Newest papers

CALLIOPE: Private Insights into AI Use, Without Embeddings

Edith Cohen, Vadym Doroshenko, Badih Ghazi, Charlie Harrison, Pritish Kamath, Alexander Knop, Ravi Kumar, Ethan Leeman · 2026 · arXiv

DOI: 10.48550/arXiv.2610.28797 · open-access PDF

Portuguese

Introduzimos o CALLIOPE, um framework top-down para a descoberta de hierarquias de tópicos em grandes corporações de dados não estruturados (por exemplo, conversas LLM), ao mesmo tempo em que apoiamos a privacidade diferencial (DP). Ao contrário das abordagens tradicionais e até recentes aumentadas pelo LLM que dependem, em última análise, de inserções de vetores agrupadas, o nosso framework aproveita as capacidades do LLM diretamente para todo o pipeline. Os experimentos demonstram que esta abordagem oferece uma melhor utilidade em relação aos métodos existentes (baseados em inserções) na configuração DP, especialmente no regime de alta privacidade.

Machine translation (Meta M2M100); the English original follows.

Abstract

We introduce CALLIOPE, a top-down framework for discovering topic hierarchies in large corpora of unstructured data (e.g., LLM conversations), while supporting differential privacy (DP). Unlike traditional and even recent LLM-augmented approaches that ultimately rely on clustering vector embeddings, our framework leverages LLM capabilities directly for the entire pipeline. Experiments demonstrate that this approach provides better utility over existing (embedding-based) methods in the DP setting, especially in the high privacy regime. We evaluate the quality of CALLIOPE outputs using a variety

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28797, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange