3 papers
cs.LG2026
Models Know Their Shortcuts: Deployment-Time Shortcut Mitigation
Jiayi Li, Shijie Tang, Gün Kaynar +2
Pretrained text encoders are prone to shortcut learning, relying on token-label correlations that fail once the distribution shifts in deployment. Existing shortcut mitigation meth…
cs.LG2026
Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priors
Shiyi Du, Jiayuan Liu, Weihua Du +6
Automated agentic workflow design currently relies on per-task iterative search, which is computationally prohibitive and fails to reuse structural knowledge across tasks. We obser…
q-bio.GN2025
CodonMoE: DNA Language Models for mRNA Analyses
Shiyi Du, Litian Liang, Jiayi Li +1
Genomic language models (gLMs) face a fundamental efficiency challenge: either maintain separate specialized models for each biological modality (DNA and RNA) or develop large mult…