SPIA benchmark reveals that subject-level inference protection falls to as low as 33% even after masking over 90% of PII spans, with non-target subjects remaining highly exposed under target-focused anonymization.
Pii-bench: Evaluating query-aware privacy protection systems
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
LLMs support taxonomy-agnostic detection and value extraction of PII in HTTP traffic via a deterministic pre-processing plus classification pipeline, plus an LLM generator for synthetic labeled traffic.
FinSec is a multi-stage detection system for financial LLM dialogues that reaches 90.13% F1 score, cuts attack success rate to 9.09%, and raises AUPRC to 0.9189.
citing papers explorer
-
Subject-level Inference for Realistic Text Anonymization Evaluation
SPIA benchmark reveals that subject-level inference protection falls to as low as 33% even after masking over 90% of PII spans, with non-target subjects remaining highly exposed under target-focused anonymization.
-
Addressing Labelled Data Scarcity: Taxonomy-Agnostic Annotation of PII Values in HTTP Traffic using LLMs
LLMs support taxonomy-agnostic detection and value extraction of PII in HTTP traffic via a deterministic pre-processing plus classification pipeline, plus an LLM generator for synthetic labeled traffic.
-
Conversations Risk Detection LLMs in Financial Agents via Multi-Stage Generative Rollout
FinSec is a multi-stage detection system for financial LLM dialogues that reaches 90.13% F1 score, cuts attack success rate to 9.09%, and raises AUPRC to 0.9189.