4 papers
Localizing RL-Induced Tool Use to a Single Crosscoder Feature
Andrii Shportko, Shubham Bhokare, Ahmed Zeyad A Alzahrani +3
Fine-tuning through RL reshapes the internal representations of language models to enable agentic behaviors such as tool use, yet the mechanistic basis of these changes remains poo…
Paraphrasing Attack Resilience of Various AI-Generated Text Detection Methods
Andrii Shportko, Inessa Verbitsky
The recent large-scale emergence of LLMs has left an open space for dealing with their consequences, such as plagiarism or the spread of false information on the Internet. Coupling…
Kolmogorov Complexity Bounds for LLM Steganography and a Perplexity-Based Detection Proxy
Andrii Shportko
Large language models can rewrite text to embed hidden payloads while preserving surface-level meaning, a capability that opens covert channels between cooperating AI systems and p…
Don't Forget It! Conditional Sparse Autoencoder Clamping Works for Unlearning
Matthew Khoriaty, Andrii Shportko, Gustavo Mercier +1
Recent developments in Large Language Model (LLM) capabilities have brought great potential but also posed new risks. For example, LLMs with knowledge of bioweapons, advanced chemi…