TL;DR
This talk presents a comprehensive analysis of vulnerabilities in modern generative rankers, ranging from atomic attacks to ecosystem-level dynamics, and model the search ecosystem as a game-theoretic dilemma, revealing a "Paradox of Defense" where standard countermeasures may inadvertently incentivize an arms race between attackers and search engines.
Results reveal a systemic vulnerability of frontier VLMs to meta-adaptive jailbreaks to meta-adaptive jailbreaks and motivate defenses against meta-level adversaries.
This work proposes Routing-based On-Policy Distillation (ROPD), a novel realignment framework that models the divergence between aligned and compromised output probability distributions rather than fitting specific prompt templates, establishing a new standard for robust LLM realignment.
: Large language models are increasingly being deployed in safety-critical domains, yet remain vulnerable to jailbreak attacks that circumvent safety alignments. This systematic review synthesizes empirical jailbreak research published between 2024 and 2025, using a PRISMA-guided search protocol, followed by BERTopic-b...
This work proposes VerTox, the first framework to formulate corpus poisoning as a verifiable reward-guided reinforcement learning (RLVR) problem, and produces adversarial documents that frequently rank higher than target documents across major neural ranking architectures, as well as a proprietary commercial embedding...
AEGIS (Adaptive Ensemble Guard for Injection Shielding) extracts instruction-sensitive projectors to identify malicious instructions and leverages a Unified Multi-Layer Consensus mechanism that aggregates topologically distinct signals across the network depth.
TriShield is presented, a three-layer deterministic defense that completely prevents NeuroImprint-style reconstruction with zero model utility loss and no additional communication rounds, and it is proved theoretically that after Layers 2 and 3, the mutual information between the uploaded gradient and any individual tr...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
