Hao Zhou

The paper proposes PAC-DP, a personalized adaptive clipping framework that dynamically adjusts gradient clipping thresholds based on the desired privacy budget, significantly improving the privacy-utility trade-off in federated learning.

The paper introduces Safety Bottleneck Regularization (SBR), a novel defense mechanism that anchors LLM safety by constraining the unembedding layer, effectively preventing harmful fine-tuning (HFT) even when other defenses fail.

The paper introduces MTAVG-Bench 2.0, a new benchmark designed to diagnose high-level failure modes of cinematic expressiveness in multi-talker audio-video generation, showing that even advanced models struggle with complex scene-level failures.

MindZero introduces a self-supervised reinforcement learning framework that trains multimodal large language models (MLLMs) for efficient and robust online mental reasoning without requiring explicit mental state annotations.

The paper introduces AMix-2, a novel protein-text foundation model that unifies protein understanding and sequence design by embedding both modalities in a shared token space, achieving state-of-the-art performance on comprehensive benchmarks.

UniD$^3$ is a novel Knowledge Graph-enhanced RAG framework that processes vast biomedical literature to systematically extract, organize, and validate comprehensive drug-disease knowledge, achieving high accuracy in structured data generation.

The paper introduces GeoCoupling, a framework that systematically optimizes the temporal coupling between heterogeneous modalities to improve the co-design of biomolecules, outperforming fixed synchronous coupling methods.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.