13 papers
TWIN-MoE: Two-stage Intra-Expert Imputation Network for Incomplete Multimodal Learning
Liang Nathan Zheng, Jiyuan Liu, Wenhao Liang +3
Recent Mixture-of-Experts (MoE) approaches address missing modalities either by routing each missingness pattern to a dedicated expert, which scales poorly as the number of modalit…
Spatial Supervision Without Attribution Optimization: Improving Post-Hoc Class Activation Maps via Box-Guided Evidence Routing
Wenhao Liang, Liangwei Nathan Zheng, Lin Yue +3
Post-hoc class activation maps (CAMs) are a standard tool for inspecting the evidence behind an image classifier's predictions, yet nothing in ordinary training encourages these ma…
Routing Probes Can Improve Without New Information: An Exact-Null Audit of Uncertainty Beyond Model Outputs
Wenhao Liang, Lin Yue, Wei Emma Zhang +3
Routing signals of modern vision transformers -- expert gates, attention-residual weights and halting scores -- often improve probes that predict whether the model is correct, and…
Signal-Routed Temperature Scaling: Low-Capacity Risk-Conditioned Calibration for Small Validation Budgets
Wenhao Liang, Liangwei Nathan Zheng, Lin Yue +4
When a classifier is recalibrated from only a few thousand held-out examples, the capacity of the calibration map becomes a statistical design choice rather than a purely architect…
How Much Can Reliability Drift Under a Fixed Confidence Distribution?
Wenhao Liang, Lin Yue, Wei Emma Zhang +3
A classifier's conditional accuracy can change while its confidence distribution stays exactly the same. We study the worst-case movement of the reliability relation under covariat…
Probing Routing-Conditional Calibration in Attention-Residual Transformers
Wenhao Liang, Lin Yue, Wei Emma Zhang +4
Post-hoc calibration is usually evaluated as a function of logits or softmax confidence alone, even as routing-augmented architectures increasingly accompany predictions with sampl…
Lifting Manifolds to Mitigate Pseudo-Alignment in LLM4TS
Liangwei Nathan Zheng, Wenhao Liang, Wei Emma Zhang +3
Pseudo-Alignment is a pervasive challenge in many large language models for time series (LLM4TS) models, often causing them to underperform compared to linear models or randomly in…
Calibration Attention: Learning Reliability-Aware Representations for Vision Transformers
Wenhao Liang, Wei Emma Zhang, Lin Yue +4
Most calibration methods operate at the logit level, implicitly assuming that miscalibration can be corrected without changing the underlying representation. We challenge this assu…
PostHoc FREE Calibrating on Kolmogorov Arnold Networks
Wenhao Liang, Wei Emma Zhang, Lin Yue +3
Kolmogorov Arnold Networks (KANs) are neural architectures inspired by the Kolmogorov Arnold representation theorem that leverage B Spline parameterizations for flexible, locally a…
We Care Each Pixel: Calibrating on Medical Segmentation Model
Wenhao Liang, Wei Zhang, Lin Yue +3
Medical image segmentation is fundamental for computer-aided diagnostics, providing accurate delineation of anatomical structures and pathological regions. While common metrics suc…
Focal Calibration Loss: Controlling Posterior Distortion in Deep Neural Classifiers
Wenhao Liang, Liangwei Zheng, Wei Zhang +1
Confidence calibration matters wherever a classifier's probabilities, not just its labels, are consumed downstream. We study Focal Calibration Loss (FCL), which adds a squared prob…
Correlation Analysis of Adversarial Attack in Time Series Classification
Zhengyang Li, Wenhao Liang, Chang Dong +2
This study investigates the vulnerability of time series classification models to adversarial attacks, with a focus on how these models process local versus global information unde…
Enhancing Financial Market Predictions: Causality-Driven Feature Selection
Wenhao Liang, Zhengyang Li, Weitong Chen
This paper introduces the FinSen dataset that revolutionizes financial market analysis by integrating economic and financial news articles from 197 countries with stock market data…