Announcement_19

Four papers accepted at ICML 2026! “Discretized Density-Guided Source-Free Domain Adaptation for Regression” (Spotlight, acceptance rate ≈ 2.2%), “FUSE: Full-spectrum Unlearnable Examples via Spectral Equalization”, “Baguan-TS: A Sequence-Native In-Context Learning Model for Time Series Forecasting with Covariates”, and “Attention with Routed-Memory for Learnable Sparse Control”. Congratulations to all collaborators!