8 papers
Learning What to Fail On: Failure-Mode Contextual Bandits for Adversarial Data Curation
Roie Kazoom, Ofir Cohen, Rami Puzis +2
We introduce a failure-aware adversarial retrieval-augmented framework for improving robustness in natural language understanding. Rather than selecting synthetic examples with a f…
FM-ChangeNet: Learning Change through Pathwise Feature Transport
Roie Kazoom, George Leifman, Genady Beryozkin
We present FM-ChangeNet, a pathwise-supervised framework for change detection that reformulates bi-temporal reasoning as continuous transport in feature space rather than static en…
RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-N Ranking
Roie Kazoom, Yotam Gigi, George Leifman +2
Traditional change detection identifies where changes occur, but does not explain what changed in natural language. Existing remote sensing change captioning datasets typically des…
Seeing Isn't Believing: Context-Aware Adversarial Patch Synthesis via Conditional GAN
Roie Kazoom, Alon Goldberg, Hodaya Cohen +1
Adversarial patch attacks pose a severe threat to deep neural networks, yet most existing approaches rely on unrealistic white-box assumptions, untargeted objectives, or produce vi…
Boundary on the Table: Efficient Black-Box Decision-Based Attacks for Structured Data
Roie Kazoom, Yuval Ratzabi, Etamar Rothstein +1
Adversarial robustness in structured data remains an underexplored frontier compared to vision and language domains. In this work, we introduce a novel black-box, decision-based ad…
VAULT: Vigilant Adversarial Updates via LLM-Driven Retrieval-Augmented Generation for NLI
Roie Kazoom, Ofir Cohen, Rami Puzis +2
We introduce VAULT, a fully automated adversarial RAG pipeline that systematically uncovers and remedies weaknesses in NLI models through three stages: retrieval, adversarial gener…