13 papers
OmniPhys: Knowledge-Graph-Driven Benchmarking and Collective Optimization for Physical Commonsense in Text-to-Image Generation
Yajing Xu, Yarong Lan, Jiaoyan Chen +6
The paper presents OmniPhys, a knowledge-graph-based benchmark for evaluating physical commonsense in text-to-image models, and OmniPrompt, an iterative optimization framework that…
Scaling LLM Knowledge Boundaries via Distribution-Optimized Synthesis
Songze Li, Yarong Lan, Zhongpu Bo +16
Knowledge injection via synthetic data is crucial for enhancing Large Language Models (LLMs). However, current synthesis methods simply stop at preset token counts or fixed data ra…
Time-Unconditional Generative Speech Enhancement via Autonomous Rectified Flow
Wen Zhang, Wenbin Jiang, Yang Zhang +1
Most generative speech enhancement methods rely on explicit time-step embeddings for temporal conditioning. In this paper, we propose the Autonomous Rectified Flow framework, which…
Structured and Abstractive Reasoning on Multi-modal Relational Knowledge Images
Yichi Zhang, Zhuo Chen, Lingbing Guo +2
Understanding and reasoning with abstractive information from the visual modality presents significant challenges for current multi-modal large language models (MLLMs). Among the v…
Collaboration of Fusion and Independence: Hypercomplex-driven Robust Multi-Modal Knowledge Graph Completion
Zhiqiang Liu, Yichi Zhang, Mengshu Sun +2
Multi-modal knowledge graph completion (MMKGC) aims to discover missing facts in multi-modal knowledge graphs (MMKGs) by leveraging both structural relationships and diverse modali…
UniHR: Hierarchical Representation Learning for Unified Knowledge Graph Link Prediction
Zhiqiang Liu, Yin Hua, Mingyang Chen +4
Real-world knowledge graphs (KGs) contain not only standard triple-based facts, but also more complex, heterogeneous types of facts, such as hyper-relational facts with auxiliary k…