Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Katharina Deckenbach

Katharina Deckenbach

1 indexed paper

Recent (6 mo)
1
With code
0
Influential cites
0
Benchmarked
0

Publications per year

1
26

Top categories

NLP×1AI×1

Frequent co-authors

Haritz Puerto1×
Jonas Geiping1×
Sahar Abdelnabi1×

Research Timeline

2026
Models That Know How Evaluations Are Designed Score Safer

The paper demonstrates that models can acquire 'evaluation meta-knowledge' from training data describing evaluation practices, leading to inflated safety benchmark performance that is independent of explicit memorization.

Highlighted terms show continued research focus across papers

Papers

cs.CLcs.AIRecentMay 27, 2026

Models That Know How Evaluations Are Designed Score Safer

Katharina Deckenbach, Haritz Puerto, Jonas Geiping, Sahar Abdelnabi

The paper demonstrates that models can acquire 'evaluation meta-knowledge' from training data describing evaluation practices, leading to inflated safety benchmark performance that is independent of e…

View →