13 papers
When Does Permutation Instability Generalize? Independent-View Validation for Listwise LLM Reranking
Wenzhang Du
Listwise language-model rerankers often disagree across equivalent candidate permutations. Finite instability diagnostics are therefore used to motivate additional sampling, aggreg…
Which Reranking Conclusions Survive the Answer Interface? A Prospective Finite-Orbit Audit
Wenzhang Du
Rerankers are increasingly evaluated through downstream language-model answers. This raises a retrieval-measurement question: if only the reader's answer interface changes, should…
What Would Fix This RAG Failure? Auditing Counterfactual Response with Paired Evidence Interventions
Wenzhang Du
A failed retrieval-augmented generation (RAG) answer can be consistent with several unseen responses to evidence repair. We introduce Pair-ID, an offline audit that holds one query…
Optimistic Feasible Search for Closed-Loop Fair Threshold Decision-Making
Wenzhang Du
Closed-loop decision-making systems (e.g., lending, screening, or recidivism risk assessment) often operate under fairness and service constraints while inducing feedback effects:…
Contract-Driven QoE Auditing for Speech and Singing Services: From MOS Regression to Service Graphs
Wenzhang Du
Subjective mean opinion scores (MOS) remain the de-facto target for non-intrusive speech and singing quality assessment. However, MOS is a scalar that collapses heterogeneous user…
Contract-Governed Training for Earth Observation: Observed Service Agreement Graphs and Coverage-Accuracy Trade-offs
Wenzhang Du
Earth observation (EO) models are frequently trained under implicit sampling policies that optimize global accuracy but provide no explicit guarantees on who (which regions, classe…
Retrieval-Augmented Memory for Online Learning
Wenzhang Du
Retrieval-augmented models couple parametric predictors with non-parametric memories, but their use in streaming supervised learning with concept drift is not well understood. We s…
City-Conditioned Memory for Multi-City Traffic and Mobility Forecasting
Wenzhang Du
Deploying spatio-temporal forecasting models across many cities is difficult: traffic networks differ in size and topology, data availability can vary by orders of magnitude, and n…
Prediction-space knowledge markets for communication-efficient federated learning on multimedia tasks
Wenzhang Du
Federated learning (FL) enables collaborative training over distributed multimedia data but suffers acutely from statistical heterogeneity and communication constraints, especially…
When +1% Is Not Enough: A Paired Bootstrap Protocol for Evaluating Small Improvements
Wenzhang Du
Recent machine learning papers often report 1-2 percentage point improvements from a single run on a benchmark. These gains are highly sensitive to random seeds, data ordering, and…
Mitigating Catastrophic Forgetting in Streaming Generative and Predictive Learning via Stateful Replay
Wenzhang Du
Many deployed learning systems must update models on streaming data under memory constraints. The default strategy, sequential fine-tuning on each new phase, is architecture-agnost…
Step-E: A Differentiable Data Cleaning Framework for Robust Learning with Noisy Labels
Wenzhang Du
Training data collected in the wild often contain noisy labels and outliers that substantially degrade the performance and reliability of deep neural networks. While data cleaning…
CAO: Curvature-Adaptive Optimization via Periodic Low-Rank Hessian Sketching
Wenzhang Du
First-order optimizers are reliable but slow in sharp, anisotropic regions. We study a curvature-adaptive method that periodically sketches a low-rank Hessian subspace via Hessian-…