Tag: AI model robustness
Browse our exclusive articles!
Robust Reasoning Benchmark for LLMs: Key Insights
Explore the Robust Reasoning Benchmark evaluating LLMs' resilience to perturbations and uncover critical insights on improving AI reasoning accuracy.
Boost Sparse Autoencoder Robustness with Masked Regularization
Enhance sparse autoencoder reliability using masked regularization to reduce feature absorption and improve out-of-distribution performance.
Stable Predictors from Weak Supervision under Shift
Explore learning stable predictors from weak supervision amid distribution shifts and supervision drift for robust machine learning models.
Zero-Shot Quantization: Boost Model Robustness Easily
Discover zero-shot quantization using weight-space arithmetic to enhance model robustness without extra training or data, improving low-bit deployments.
Training-Free Visual Concept Unlearning in VLMs: Benchmarking
Explore VLM-UnBench, the first benchmark for training-free visual concept unlearning in VLMs, revealing key challenges and insights for ethical AI.