arXiv:2608. 02616v1 Announce Type: cross Abstract: We present the first independent, systematic evaluation of OpenAI's Privacy Filter (OPF), a 1.
arXiv:2608. 02616v2 Announce Type: replace-cross Abstract: We present what is, to our knowledge, the first systematic evaluation of OpenAI's Privacy Filter (OPF), a 1.
arXiv:2608. 05163v1 Announce Type: cross Abstract: A common assumption holds that switching to a non-English language makes a multilingual RAG system easier to attack for personal information.
arXiv:2609.38630v1 Announce Type: new Abstract: Privacy redaction must remove personal information while preserving relationships expressed in text. We develop a multilingual named-entity tagger with...
arXiv:2601. 10161v3 Announce Type: replace-cross Abstract: Named Entity Recognition (NER) and Personally Identifiable Information (PII) anonymization are critical tasks in Natural Language Processing (NLP) for information extraction and privacy preservation.
The OpenAI Privacy Filter is a 1.5‑billion‑parameter, bidirectional token‑classification model that detects and redacts personally identifiable information and secrets in unstructured text. It is built from an autoregressive checkpoint, converted into a banded‑attention classifier, and uses a constrained Viterbi decoder to produce coherent spans across eight privacy categories in a single forward pass. The model supports configurable precision‑recall tradeoffs, a 128,000‑token context window, and is designed for efficient local deployment and domain‑specific fine‑tuning as a data‑minimization component within layered privacy workflows.
arXiv:2606. 18782v1 Announce Type: cross Abstract: Large Language Models are increasingly applied to sensitive domains that require redaction of personally identifiable information (PII).
arXiv:2607. 02079v1 Announce Type: cross Abstract: We present HaloGuard 1.
The paper presents PersianAnonymizer, a method for anonymizing Persian customer chats by training a compact NER model using supervision from large language models (LLMs). Three instruction‑tuned LLMs—DeepSeek‑V3‑0324, GPT‑OSS‑120B, and Qwen3‑235B‑A22B‑Instruct‑2507—were used to generate span annotations, producing four corpora. A MatinaRoberta token‑classifier trained on each corpus achieved high macro‑F1 and Label Coverage Recall, with the OSS_ZeroShot‑derived NER labeling a 40K‑message test set in about two minutes on a single RTX 3090, demonstrating a practical, low‑cost approach to Persian data anonymization.
OpenSanctions Pairs is the first large‑scale public benchmark for entity matching on sanctions and OSINT data, comprising 755,540 expert‑labeled pairs drawn from over 1 million entities across 293 source datasets and 45 jurisdictions. The dataset spans multiple languages and writing systems, inconsistent structures, and time‑varying provenance, making it far more heterogeneous than prior benchmarks. Baseline experiments show a rule‑based matcher achieving 91.3 % F1, GPT‑4o reaching 99.0 % F1, and a locally deployable open‑source model scoring 98.2 % F1, with complementary failure modes that highlight the need to focus on downstream pipeline components.
arXiv:2608. 19957v1 Announce Type: new Abstract: Natural language code retrieval is a rapidly evolving task in computer science.
arXiv:2609.01046v1 Announce Type: cross Abstract: Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigat...
arXiv:2607. 22695v1 Announce Type: new Abstract: Large Language Models (LLMs) are capable of generalizing human language for the completion of never-before-seen tasks, leading to widespread deployment.