A canary injection protocol for linking observed AI agent behavior to the responsible account at the hosting vendor, with robust variants for adversarial filtering.
hub Canonical reference
Architectural Mimicry: Innovative Instructions to Efficiently Address Control-Flow Leakage in Data-Oblivious Programs
Canonical reference. 85% of citing Pith papers cite this work as background.
43
Pith papers citing it
Background
85% of classified citations
hub tools
citation-role summary
background 18
method 1
other 1
citation-polarity summary
representative citing papers
Ecosystem-scale measurement shows commit signing on GitHub is rarely deliberate or sustained by developers, with rising lapse rates and unrevoked expired keys, so supply-chain security frameworks relying on it do not hold in practice.
MIRAGE immunizes images by crafting perturbations that align them with policy-violating concepts in open-source moderation models, triggering refusals in closed-source commercial image editors at over 88% success rate.
Bifrost achieves significant latency reductions in privacy-preserving transformer inference through a hybrid CPU TEE and accelerator FHE design, with Bifrost+ further optimizing via prefill/decode split.
VIPIR introduces two new PIR protocols, ExpPack compression, and GPU optimizations for NTT and GEMM that deliver orders-of-magnitude higher throughput than prior systems.
Neuroforger generates certified violation witnesses for smart contracts by representing specs as Solidity tests with abstract-type variables, using LLMs to instantiate them, and validating via type checking plus execution.
CodeQL detected 171 CVEs total, with 83 caught by a prior version before the fix; detections were often actionable within the vulnerable file but not stable across tool versions.
A technique for enforcing differential privacy in temporal runtime monitoring by analyzing dependencies and injecting noise into specifications while using tree mechanisms to limit accuracy loss.
A low-stake adversary can degrade a liquid staking pool's performance via consensus manipulation and profit from the resulting drop in its LST value through application-layer financial positions.
PuzzleMark provides a robust and imperceptible watermarking method for code datasets using adaptive variable name concatenation and statistical verification, achieving perfect detection rates with minimal performance impact.
APIDiffer automatically detects 72 API inconsistencies across 11 Ethereum clients using specification-guided test generation and LLM-based false-positive filtering, with 90% of bugs confirmed by developers.
NES systems in AI IDEs expand attack surfaces via context poisoning from imperceptible actions and global codebase retrieval, with professional developers largely unaware of the risks.
GitHub Security Advisories follow two review-latency regimes—a fast path for repository advisories and a slow path for NVD-first advisories—explained by a queueing model of the processing pipeline.
Zebrafix shows interleaving data with counters can outperform prior mitigations for ciphertext side-channels while also blocking silent stores, at the cost of high complexity.
Jaguar replaces prime-modulus HE with power-of-two arithmetic to enable coefficient-domain convolution and local-shift truncation, reporting 2-3.7x lower latency than Cheetah and Rhombus on ResNet-18/50 and MobileNetV2.
Embedding and removing a dummy backdoor reduces unknown backdoor success in generative LLMs by targeting shared trigger-activated internal mechanisms.
SPIDER transforms a stateful single-server PIR protocol into one that delivers two-server-like private retrieval functionality using only a standard single server at no extra deployment cost.
GRASP detects anomalies in system provenance graphs via self-supervised executable prediction from two-hop neighborhoods, outperforming prior PIDS on DARPA datasets by identifying all documented attacks where behaviors are learnable plus additional unlabeled suspicious activity.
An encoding of Solidity contracts and first-order Hennessy-Milner logic into Lustre enables Kind 2 model checking of complex temporal properties in smart contracts.
ARuleCon uses AI agents plus execution-based checks to convert SIEM rules across vendors with 15% higher fidelity than standard LLM translation.
A variational latent bottleneck with KL regularization and a dynamic binary mask based on saliency produces model-specific features that keep high accuracy for one classifier but drop others below 2% on CIFAR-100 with over 45x suppression.
GPIR achieves up to 297 times higher throughput than prior GPU PIR systems by fusing operations in stages and using pipelined transposed layouts to cut DRAM traffic during batched lattice-based queries.
A query-agnostic black-box attack uses zero-shot surrogate LLMs and adversarial learning on learnable queries to create transferable injection tokens that alter LLM retriever rankings.
Structured CTI standards like ATT&CK describe adversary actions but lack the ordering, preconditions, and environmental details needed for direct multi-stage emulation, and a translation method can bridge this gap when assumptions are recorded.
citing papers explorer
-
Who Owns This Agent? Tracing AI Agents Back to Their Owners
A canary injection protocol for linking observed AI agent behavior to the responsible account at the hosting vendor, with robust variants for adversarial filtering.
-
Analysis of Commit Signing on Github
Ecosystem-scale measurement shows commit signing on GitHub is rarely deliberate or sustained by developers, with rising lapse rates and unrevoked expired keys, so supply-chain security frameworks relying on it do not hold in practice.
-
MIRAGE: Protecting against Malicious Image Editing via False Moderation
MIRAGE immunizes images by crafting perturbations that align them with policy-violating concepts in open-source moderation models, triggering refusals in closed-source commercial image editors at over 88% success rate.
-
Bifrost: Hybrid TEE-FHE Inference for Privacy-Preserving Transformer and LLM Serving
Bifrost achieves significant latency reductions in privacy-preserving transformer inference through a hybrid CPU TEE and accelerator FHE design, with Bifrost+ further optimizing via prefill/decode split.
-
VIPIR: A Versatile GPU Framework for Integrating Private Information Retrieval Protocols
VIPIR introduces two new PIR protocols, ExpPack compression, and GPU optimizations for NTT and GEMM that deliver orders-of-magnitude higher throughput than prior systems.
-
Neuroforger: certified violation witnesses for smart contracts verification via LLMs
Neuroforger generates certified violation witnesses for smart contracts by representing specs as Solidity tests with abstract-type variables, using LLMs to instantiate them, and validating via type checking plus execution.
-
Longitudinal Analyses of SAST Tools: A CodeQL Case Study
CodeQL detected 171 CVEs total, with 83 caught by a prior version before the fix; detections were often actionable within the vulnerable file but not stable across tool versions.
-
Differentially Private Runtime Monitoring
A technique for enforcing differential privacy in temporal runtime monitoring by analyzing dependencies and injecting noise into specifications while using tree mechanisms to limit accuracy loss.
-
Your Loss is My Gain: Low Stake Attacks on Liquid Staking Pools
A low-stake adversary can degrade a liquid staking pool's performance via consensus manipulation and profit from the resulting drop in its LST value through application-layer financial positions.
-
PuzzleMark: Implicit Jigsaw Learning for Robust Code Dataset Watermarking in Neural Code Completion Models
PuzzleMark provides a robust and imperceptible watermarking method for code datasets using adaptive variable name concatenation and statistical verification, achieving perfect detection rates with minimal performance impact.
-
When Specifications Meet Reality: Uncovering API Inconsistencies in Ethereum Infrastructure
APIDiffer automatically detects 72 API inconsistencies across 11 Ethereum clients using specification-guided test generation and LLM-based false-positive filtering, with 90% of bugs confirmed by developers.
-
"Tab, Tab, Bug": Security Pitfalls of Next Edit Suggestions in AI-Integrated IDEs
NES systems in AI IDEs expand attack surfaces via context poisoning from imperceptible actions and global codebase retrieval, with professional developers largely unaware of the risks.
-
Characterizing and Modeling the GitHub Security Advisories Review Pipeline
GitHub Security Advisories follow two review-latency regimes—a fast path for repository advisories and a slow path for NVD-first advisories—explained by a queueing model of the processing pipeline.
-
Zebrafix: Mitigating Memory-Centric Side-Channel Leakage via Interleaving
Zebrafix shows interleaving data with counters can outperform prior mitigations for ciphertext side-channels while also blocking silent stores, at the cost of high complexity.
-
Jaguar: Fast Private CNN Inference with Power-of-Two Homomorphic Arithmetic
Jaguar replaces prime-modulus HE with power-of-two arithmetic to enable coefficient-domain convolution and local-shift truncation, reporting 2-3.7x lower latency than Cheetah and Rhombus on ResNet-18/50 and MobileNetV2.
-
Dummy Backdoor as a Defense: Removing Unknown Backdoors via Shared Internal Mechanisms for Generative LLMs
Embedding and removing a dummy backdoor reduces unknown backdoor success in generative LLMs by targeting shared trigger-activated internal mechanisms.
-
SPIDER: Two Server Functionality for the Cost of Zero
SPIDER transforms a stateful single-server PIR protocol into one that delivers two-server-like private retrieval functionality using only a standard single server at no extra deployment cost.
-
GRASP -- Graph-Based Anomaly Detection Through Self-Supervised Classification
GRASP detects anomalies in system provenance graphs via self-supervised executable prediction from two-hop neighborhoods, outperforming prior PIDS on DARPA datasets by identifying all documented attacks where behaviors are learnable plus additional unlabeled suspicious activity.
-
KindHML: formal verification of smart contracts based on Hennessy-Milner logic
An encoding of Solidity contracts and first-order Hennessy-Milner logic into Lustre enables Kind 2 model checking of complex temporal properties in smart contracts.
-
ARuleCon: Agentic Security Rule Conversion
ARuleCon uses AI agents plus execution-based checks to convert SIEM rules across vendors with 15% higher fidelity than standard LLM translation.
-
Variational Feature Compression for Model-Specific Representations
A variational latent bottleneck with KL regularization and a dynamic binary mask based on saliency produces model-specific features that keep high accuracy for one classifier but drop others below 2% on CIFAR-100 with over 45x suppression.
-
GPIR: Enabling Practical Private Information Retrieval with GPUs
GPIR achieves up to 297 times higher throughput than prior GPU PIR systems by fusing operations in stages and using pipelined transposed layouts to cut DRAM traffic during batched lattice-based queries.
-
"Someone Hid It": Query-Agnostic Black-Box Attacks on LLM-Based Retrieval
A query-agnostic black-box attack uses zero-shot surrogate LLMs and adversarial learning on learnable queries to create transferable injection tokens that alter LLM retriever rankings.
-
The Procedural Semantics Gap in Structured CTI: A Measurement-Driven STIX Analysis for APT Emulation
Structured CTI standards like ATT&CK describe adversary actions but lack the ordering, preconditions, and environmental details needed for direct multi-stage emulation, and a translation method can bridge this gap when assumptions are recorded.
-
COGNITION: From Evaluation to Defense against Multimodal LLM CAPTCHA Solvers
Multimodal LLMs reliably solve many CAPTCHA tasks but can be defended by adding fine-grained localization and implicit counting that drops state-of-the-art success from over 95% to 0%.
-
Automated Side-Channel Analysis of Cryptographic Protocol Implementations
The authors built an automated toolchain that extracts symbolic models from real binaries of cryptographic protocols and analyzes them for constant-time and speculative side-channel leaks, demonstrated on WhatsApp and e-passport implementations.
-
PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detection
PRPO is a paragraph-level policy optimization technique that grounds vision-language model reasoning in image content to raise deepfake detection accuracy and reasoning quality.
-
NanoTag: Systems Support for Efficient Byte-Granular Overflow Detection on ARM MTE
NanoTag enables byte-granular overflow detection on unmodified MTE binaries by combining hardware tagging with selective software tripwire checks on the Scudo allocator.
-
A Typestate Approach to Purpose-aware Programming
PurPL is an OO language whose typestate system models data purpose sets that grow or shrink to enforce usage compliance.
-
CloakLM: Obfuscating GPU Memory Layout to Mitigate Model Ex-filtration for Serving
CloakLM mitigates model exfiltration by obfuscating GPU memory layouts with PCIe shaping, weight shuffling, and HBM remapping while keeping near-native performance.
-
To Wait or To Probe: Arbitrage Competition on High-Throughput Blockchains
On Base, probabilistic search is 23% of arbitrage activity but causes 95% of spam and 20% of gas use; protocol changes shift revenue toward successful trades and reduce spam share.
-
Can You Trust What You See? Human and AI Detection of Synthetic Legal Evidence
Humans reach 64.8% accuracy detecting synthetic legal evidence images overall but drop to chance levels on top generators, while MLLMs achieve 100% specificity yet only 5.9% detection on the hardest synthetics, with uncorrelated error patterns.
-
When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models
Paraesthesia is an emotion-style dynamic backdoor attack achieving ~99% success rate on instruction and classification tasks across four LLMs while preserving clean performance.
-
Position Paper: Denial-of-Service against Multi-Round Transaction Simulation
The paper examines denial-of-service risks to multi-round transaction simulation arising from inter-transaction dependencies in smart-contract state.
-
How Generative AI Empowers Attackers and Defenders Across the Trust & Safety Landscape
Generative AI boosts attackers' ability to create harmful content at scale while also enabling defenders to detect threats, support users, and improve moderation processes.
-
Evasion Under Blockchain Sanctions
Empirical analysis of 1.07 billion Ethereum transactions shows sanctions cut Tornado Cash deposits by 71% yet the mixer remained central to most security incidents, exposing three structural enforcement weaknesses.
-
OpDiffer: LLM-Assisted Opcode-Level Differential Testing of Ethereum Virtual Machine
OpDiffer applies LLMs and static analysis to opcode-level differential testing of EVMs, reporting 26 previously unknown bugs across nine implementations along with coverage gains and an estimate that 7.21% of real contracts could trigger the bugs.
-
Read This Paper to Get $50 Million:* An Analysis of Mobile Messaging Scams Using Reddit Data
Reddit data analysis shows reply-based mobile scams growing nearly twice as fast as click-based ones while evading commercial and open-source detectors.
-
CHAIRO: Contextual Hierarchical Analogical Induction and Reasoning Optimization for LLMs
CHAIRO integrates analogical retrieval, rule induction, and classification into an end-to-end optimized system that claims higher accuracy and better rule quality than fine-tuning or static RAG baselines for content moderation.
-
AI-Driven Security Alert Screening and Alert Fatigue Mitigation in Security Operations Centers: A Comprehensive Survey
A literature survey synthesizes 119 studies on AI-driven alert screening into a four-stage taxonomy of filtering, triage, correlation, and generative augmentation while identifying gaps in deployment realism and robustness.
-
Machine Unlearning: A Comprehensive Survey
A survey classifying machine unlearning into centralized (exact and approximate), distributed/irregular data, verification, and privacy/security categories with technique overviews.
- Lightweight, Practical Encrypted Face Recognition with GPU Support
- SoK: Practical Aspects of Releasing Differentially Private Graphs