Lira
PulseAugur coverage of Lira — every cluster mentioning Lira across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Rebuff 0.1.1 guard module bypassed by simple phrase modification
An audit of the Rebuff 0.1.1 guard module's default heuristic layer revealed a 0.75 threshold that can be bypassed with simple modifications. When tested with its own documented example, the module flagged the phrase wi…
-
New method estimates AI model privacy risks without reference models
Researchers have developed a new method to estimate the vulnerability of AI models to membership inference attacks (MIAs) without needing to train separate reference models. This approach analyzes the loss distributions…
-
New framework unifies privacy attacks, introduces Bayesian approach
Researchers have unified three leading membership inference attacks (MIAs) – LiRA, RMIA, and BASE – under a single exponential-family log-likelihood ratio framework. This unification reveals a hierarchy of model complex…
-
New research audits fairness-privacy trade-offs in ML subpopulations
A new study published on arXiv investigates the complex interplay between fairness-enhancing algorithms and privacy leakage in machine learning models. Researchers adapted the Likelihood Ratio Attack (LiRA) to audit pri…
-
Alibaba unveils Qwen-Robot embodied AI model series
Alibaba has launched the Qwen-Robot series, its first comprehensive embodied AI model family. This series includes three distinct models: Qwen-RobotManip for VLA (Vision-Language-Action) operations, Qwen-RobotNav for VL…
-
EEG Foundation Models Leak Data Despite Standard Audits
Researchers have developed a new auditing framework for EEG foundation models that goes beyond single-endpoint evaluations. This framework jointly audits multiple endpoints, revealing that models cleared by individual t…
-
New method traces AI model training data via semantic correlations
Researchers have developed a new method called idSCD to identify specific datasets used in training AI models. This technique analyzes the semantic correlation structure learned by a model, looking for incidental regula…