不用教它撒谎，LLM也会“心口不一”：上海AI Lab揭露高风险下的模型欺骗行为
"数据污染" 相关文章
用「进化+压力测试」自动生成的竞赛级编程题，各家大模型谁更hold住？
Evaluating Latent Knowledge of Public Tabular Datasets in Large Language Models
Evaluating Latent Knowledge of Public Tabular Datasets in Large Language Models
德克萨斯A&M大学：LLM阅读社交媒体低质内容致认知能力衰退
Deep Associations, High Creativity: A Simple yet Effective Metric for Evaluating Large Language Models
A small amount of bad data can ‘poison’ even the largest AI models, researchers warn
Detecting Data Contamination from Reinforcement Learning Post-training for Large Language Models
Detecting Data Contamination from Reinforcement Learning Post-training for Large Language Models
RADAR: Mechanistic Pathways for Detecting Data Contamination in LLM Evaluation
RADAR: Mechanistic Pathways for Detecting Data Contamination in LLM Evaluation
RADAR: Mechanistic Pathways for Detecting Data Contamination in LLM Evaluation
Detecting Distillation Data from Reasoning Models
Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluation
A Practical Guide to Maintaining Machine Learning in Production
How Google’s $60M Reddit Deal Undermines the Future of Knowledge
How Generative Models Are Ruining Themselves
How Generative Models Are Ruining Themselves
Both Text and Images Leaked! A Systematic Analysis of Data Contamination in Multimodal LLM
VisMoDAl: Visual Analytics for Evaluating and Improving Corruption Robustness of Vision-Language Models
