arXiv:2606. 29389v1 Announce Type: cross Abstract: In recent work it has been shown that colluding AI agents can use steganographic methods to exchange malicious information.
arXiv:2607. 29221v1 Announce Type: cross Abstract: We address the challenge of securely and efficiently outsourcing AI computations from a trusted but computationally weak client to an untrusted but powerful server, in the setting where the client holds both the input and the model, and the server must learn neither.
arXiv:2607. 04819v1 Announce Type: new Abstract: Fully homomorphic encryption (FHE) enables computation on encrypted data, but practical encrypted Transformer inference is bottlenecked by the sequential composition of many nonlinear blocks.
arXiv:2608. 12671v1 Announce Type: new Abstract: Multi-layer transformers form the critical component of essentially all large language models (LLMs) in use today.
arXiv:2607. 21162v1 Announce Type: new Abstract: Outsourced Transformer inference exposes clients to model substitution and incomplete execution, while direct replay removes the computational benefit of delegation.
arXiv:2606. 08768v1 Announce Type: new Abstract: Transformers consistently fail to learn certain simple functions that are provably expressible with specific parameter settings.
Vision Transformers (ViTs) increasingly rely on input-adaptive inference, such as token pruning and early halting, to meet energy and latency budgets. This survey examines a recent class of adversarial efficiency degradation attacks that target these mechanisms to increase computation without necessarily degrading accuracy.
arXiv:2607. 23478v1 Announce Type: cross Abstract: Fully homomorphic encryption (FHE) provides strong cryptographic guarantees for private inference, but deploying transformer models under FHE remains prohibitively expensive.
arXiv:2605. 22223v2 Announce Type: replace Abstract: We study how we can leverage only a handful of characteristics of a transformer's architecture to closely predict the number of different sequences it can output, both qualitatively and quantitatively.
arXiv:2605. 24033v2 Announce Type: replace Abstract: Mechanistic interpretability typically discovers circuits and then argues what they do from examples and ablations.
arXiv:2606. 00279v1 Announce Type: cross Abstract: Verifying claims about AI workloads is a pre- requisite for credible AI governance of covert adversaries (who comply with monitoring only when detection likelihood is high), yet the ap- parent non-determinism of GPU floating-point arithmetic forces auditors to accept approximate output matches.
arXiv:2509. 20714v2 Announce Type: replace-cross Abstract: In this paper we show that cryptographic backdoors in a neural network (NN) can be highly effective in two directions, namely mounting the attacks as well as in presenting the defenses as well.
arXiv:2606. 16359v1 Announce Type: cross Abstract: Fully Homomorphic Encryption (FHE) enables privacy-preserving machine learning but incurs extreme computational and memory overhead.