ArXiv TLDR

Yong Li

6 papers ยท Latest:

Computer Vision

LoViF 2026 The First Challenge on Holistic Quality Assessment for 4D World Model (PhyScore)

The LoViF 2026 PhyScore challenge evaluates 4D world models on holistic quality, including physical realism and temporal consistency.

2605.05187
Computer Vision

A Benchmark for Interactive World Models with a Unified Action Generation Framework

iWorld-Bench is a new benchmark with a large dataset and unified framework for evaluating interactive world models' physical interaction abilities.

2605.03941
Human-Computer Interaction

AgentEconomist: An End-to-end Agentic System Translating Economic Intuitions into Executable Computational Experiments

AgentEconomist is an agentic system that translates economic intuitions into executable computational experiments, improving research verification.

2604.27725
Information Retrieval

Explaining the "Why": A Unified Framework for the Additive Attribution of Changes in Arbitrary Measures

This paper introduces a unified, game theory-based framework to explain why aggregated measures change, offering general and rigorous attribution.

2604.26266
Computer Vision

Beyond Text Prompts: Precise Concept Erasure through Text-Image Collaboration

TICoE is a text-image collaborative framework that precisely erases undesirable concepts from generative models while preserving content fidelity.

2604.15829
Information Retrieval

Enhancing Local Life Service Recommendation with Agentic Reasoning in Large Language Model

This paper proposes an LLM-based framework for local life service recommendation that unifies living need prediction and service recommendation.

2604.14051

๐Ÿ“ฌ Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week โ€” summarized, scored, and delivered to your inbox every Monday.