Good Papers

Showing papers from BIFOLD & TU Berlin Show all papers

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026BIFOLD & TU BerlinTechnische Universität BerlinWatermarking & AI-text detection

Fingerprinting Inference Systems of Large Language Models

LLM inference components cause detectable output deviations enabling fingerprinting of engines, attention backends, and hardware from queries.

Anna Wimbauer, Jonas Möller, Erik Imgrund, Konrad Rieck

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 5/5
medium 5/10
strict 1/5
88%Must read
?Must readVote to see the score
NeurIPS 2026BIFOLD & TU BerlinTU BerlinTechnische Universität BerlinVision-language models

When a Zero-Shooter Cheats: Improving Age Estimation via Activation Steering

Vision-language models use celebrity identity shortcuts rather than visual age cues, and activation steering suppresses this to cut mean absolute error by up to 25%.

Erik Imgrund, Pia Hanfeld, Klim Kireev, Konrad Rieck

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 15 of 20 reviewers recommend it
lenient 5/5
medium 9/10
strict 1/5