{
  "id": 1220,
  "url": "https://arxiv.org/abs/2606.10461v1",
  "title": "ERAlign: Energy-based Representation Alignment of GNNs and LLMs on Text-attributed Graphs",
  "summary": "Text-attributed Graphs (TAGs) incorporate textual node attributes with graph structures to describe rich relational semantics. Recent efforts to integrate Graph Neural Networks (GNNs) and Large Language Models (LLMs) have shown promise for learning on TAGs, yet achieving well-aligned representations remains challenging. Prior studies largely rely on heuristics that perform coarse-grained matching. They lack sufficient constraints and ignore distributional alignment, leading to representation dri",
  "authors": "Xianlin Zeng, Fan Xia, Xiangyu Chen",
  "category": "research",
  "topics": "safety-alignment,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-09T06:16:40.000Z",
  "fetched_at": "2026-07-14T14:15:07.843Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/1220",
  "original_url": "https://arxiv.org/abs/2606.10461v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}