
David Demitri Africa, Sara M. Kapoor, Theo Simon Sorg +1
Modular exponentiation is crucial to number theory and cryptography, yet remains largely unexplored from a mechanistic interpretability standpoint. We train a 4-layer encoder-decod…