Chang Liu
6 papers · Latest:
Self-organized MT Direction Maps Emerge from Spatiotemporal Contrastive Optimization
A spatiotemporal TDANN model, trained with self-supervised learning, spontaneously generates brain-like direction maps in the visual cortex.
Count Anything at Any Granularity
This paper introduces multi-grained open-world object counting, a new dataset (KubriCount), and a model (HieraCount) to improve counting accuracy.
DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation
DCGL uses dual-channel graph learning with LLMs for knowledge-aware recommendation, improving performance by decoupling semantics and behavior.
The Eye of Sauron in SN 2025ngs: a Short-plateau Cousin of SN 1998S with Evidence for a Ring-like Circumstellar Medium
SN 2025ngs is a peculiar short-plateau supernova with evidence for a ring-like circumstellar medium, spectroscopically resembling SN 1998S.
Quotient-Space Diffusion Models
Quotient-Space Diffusion Models simplify learning and improve generative AI by formally incorporating symmetries, outperforming prior methods for molecular structure generation.
Into the Gray Zone: Domain Contexts Can Blur LLM Safety Boundaries
Jargon exploits LLM safety boundaries by leveraging domain contexts, achieving >93% attack success on frontier models, and proposes a policy-guided safeguard.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.