PulseAugur
EN
LIVE 13:57:26

New benchmarks tackle privacy risks in large language models

Researchers have developed new methods to evaluate membership inference attacks (MIAs) against large language models (LLMs), particularly focusing on audio and text modalities. The first study introduces a systematic evaluation for Large Audio-Language Models (LALMs) using "Multi-modal Blind Baselines" to control for distribution shifts, revealing that memorization is cross-modal and linked to speaker vocal identity. The second paper, CheckMIABench, proposes a framework for principled MIA evaluation on LLMs by leveraging intermediate training checkpoints and public data, demonstrating its application on Pythia and OLMo models and releasing a modular library for further research. AI

IMPACT These new evaluation frameworks and findings are crucial for developing more private LLMs and establishing robust auditing standards.

RANK_REASON The cluster contains two academic papers published on arXiv detailing new research methodologies for evaluating privacy risks in language models.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.AI TIER_1 English(EN) · Jia-Kai Dong, Yu-Xiang Lin, Hung-Yi Lee ·

    Membership Inference Attacks against Large Audio Language Models

    arXiv:2603.28378v2 Announce Type: replace-cross Abstract: We present the first systematic Membership Inference Attack (MIA) evaluation of LALMs. Using Multi-modal Blind Baselines based on textual, spectral and prosodic features, we demonstrate that common audio datasets exhibit n…

  2. arXiv cs.LG TIER_1 English(EN) · Jeffrey G. Wang, Jason Wang, Marvin Li, Seth Neel ·

    CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models

    arXiv:2606.17464v1 Announce Type: new Abstract: Membership inference attacks (MIAs) are a canonical way to assess a machine learning model's privacy properties. Although several attempts have been made to evaluate MIAs on language models, the extant literature has suffered numero…