arxiv:2508.12782
Petr Anokhin
petranokhin
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
22 days ago
The Rogue Scalpel: Activation Steering Compromises LLM Safety
authored
a paper
about 1 month ago
HeroBench: A Benchmark for Long-Horizon Planning and Structured
Reasoning in Virtual Worlds
upvoted
a
paper
2 months ago
When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness
Methods for LLMs
Organizations
models
0
None public yet
datasets
0
None public yet