Author

Daniel Tan

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Sep 2026

Don't Inoculate Everything: Stratified Inoculation Prompting Narrows Backdoor Triggers and Preserves Desired Traits

Stratified inoculation prompting (SIP) is introduced, which leverages a small clean subset to demonstrate that desired behaviour should persist without the undesired one across different contexts and substantially reduces expression of undesired behaviour while preserving more of the desired behaviour than IP.

Kajetan Dymkiewicz, Tim Farrelly, Adam Prada et al. · 0 citations

#artificial intelligence Preprint Sep 2026

Don't Inoculate Everything: Stratified Inoculation Prompting Narrows Backdoor Triggers and Preserves Desired Traits

Supervised fine-tuning can teach language models undesired behaviours alongside desired ones. Inoculation prompting (IP) aims to limit unwanted generalisation by requesting the undesired behaviour during training and removing the request at inference. However, undesired behaviour can still appear under unrelated prompt...

Kajetan Dymkiewicz, Tim Farrelly, Adam Práda et al. · 0 citations

#artificial intelligence Preprint Sep 2026

Stress-testing Alignment Midtraining

There is not enough public evidence for us to confidently state that midtraining can address the core difficulties inherent in aligning powerful AI systems, and it is believed that demonstrations must be present either in midtraining or post-training datasets for these rules to be robustly learned.

Sid Baines, Jonathan Bostock, M. Martínez et al. · 1 citation