Hui Xiong
6 papers · Latest:
SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing
SLIM enhances LLM molecular editing by using sparse latent steering to precisely control properties and improve success rates.
SCPRM: A Schema-aware Cumulative Process Reward Model for Knowledge Graph Question Answering
SCPRM is a new reward model for Knowledge Graph Question Answering that uses schema-aware cumulative rewards to improve multi-hop reasoning accuracy.
LLM-Oriented Information Retrieval: A Denoising-First Perspective
This paper argues that denoising is the primary bottleneck in LLM-oriented information retrieval, proposing a framework and techniques.
Geometry-Calibrated Conformal Abstention for Language Models
A post-hoc framework, Geometry-Calibrated Conformal Abstention, enables LMs to selectively abstain from answering when uncertain, boosting correctness.
Discrete Preference Learning for Personalized Multimodal Generation
DPPMG learns discrete modal-specific preferences to generate personalized and consistent multimodal content from user interactions.
Scaling Human-AI Coding Collaboration Requires a Governable Consensus Layer
This paper introduces Agentic Consensus, a new paradigm where a governable, graph-based world model replaces code as the primary artifact in human-AI coding.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.