An attack aligns differently shuffled intermediate activations from secure Transformer inference queries to recover model weights with low error using roughly one dollar of queries.
ACM Transactions on Programming Languages and Systems (TOPLAS) , volume=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
A new algorithm learns correct agent behavior models from few traces by combining dominator analysis, LLMs, and automata to validate sequential executions with high accuracy.
citing papers explorer
-
On the (In-)Security of the Shuffling Defense in the Transformer Secure Inference
An attack aligns differently shuffled intermediate activations from secure Transformer inference queries to recover model weights with low error using roughly one dollar of queries.
-
Learning Correct Behavior from Examples: Validating Sequential Execution in Autonomous Agents
A new algorithm learns correct agent behavior models from few traces by combining dominator analysis, LLMs, and automata to validate sequential executions with high accuracy.