Explore primary-source AI security research, evaluated models, attack techniques, and defensive evidence.
985 research findings · 1119 evaluated models
Matches every word across titles, descriptions, sources, affected systems, and models.
Deleting a memory record can leave its information in an agent's summaries, pending plans and KV cache. The paper evaluates revocation across this derived execution state.
Document-image PII detection can miss identifiers despite improving average localization scores. LeakageBench evaluates 500 pages with 11,954 annotations.
Context assembly can promote repository, tool or skill content into higher-priority instructions or persistent state. The paper studies 12 pinned agent-harness versions.
A compromised serving framework can violate user-data isolation through shared GPU state. GIFT evaluates per-user information-flow enforcement in vLLM and DistServe.
API gateways can pool distinct customers into one upstream cache identity. The authors observe cross-customer cache reuse in five deliberately pooled gateway configurations through two provider interfaces.
SRE-Bench evaluates whether cybersecurity agents can recover the behavior of realistic binary-only software without relying on source-code memorization. The authors construct 19 private programs, 44 anti-analysis primitives, 262 binary instances, and 1,572 deterministic grading tasks covering security-relevant reverse-engineering scenarios.
The authors report historical isolation failures in client-held encrypted reasoning blocks across compatible provider API contexts.
Self-evolving agent-skill systems may promote recurring, attacker-controlled execution records into persistent reusable instructions. The paper evaluates whether poisoned but task-plausible trajectories survive aggregation and appear in generated skill artifacts, using inert canary behaviors and two structurally different skill-evolution pipelines.
SkillSentry evaluates third-party agent skills by constructing source-grounded decoy environments and comparing matched executions with and without the tested skill. The method requires completed, observable, skill-attributed side effects rather than treating suspicious text, ordinary privileged operations, or unexecuted paths as proven malicious behavior.
Third-party fine-tuning adapters may contain backdoors. Z-PEFT screens adapter weights using spectral features, evaluated on PADBench's 13,300 adapters.
Entries summarize publicly available primary-source security research. Model names reflect only systems explicitly evaluated by the cited paper, and measurements are research-reported unless independent verification is stated.
Discovery and publication are separate: new papers enter a candidate queue, then require human review and explicit approval before publication. Updates follow that review process, not a guaranteed daily publishing schedule. “Last analyzed” is the newest analysis date among published entries; it does not mean that all recent research has been covered.