Robustness
The ability of an AI system to maintain stable outputs when confronted with input perturbations, intermediate-step errors, or environmental changes, including the capacity to detect contradictions and perform self-correction. Robustness serves as an important metric for evaluating the reliability of large language models.
Timeline (last 90 days)
Multi-value alignment objectives need to assess the similarity and conflict between values as a prior indicator of robustness.
An engineeringly reliable embedding model should exhibit minimal vector change under character-level and lexical-level perturbations, highly similar vectors under semantic-level paraphrasing, and a perceptible distance change when there is a substantive semantic shift (such as negation).
The error recovery capability corresponds to the concepts of Robustness and Self-correction in AI system evaluation.
Robustness does not require a function to be completely consistent every time, but does require it to be sufficiently intelligent every time (a lower bound on intelligent decision quality).
All Facts (5)
鲁棒性（Robustness）概念源自控制理论，指系统在面对输入扰动或环境变化时维持正常功能的能力
90%UnverifiedMulti-value alignment objectives need to assess the similarity and conflict between values as a prior indicator of robustness.
50%UnverifiedAn engineeringly reliable embedding model should exhibit minimal vector change under character-level and lexical-level perturbations, highly similar vectors under semantic-level paraphrasing, and a perceptible distance change when there is a substantive semantic shift (such as negation).
50%UnverifiedThe error recovery capability corresponds to the concepts of Robustness and Self-correction in AI system evaluation.
50%UnverifiedRobustness does not require a function to be completely consistent every time, but does require it to be sufficiently intelligent every time (a lower bound on intelligent decision quality).
50%