6 papers · 1 filter
Margin-Drop Coordinates for Cross-Budget Robustness Evaluation
Yanliang Huang, Zhen Zhang, Peng Xie +4
Fixed-budget robustness evaluation can select the wrong frozen vision encoder. An encoder that survives a shallow attack may lose most of that robustness when the same evaluation i…
Cross-View Correspondence Is a Measurement Intervention: Two-Sided Validation for Agent Evaluation and Credit Assignment
Zhen Zhang, Ahmad Hafez, Amr Alanwar
Agent evaluations and trace-based learning often compare outputs across transformed views through a post-response correspondence treated as neutral preprocessing. We show that this…
Certified Mechanistic Interpretability: Lifting Single-Input Findings to Bounded Neighbourhoods
Zhen Zhang, Yanliang Huang, Peng Xie +2
Mechanistic interpretability reverse-engineers transformer circuits one input at a time, leaving observed mechanisms without guarantees over bounded input neighbourhoods. We addres…
Matrix Zonotopic Attention: A Context-Adaptive Value Projection for Set Transformers
Zhen Zhang, Amr Alanwar
Multi-head attention combines an input-dependent softmax routing with an input-independent linear value projection, so the per-sample operator mapping aggregated values to outputs…
Hybrid Probabilistic Zonotopes for Identifiable and Refinable Predictive Uncertainty
Zhen Zhang, Amr Alanwar
Probabilistic prediction heads in neural networks typically output either a Gaussian mixture or a single conformal region. Neither separates the distinct sources of uncertainty oft…
DiffSlack: Learning under Nonlinear Inequality Constraints via Learnable Slack Variables
Ziqian Wang, Chenxi Fang, Zhen Zhang
Enforcing nonlinear inequality constraints in neural networks remains challenging, especially when the output is subject to many coupled constraints. Existing hard constraint metho…