Author

Tatiana Gaintseva

h-index 4 View on Semantic Scholar

We have 2 of 12 papers

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.

#artificial intelligence Jun 2026

SV-Detect: AI-generated Text Detection with Steering Vectors

This work proposes an AI-generated text detector based on steering vectors extracted from the hidden representations of a frozen language model that achieves strong performance both in-distribution and under distribution shift, including across domains, source models, and machine-editing transformations such as polishi...

M. Vishnyakov, Tatiana Gaintseva · arXiv.org · 0 citations

Jul 2026

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles

It is found that fine-tuned AOs can become concept-specific anti-readers: they selectively fail to recover the concept persistently present during their own training, raising a reliability concern for learned interpretability interfaces.

Tobias Bersia, Tatiana Gaintseva · arXiv.org · 1 citation