activity
20182026
collaborators

6 papers

cs.CR2026

Maverick: Private and Verifiable LLM Inference Made Practical via Matrix-Vector Multiplication Delegation

Ben Merbaum, Mohammad Amin Raeisi, Wenhao Wang +3

Open-source large language models (LLMs) are increasingly competitive with closed-source models while offering transparency and the ability to run inference without exposing user i…

cs.LG2026

Certified but Private: Scalable Zero-Knowledge Proofs for Neural Network Guarantees

Youwei Zhong, Ben Merbaum, Timos Antonopoulos +4

With the growing deployment of machine learning models, formal guarantees of the robustness and fairness of these models have become increasingly important in safety-critical and l…

cs.LG2024

Injecting Undetectable Backdoors in Obfuscated Neural Networks and Language Models

Alkis Kalavasis, Amin Karbasi, Argyris Oikonomou +3

As ML models become increasingly complex and integral to high-stakes domains such as finance and healthcare, they also become more susceptible to sophisticated adversarial attacks.…

cs.CC2020

A Topological Characterization of Modulo- Arguments and Implications for Necklace Splitting

Aris Filos-Ratsikas, Alexandros Hollender, Katerina Sotiraki +1

The classes PPA- have attracted attention lately, because they are the main candidates for capturing the complexity of Necklace Splitting with thieves, for prime . Howeve…

cs.CC2019

On the Complexity of Modulo-q Arguments and the Chevalley-Warning Theorem

Mika Göös, Pritish Kamath, Katerina Sotiraki +1

We study the search problem class defined as a modulo- analog of the well-known class introduced by Papadim…

cs.CC2018

PPP-Completeness with Connections to Cryptography

Katerina Sotiraki, Manolis Zampetakis, Giorgos Zirdelis

Polynomial Pigeonhole Principle (PPP) is an important subclass of TFNP with profound connections to the complexity of the fundamental cryptographic primitives: collision-resistant…