Zi-Han Wang

RAPID is proposed, a self-referenced latent maximization framework that removes external dependencies and encourages the same model update to induce consistently disruptive effects across training samples, thereby improving reachability and establishing a paradigm for real-time protection against unauthorized distillation in deployed T2I systems.

This work establishes a skill anti-distillation paradigm for protecting personal privacy and labor autonomy against unauthorized skill distillation, and introduces CounterPersona, which constructs targeted counter-persona evidence, packs compatible behavioral states into compact realization units, and strengthens them through rationale-guided consistency rewriting.

This work introduces MindTopo, a benchmark of topological intuition across five properties grounded in cognitive science and formal topology: continuity, separation, order, enclosure, and knots, which performs better on reasoning than on planning, and the best-performing model remains far below observed human performance.

LLMs falter in maintaining coherent multi-turn interactions, with only 37.3% successfully completing five-turn episodes, revealing a major gap in generative UI capabilities.

NeuralParker achieves superior parking performance by retaining essential route context in complex environments, outperforming traditional planners.

Morphology-derived tumor states reveal critical progression information that traditional diagnostic labels overlook, linking spatial histopathology to multi-omics insights.

Achieving state-of-the-art pose-free novel view synthesis by reusing a single 3D scene across multiple complementary roles transforms how we approach 3D reconstruction and rendering.

HarmTrace reveals that fine-grained target identification can drastically improve harmful meme detection accuracy, closing a critical gap in multimodal AI understanding.

Malicious modifications to router weights can turn Mixture-of-Experts models into trigger-controlled bottlenecks, revealing a critical vulnerability in AI serving architectures.

ATLAS outperforms existing systems in medication safety for older adults, achieving a 53.73-point lead in decision-making accuracy over top proprietary models.

LLM agents can have their proprietary skills stolen with just 3 interactions, exposing a major copyright vulnerability in the burgeoning skill marketplace.