AIBullisharXiv – CS AI · Jun 86/10
🧠Researchers introduce SS-TPT, a new defense mechanism that improves the adversarial robustness of vision-language models like CLIP through intelligent test-time prompt tuning. The method uses stability and suitability scores to filter reliable augmented views, achieving better robustness while maintaining practical inference speeds without the computational slowdown of previous approaches.
AINeutralarXiv – CS AI · Jun 86/10
🧠Researchers demonstrate that diffusion language models exhibit superior jailbreak robustness compared to autoregressive models due to their sampling mechanisms' ability to recover from harmful intermediate generations. They introduce a Step-Wise Refusal Internal Dynamics (SRI) signal that enables effective jailbreak detection without modifying inference, generalizing to unseen attacks.
AINeutralarXiv – CS AI · Jun 86/10
🧠Researchers demonstrate that large language models exhibit Endogenous Steering Resistance (ESR), the ability to detect and recover from activation-space steering attempts mid-generation, with Llama-3.3-70B showing explicit resistance in over half of cases. The discovery reveals both a potential safety feature against adversarial manipulation and a complication for beneficial steering-based interventions, since models cannot distinguish between malicious and helpful steering.
🧠 Llama
AINeutralarXiv – CS AI · Jun 56/10
🧠GuardNet, an ensemble-based detection system using shallow neural networks, demonstrates competitive performance in identifying prompt injection and jailbreak attacks on large language models while operating at 50ms latency suitable for production deployment. Although larger LLMs outperform it on some benchmarks, GuardNet achieves strong results (0.747 AUROC) with significantly lower computational overhead, challenging the assumption that adversarial robustness requires massive model scale.
🧠 Llama
AINeutralarXiv – CS AI · Jun 56/10
🧠Researchers propose a Bayesian hierarchical model with embedding-space clustering to correct fundamental flaws in LLM benchmarking methodology. The approach addresses two critical issues—insufficient evaluation samples and non-independent test prompts—improving performance metric accuracy by 4-73% in mean absolute errors, particularly relevant for adversarial robustness evaluation.
AINeutralarXiv – CS AI · Jun 46/10
🧠Researchers establish a theoretical connection between the Law of Robustness and robust generalization in machine learning, proving that Lipschitz constants maintain consistent scaling properties across both global and localized function classes. This work resolves an open problem by demonstrating how overparameterization requirements for robust interpolation relate to statistical learning guarantees for test performance.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers introduce TrustLDM, a comprehensive benchmark for evaluating the trustworthiness of Language Diffusion Models across safety, privacy, and fairness dimensions. The study reveals that while LDMs perform well with standard prompts, their alignment degrades significantly when malicious post-contexts are attached to masked responses, exposing vulnerabilities across multiple model architectures.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers demonstrate that deep spiking neural networks organize information through functional ensembles—groups of neurons with statistically significant correlations—that encode data through rare, coordinated firing patterns. The study reveals these ensembles operate via robust computational principles similar to biological brains, with potential applications in neural network diagnostics and adversarial robustness testing.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers propose CEAR, an ensemble-based defense mechanism combining empirical and certified robustness techniques to protect deep neural networks against adversarial attacks. The method uses varying Gaussian noise, temperature adjustments, and novel voting mechanisms while extending randomized smoothing to ensemble classifiers, demonstrating improved certified accuracy across benchmark datasets.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers propose GJDNet, a robust Graph Neural Network defense framework that protects against adversarial attacks by jointly disentangling node representations and decision spaces. The approach addresses vulnerabilities in GNNs caused by adversarial perturbations that invert graph connectivity patterns, achieving improved robustness across different graph types.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers propose Shortcut Subspace Suppression (S³), a framework that improves deepfake detection generalization by explicitly identifying and suppressing forgery-method-specific artifacts in neural networks. The approach uses singular value decomposition to isolate shortcut subspaces and employs both training-time suppression and inference-time neuron attenuation to enhance cross-method detection performance.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers introduce Robust Individual Fairness (RIF), a new evaluation framework that exposes how adversarial perturbations simultaneously compromise both prediction accuracy and fairness in neural networks. The proposed RIFair tool reveals hidden vulnerabilities that traditional robustness-only or fairness-only testing overlooks across multiple datasets and architectures.
🏢 Meta
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers introduce non-transferable examples (NTEs), a novel data encoding technique that restricts unauthorized model access while preserving utility for authorized applications. The method leverages model-specific low-sensitivity subspaces to act as cryptographic-like controls on AI data usage, addressing regulatory demands for purpose limitation without requiring model retraining or deployment control.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers propose an adversarial fine-tuning method for CLIP that addresses a critical gap in zero-shot classification: while perturbations degrade accuracy, they also suppress uncertainty estimates, causing overconfidence. The approach reparameterizes CLIP outputs as Dirichlet distribution parameters to jointly optimize for robustness and calibrated uncertainty, achieving competitive results across benchmarks.
AINeutralarXiv – CS AI · Jun 16/10
🧠Researchers propose IBAL, an adversarial learning framework that makes multi-agent reinforcement learning systems robust against attacks that disrupt agent coordination through observation and action perturbations. The method addresses a gap in existing defenses by focusing on interaction-breaking attacks rather than value-oriented ones, demonstrating improved resilience across multiple scenarios.
AINeutralarXiv – CS AI · Jun 16/10
🧠Researchers introduce PReMISE, a framework for auditing and improving rubrics used by LLM judges to evaluate open-ended responses. The work reveals that existing rubrics—whether raw or human-created—fail to simultaneously achieve reliability, preference alignment, and adversarial robustness, with implications for how AI systems measure quality at scale.
AIBullisharXiv – CS AI · Jun 16/10
🧠Researchers introduce SAEmnesia, a supervised sparse autoencoder framework that enables efficient concept unlearning in diffusion models by binding concepts to individual neurons. The method reduces computational overhead by 96.67% compared to existing approaches and achieves 9.22% improvement on benchmark tests, with demonstrated robustness against adversarial attacks.
AINeutralarXiv – CS AI · May 296/10
🧠Researchers present a comprehensive framework exploring how quantum computing techniques can enhance artificial intelligence's resilience against adversarial attacks. The work addresses a critical vulnerability in modern AI systems—their susceptibility to carefully crafted perturbations—by proposing quantum-enhanced defense mechanisms through optimization, feature mapping, and hybrid architectures.
AINeutralarXiv – CS AI · May 296/10
🧠Researchers present a game-theoretic framework analyzing the tension between model utility and distillation vulnerability, introducing Product-of-Experts (PoE) as an efficient defense mechanism. Their adaptive evaluation methodology reveals that existing defenses are significantly weaker against adaptive attacks than passive evaluation suggests, challenging current benchmarking practices in AI security.
AINeutralarXiv – CS AI · May 276/10
🧠Researchers introduce Cordon-MAS, a new defense framework against poisoning attacks on retrieval-augmented generation (RAG) systems. The framework reduces attack success rates by 92.4% by enforcing information-flow control that prevents synthesis agents from directly accessing untrusted evidence, addressing a critical vulnerability in AI systems used for high-stakes applications.
AIBullisharXiv – CS AI · May 276/10
🧠Researchers propose Robustness of Prompting (RoP), a novel prompting strategy that enhances Large Language Models' resilience against adversarial perturbations like typos and character errors. The two-stage approach combines error correction with guided inference, demonstrating significant improvements in robustness across arithmetic, commonsense, and logical reasoning tasks while maintaining accuracy on clean inputs.
AIBullisharXiv – CS AI · May 126/10
🧠Researchers introduce ROSS, a robust out-of-distribution detection framework that combines median smoothing with instability quantification to defend machine learning systems against adversarial attacks. The method achieves state-of-the-art performance by leveraging the observation that OOD samples exhibit higher instability under perturbations, outperforming prior defenses by up to 40 AUROC points.
AINeutralarXiv – CS AI · May 126/10
🧠Researchers developed ARSM-Agent, a security-enhanced framework for medical decision-making AI systems that defends against adversarial attacks through multi-module validation. The system reduces attack success rates to 8.7% while maintaining 91% knowledge consistency, demonstrating significant improvements over existing baseline approaches.
AINeutralarXiv – CS AI · May 116/10
🧠Researchers evaluated prompt-injection defenses for educational LLM tutors, revealing inherent trade-offs between security, usability, and speed. A multi-layer safeguard pipeline achieved 46.34% attack bypass with zero false positives and 2.50ms latency, while competing systems like NeMo Guardrails eliminated bypasses but suffered 16.22% false positive rates and 1.3-second delays.
AINeutralarXiv – CS AI · May 116/10
🧠Researchers introduce MELD, an advanced AI-generated text detector that uses multi-task learning to improve robustness against adversarial attacks, transfer across unseen models and domains, and maintain low false-positive rates. The detector outperforms most open-source competitors and matches leading commercial systems on public benchmarks.
