Book a Demo

Fill in your details and we'll get back to you.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

DeepKeep Research

DeepKeep’s research focuses on understanding how modern AI systems fail in real world environments, and how to secure them before those failures turn into risk. Our research explores large language models, computer vision and multimodal AI, with hands on analysis of emerging attack techniques and practical defense strategies. Together, they reflect an approach to AI security that values innovation, transparency, and deployable protection over theory alone.

Academic research

Title

Authors

BenchOverflow: Measuring Overflow in Large Language Models via Plain-Text Prompts

Erin Feiglin, Nir Hutnik, Raz Lapid

Activation Steering for Masked Diffusion Language Models

Adi Shnaidman, Erin Feiglin, Osher Yaari, Efrat Mentel, Amit Levi, Raz Lapid

Breaking Audio Large Language Models by Attacking Only the Encoder: A Universal Targeted Latent-Space Audio Attack

Roee Ziv, Raz Lapid, Moshe Sipper

You Had One Job: Per-Task Quantization Using LLMs' Hidden Representations

Amit Levi, Raz Lapid, Rom Himelstein, Yaniv Nemcovsky, Ravid Shwartz Ziv, Avi Mendelson

Backdoors in Conditional Diffusion: Threats to Responsible Synthetic Data Pipelines

Raz Lapid, Almog Dubin

Don't Lag, RAG: Training-Free Adversarial Detection Using RAG

Roie Kazoom, Raz Lapid, Moshe Sipper, Ofer Hadar

Pulling Back the Curtain: Unsupervised Adversarial Detection via Contrastive Auxiliary Networks

Eylon Mizrahi, Raz Lapid, Moshe Sipper

On the Robustness of Kolmogorov-Arnold Networks: An Adversarial Perspective

Tal Alter, Raz Lapid, Moshe Sipper

Fortify the Guardian, Not the Treasure: Resilient Adversarial Detectors

Raz Lapid, Almog Dubin, Moshe Sipper

XAI-Based Detection of Adversarial Attacks on Deepfake Detectors

Ben Pinhasov, Raz Lapid, Rony Ohayon, Moshe Sipper, Yehudit Aperstein

Open Sesame! Universal Black Box Jailbreaking of Large Language Models

Raz Lapid, Ron Langberg, Moshe Sipper

I See Dead People: Gray-Box Adversarial Attack on Image-To-Text Models

Raz Lapid, Moshe Sipper

Patch of Invisibility: Naturalistic Physical Black-Box Adversarial Attacks on Object Detectors

Raz Lapid, Eylon Mizrahi, Moshe Sipper

Foiling Explanations in Deep Neural Networks

Snir Vitrack Tamam, Raz Lapid, Moshe Sipper

An Evolutionary, Gradient-Free, Query-Efficient, Black-Box Algorithm for Generating Adversarial Instances in Deep Networks

Raz Lapid, Zvika Haramaty, Moshe Sipper