Quantifying perturbation impacts for large language models
원문 초록
We consider the problem of quantifying how an input perturbation impacts the outputs of large language models (LLMs), a fundamental task for model reliability and post-hoc interpretability. A key obstacle in this domain is disentangling the meaningful changes in model responses from the intrinsic stochasticity of LLM outputs. To overcome this, we introduce Distribution-Based Perturbation Analysis (DBPA), a framework that reformulates LLM perturbation analysis as a frequentist hypothesis testing problem. DBPA constructs empirical null and alternative output distributions within a low-dimensional semantic similarity space via Monte Carlo sampling. Comparisons of Monte Carlo estimates in the reduced dimensionality space enables tractable frequentist inference without relying on restrictive distributional assumptions. The framework is model-agnostic, supports the evaluation of arbitrary input perturbations on any black-box LLM, yields interpretable p-values, supports multiple perturbation testing via controlled error rates, and provides scalar effect sizes for any chosen similarity or distance metric. We demonstrate the effectiveness of DBPA in evaluating perturbation impacts, showing its versatility for perturbation analysis.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Temporally-Grounded Language Generation: Towards Real-Time Vision-Language Models
FA: Forced Prompt Learning of Vision-Language Models for Out-of-Distribution Detection
ImpliRet: Benchmarking the Implicit Fact Retrieval Challenge
Inv-Entropy: A Fully Probabilistic Framework for Uncertainty Quantification in Language Models
Trend-Aware Fashion Recommendation with Visual Segmentation and Semantic Similarity
KNN-Defense: Defense against 3D Adversarial Point Clouds using Nearest-Neighbor Search
토론
이 논문과 연결된 포럼 글이 아직 없습니다.