AI ethics evidence index Search
Source-linked daily record
Evidence record 16547 · automatically gathered

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents

Memory-augmented LLM agents rely on rich context for long-horizon reasoning and acting, yet their memory modules expose a persistent attack surface for malicious records, making the study of memory poisoning threats imperative. However, existing query-only attacks often fail to remain effective in two realistic and prevalent settings: large-scale benign memory pools and active input auditing. Consequently, current approaches fall short when facing the dual challenges of high retrieval competitiv

Record details

Published: 4 August 2026
Source: arXiv cs.AI
Category: Research
Topics: Agents & autonomy · Transparency
Retrieved: 5 August 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Shared topic: Agents & autonomy · Shared topic: Transparency

CARE-Bench: Benchmarking Patient-Facing LLM Triage

arXiv cs.AI · 4 August 2026

Shared topic: Agents & autonomy · Shared topic: Transparency

Accountability Asymmetry and Structural Trust in Autonomous AI Systems

arXiv · 4 August 2026

Shared topic: Agents & autonomy · Shared topic: Transparency

AgentForge: An Immersive Role-Playing Platform for Learning Agentic Software Engineering

arXiv cs.HC · 4 August 2026

Shared topic: Agents & autonomy · Shared topic: Transparency

WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks

arXiv · 4 August 2026

Shared topic: Agents & autonomy · Shared topic: Transparency

Humans Are More Diverse: Frontier LLMs Show Extreme Policies in Idealised AI Development Races

arXiv cs.CY · 4 August 2026

Shared topic: Agents & autonomy · Shared topic: Transparency

Hybrid AI for Explainable and Accurate Conversational Agents in eGovernment

arXiv cs.CY · 4 August 2026

How to cite this record

ethics.ai (4 August 2026), “MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents,” evidence record 16547, https://ethics.ai/record/16547 (originally published by arXiv cs.AI).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.