# H4G: Unlocking Faithful Inference for Zero-Shot Graph Learning in Hyperbolic Space

arXiv: https://arxiv.org/abs/2510.12094 · HTML: https://mlsanity.com/paper/2510.12094/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 52 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Builds on (most cited first, 22 of 22)

- [Open Graph Benchmark: Datasets for Machine Learning on Graphs](https://mlsanity.com/paper/2005.00687/) (2020)
- [Revisiting Semi-Supervised Learning with Graph Embeddings](https://mlsanity.com/paper/1603.08861/) (2016)
- [Image-based Recommendations on Styles and Substitutes](https://mlsanity.com/paper/1506.04757/) (2015)
- [One for All: Towards Training One Graph Model for All Classification Tasks](https://mlsanity.com/paper/2310.00149/) (2023)
- [Language is All a Graph Needs](https://mlsanity.com/paper/2308.07134/) (2023)
- [LLaGA: Large Language and Graph Assistant](https://mlsanity.com/paper/2402.08170/) (2024)
- [GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networks](https://mlsanity.com/paper/2302.08043/) (2023)
- [GraphText: Graph Reasoning in Text Space](https://mlsanity.com/paper/2310.01089/) (2023)
- [Graph Prototypical Networks for Few-shot Learning on Attributed Networks](https://mlsanity.com/paper/2006.12739/) (2020)
- [ZeroG: Investigating Cross-dataset Zero-shot Transferability in Graphs](https://mlsanity.com/paper/2402.11235/) (2024)
- [Fine-tuning large language models for domain adaptation: Exploration of training strategies, scaling, model merging and synergistic capabilities](https://mlsanity.com/paper/2409.03444/) (2024)
- [LLMs as Zero-shot Graph Learners: Alignment of GNN Representations with LLM Token Embeddings](https://mlsanity.com/paper/2408.14512/) (2024)
- [GraphInstruct: Empowering Large Language Models with Graph Understanding and Reasoning Capability](https://mlsanity.com/paper/2403.04483/) (2024)
- [GraphCLIP: Enhancing Transferability in Graph Foundation Models for Text-Attributed Graphs](https://mlsanity.com/paper/2410.10329/) (2024)
- [Efficient Tuning and Inference for Large Language Models on Textual Graphs](https://mlsanity.com/paper/2401.15569/) (2024)
- [ConGraT: Self-Supervised Contrastive Pretraining for Joint Graph and Text Embeddings](https://mlsanity.com/paper/2305.14321/) (2023)
- [GraphInsight: Unlocking Insights in Large Language Models for Graph Structure Understanding](https://mlsanity.com/paper/2409.03258/) (2024)
- [A Survey of Few-Shot Learning on Graphs: from Meta-Learning to Pre-Training and Prompt Learning](https://mlsanity.com/paper/2402.01440/) (2024)
- [Tree-Wasserstein Distance for High Dimensional Data with a Latent Feature Hierarchy](https://mlsanity.com/paper/2410.21107/) (2024)
- [Curvature Learning for Generalization of Hyperbolic Neural Networks](https://mlsanity.com/paper/2508.17232/) (2025)
- [Graph Contrastive Learning Meets Graph Meta Learning: A Unified Method for Few-shot Node Tasks](https://mlsanity.com/paper/2309.10376/) (2023)
- [A Geometry-Aware Algorithm to Learn Hierarchical Embeddings in Hyperbolic Space](https://mlsanity.com/paper/2407.16641/) (2024)

## Built on by (most cited first, 1 of 1)

- [ReconMIL: Synergizing Latent Space Reconstruction with Bi-Stream Mamba for Whole Slide Image Analysis](https://mlsanity.com/paper/2603.19925/) (2026)

## Abstract (arXiv)

Text-attributed graphs are widely used across domains, offering rich opportunities for zero-shot learning via graph-text alignment. However, existing methods struggle with tasks requiring fine-grained pattern recognition, particularly on heterophilic graphs. Through empirical and theoretical analysis, we identify an over-abstraction problem: current approaches operate at excessively large hyperbolic radii, compressing multi-scale structural information into uniform high-level abstractions. This abstraction-induced information loss obscures critical local patterns essential for accurate predictions. By analyzing embeddings in hyperbolic space, we demonstrate that optimal graph learning requires faithful preservation of fine-grained structural details, better retained by representations positioned closer to the origin. To address this, we propose H4G, a framework that systematically reduces embedding radii using learnable block-diagonal scaling matrices and M"obius matrix multiplication. This approach restores access to fine-grained patterns while maintaining global receptive ability with minimal computational overhead. Experiments show H4G achieves state-of-the-art zero-shot performance with 12.8% improvement on heterophilic graphs and 8.4% on homophilic graphs, confirming that radius reduction enables faithful multi-scale representation for advancing zero-shot graph learning.