Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations
研究解决了跨语言模型的安全性漂移问题，通过线性代数方法精确刻画，并提出了一种校准暴露度量来区分不同类型的漂移。
0 citationsRead paper
Selected work
Representative Papers
Interpolation Is Not Invariance: Pair Count Is Not Coverage in Transformation Audits
本文解决了转换审计中配对计数过估覆盖范围的问题，通过区分四个关键量并提出精确的块-Woodbury留一轨道更新方法来提高审计准确性和部署可靠性。
0 citationsRead paper
AI Innovation and Firm Performance in the Medical Device Industry
研究通过结合FDA许可、专利等数据，探讨了AI创新对医疗器械公司绩效的影响，发现外部AI研究合作尤其是与产业和临床的合作能有效促进AI设备引入并提高劳动生产率。
0 citationsRead paper
Capital allocation on decentralized lending platforms
本文研究了去中心化借贷平台上贷方如何在多个市场中分配固定预算，通过三种利率模型得出了解决方案，并指出在折线利率模型下，贷方和借方的最佳策略存在不对称性。
0 citationsRead paper
Spectral Rank Certification for Foundation Model Adapters
This study addresses the discrepancy between nominal and effective ranks in foundation model adapters and the lack of finite-sample statistical inference. We propose an auditing framework based on joint spectral gap calibration and empirical null hypotheses. By integrating high-dimensional spectral analysis, Monte Carlo testing, and multiple correction, we establish precise chi-square divergence and Le Cam bounds to calibrate spectral evidence. Empirical evaluation across 26 adapters demonstrates that the calibrated effective rank is significantly lower than the nominal rank and distinct from energy retention rates. These findings reveal the intrinsic low-dimensional structure of adapters, providing a rigorous statistical certification and novel evaluation paradigm for parameter-efficient fine-tuning.
0 citationsRead paper
Recent publications
Latest Papers
Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations
研究解决了跨语言模型的安全性漂移问题，通过线性代数方法精确刻画，并提出了一种校准暴露度量来区分不同类型的漂移。
0 citationsRead paper
Interpolation Is Not Invariance: Pair Count Is Not Coverage in Transformation Audits
本文解决了转换审计中配对计数过估覆盖范围的问题，通过区分四个关键量并提出精确的块-Woodbury留一轨道更新方法来提高审计准确性和部署可靠性。
0 citationsRead paper
AI Innovation and Firm Performance in the Medical Device Industry
研究通过结合FDA许可、专利等数据，探讨了AI创新对医疗器械公司绩效的影响，发现外部AI研究合作尤其是与产业和临床的合作能有效促进AI设备引入并提高劳动生产率。
0 citationsRead paper
Capital allocation on decentralized lending platforms
本文研究了去中心化借贷平台上贷方如何在多个市场中分配固定预算，通过三种利率模型得出了解决方案，并指出在折线利率模型下，贷方和借方的最佳策略存在不对称性。
0 citationsRead paper
Spectral Rank Certification for Foundation Model Adapters
This study addresses the discrepancy between nominal and effective ranks in foundation model adapters and the lack of finite-sample statistical inference. We propose an auditing framework based on joint spectral gap calibration and empirical null hypotheses. By integrating high-dimensional spectral analysis, Monte Carlo testing, and multiple correction, we establish precise chi-square divergence and Le Cam bounds to calibrate spectral evidence. Empirical evaluation across 26 adapters demonstrates that the calibrated effective rank is significantly lower than the nominal rank and distinct from energy retention rates. These findings reveal the intrinsic low-dimensional structure of adapters, providing a rigorous statistical certification and novel evaluation paradigm for parameter-efficient fine-tuning.
0 citationsRead paper