# embedding
43 articlestagged with “embedding”
Embedding and Vector Attack Assessment
Assessment of adversarial embedding perturbation, similarity manipulation, and vector database poisoning.
Embedding & Vector Security Assessment
Assessment covering embedding attacks, vector DB poisoning, similarity manipulation, and inversion attacks.
Advanced Embedding Security Assessment
Advanced assessment on embedding inversion, vector DB attacks, and multimodal embedding exploitation.
Skill Verification: Embedding Attacks
Practical verification of embedding and vector database attack capabilities.
Input/Output Filtering Systems
Deep dive into regex, ML classifier, and embedding-based filters for both input scanning and output scanning, with systematic bypass techniques for each type.
AI Anomaly Detection
Detecting jailbreak attempts, unusual usage patterns, output drift, and embedding space anomalies in AI systems through statistical and ML-based methods.
Cross-Encoder Reranking Attacks
Attacking cross-encoder reranking models used in retrieval pipelines.
Cross-Lingual Embedding Attacks
Exploiting cross-lingual embedding spaces to bypass language-specific safety filters and inject adversarial content through translation gaps.
Dense Retrieval Adversarial Attacks
Adversarial attacks against dense retrieval models used in RAG and search systems.
Dense Retrieval Attacks
Attacking dense retrieval systems by crafting adversarial passages that achieve high relevance scores for target queries while containing malicious content.
Embedding Backdoor Attacks
Inserting backdoors into embedding models that cause specific trigger inputs to produce predetermined embedding vectors for adversarial retrieval.
Embedding Drift Attacks
Causing gradual embedding drift in vector stores through repeated small manipulations.
Embedding Extraction Techniques
Methods for extracting embedding model weights and behavior through API access, including dimension reduction and reconstruction attacks.
Embedding Inversion Attacks (Embedding Vector Security)
Recovering original text from embedding vectors using inversion techniques.
Embedding Model Extraction
Extracting embedding model behavior through systematic API querying.
Embedding Poisoning Techniques
Techniques for poisoning embedding spaces to manipulate retrieval and similarity search.
Embedding Space Mapping Attacks
Using embedding space topology analysis to identify adversarial regions and craft inputs that produce targeted embedding representations.
Embedding Watermarking Attacks
Attacking and evading embedding watermarking schemes used for content tracking and intellectual property protection.
Hybrid Search Exploitation
Exploiting hybrid dense-sparse search systems through coordinated embedding manipulation.
Hybrid Search Manipulation
Attacking hybrid search systems that combine dense and sparse retrieval by exploiting score fusion and re-ranking vulnerabilities.
Multi-Vector Retrieval Attacks
Exploiting multi-vector retrieval models like ColBERT through token-level adversarial manipulation and late interaction exploitation.
Multimodal Embedding Attacks (Embedding Vector Security)
Attacking multimodal embedding spaces like CLIP for cross-modal manipulation.
RAG Retrieval Poisoning
Poisoning document collections to manipulate what gets retrieved by RAG systems, enabling indirect prompt injection at scale.
Reranker Adversarial Inputs
Crafting adversarial inputs that manipulate cross-encoder reranking models in retrieval pipelines.
Reranker Exploitation Techniques
Attacking cross-encoder rerankers used in multi-stage retrieval pipelines to promote adversarial documents past initial retrieval filtering.
Similarity Search Gaming
Techniques for crafting adversarial content that games similarity search to ensure attacker-controlled documents rank highest in retrieval results.
Similarity Search Manipulation
Manipulating similarity search results through adversarial embedding crafting.
Sparse Embedding Attacks
Exploiting sparse embedding methods (BM25, SPLADE) through keyword stuffing, term frequency manipulation, and index poisoning.
Sparse Embedding Manipulation
Manipulating sparse embeddings (BM25, SPLADE) for retrieval result poisoning.
Vector DB Access Control Bypass Techniques
Techniques for bypassing vector database access controls including namespace escaping, metadata injection, and query manipulation.
Vector Database Denial of Service
Denial of service attacks targeting vector databases through adversarial query patterns, index bloating, and resource exhaustion.
Vector Database Injection Attacks (Embedding Vector Security)
Comprehensive techniques for injecting adversarial vectors into vector databases to manipulate retrieval results and influence RAG system outputs.
Vector Database Injection Attacks (Embedding Vector Security Overview)
Injecting adversarial documents into vector databases to influence retrieval results.
Semantic Space Injection Research
Research into injections that operate in semantic embedding space rather than token space, exploiting learned representations directly.
Embedding Adversarial Perturbation
Craft adversarial inputs that produce target embeddings for retrieval manipulation.
Embedding Inversion Attack Implementation
Implement embedding inversion to recover original text from vector database embeddings.
Embedding Basics for Security
Understand text embeddings and their security relevance by generating, comparing, and manipulating embedding vectors.
Lab: Advanced Embedding Manipulation
Manipulate embedding vectors to achieve target similarity scores for RAG poisoning and retrieval manipulation.
Practical Embedding Manipulation
Manipulate text to achieve target embedding similarities for poisoning retrieval augmented generation systems.
Embedding Collision Attack Walkthrough
Craft documents that collide in embedding space with target queries to hijack RAG retrieval results.
Embedding Inversion Attack Walkthrough
Walkthrough of inverting text embeddings to recover original documents from vector databases.
Embedding Poisoning Detection System
Build a detection system for identifying poisoned documents in vector databases using statistical analysis.
RAG System Red Team Engagement
Complete walkthrough for testing RAG applications: document injection, cross-scope retrieval exploitation, embedding manipulation, data exfiltration through retrieval, and chunk boundary attacks.