Yu Cheng
6 papers · Latest:
TabEmbed: Benchmarking and Learning Generalist Embeddings for Tabular Understanding
TabEmbed introduces a generalist embedding model for tabular data, unifying classification and retrieval, alongside TabBench for evaluation.
The Keplerian disk, envelope, and streamers surrounding an early O-type protostar in the Sagittarius C cloud of the Central Molecular Zone
ALMA observations reveal a Keplerian disk, envelope, and streamers around a 40 M☉ protostar in Sgr C, detailing massive star formation in the CMZ.
Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs
PVM introduces a lightweight module for LVLMs to counteract visual signal dilution, ensuring sustained perception during deep generation.
HOTDISK. Finding Massive Protostellar Disks with Water and Refractory Molecular Species
The HOTDISK project uses ALMA to identify massive protostellar disks via vibrationally excited water, NaCl, and SiS, revealing their common presence.
ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety
ProjLens demystifies MLLM backdoors, revealing how projector fine-tuning introduces vulnerabilities and detailing their low-rank structure and activation mechanism.
RIRF: Reasoning Image Restoration Framework
A novel framework, R&R, integrates Chain-of-Thought reasoning to diagnose image degradations and guide restoration, achieving state-of-the-art performance.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.