DANCE: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs

Zekai Chen, Haodong Lu, Xunkai Li, Henan Sun, Jia Li, Hongchao Qin, Rong-Hua Li, Guoren Wang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:15861-15884, 2026.

Abstract

Federated graph learning (FGL) enables collaborative training on graph data across multiple clients. With the rise of large language models (LLMs), textual attributes in FGL graphs are gaining attention. Text-attributed graph federated learning (TAG-FGL) improves FGL by explicitly leveraging LLMs to process and integrate these textual features. However, current TAG-FGL methods face three main challenges: (1) Overhead. LLMs for processing long texts incur high token and computation costs. To make TAG-FGL practical, we introduce graph condensation (GC) to reduce computation load, but this choice also brings new issues. (2) Suboptimal. To reduce LLM overhead, we introduce GC into TAG-FGL by compressing multi-hop texts/neighborhoods into a condensed core with fixed LLM surrogates (summaries/embeddings). However, this one-shot condensation is often not client-adaptive, leading to suboptimal performance. (3) Interpretability. LLM-based condensation further introduces a black-box bottleneck: summaries lack faithful attribution and clear grounding to specific source spans, making local inspection and auditing difficult. To address the above issues, we propose DANCE, a new TAG-FGL paradigm with GC. To improve suboptimal performance, DANCE performs round-wise, model-in-the-loop condensation refresh using the latest global model. To enhance interpretability, DANCE preserves provenance by storing locally inspectable evidence packs that trace predictions to selected neighbors and source text spans. Across 8 TAG datasets, DANCE improves accuracy by 2.33% at an 8% condensation ratio, with 33.42% fewer tokens per condensed node than TAG-FGL baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26cr, title = {{DANCE}: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs}, author = {Chen, Zekai and Lu, Haodong and Li, Xunkai and Sun, Henan and Li, Jia and Qin, Hongchao and Li, Rong-Hua and Wang, Guoren}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {15861--15884}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/chen26cr/chen26cr.pdf}, url = {https://proceedings.mlr.press/v306/chen26cr.html}, abstract = {Federated graph learning (FGL) enables collaborative training on graph data across multiple clients. With the rise of large language models (LLMs), textual attributes in FGL graphs are gaining attention. Text-attributed graph federated learning (TAG-FGL) improves FGL by explicitly leveraging LLMs to process and integrate these textual features. However, current TAG-FGL methods face three main challenges: (1) Overhead. LLMs for processing long texts incur high token and computation costs. To make TAG-FGL practical, we introduce graph condensation (GC) to reduce computation load, but this choice also brings new issues. (2) Suboptimal. To reduce LLM overhead, we introduce GC into TAG-FGL by compressing multi-hop texts/neighborhoods into a condensed core with fixed LLM surrogates (summaries/embeddings). However, this one-shot condensation is often not client-adaptive, leading to suboptimal performance. (3) Interpretability. LLM-based condensation further introduces a black-box bottleneck: summaries lack faithful attribution and clear grounding to specific source spans, making local inspection and auditing difficult. To address the above issues, we propose DANCE, a new TAG-FGL paradigm with GC. To improve suboptimal performance, DANCE performs round-wise, model-in-the-loop condensation refresh using the latest global model. To enhance interpretability, DANCE preserves provenance by storing locally inspectable evidence packs that trace predictions to selected neighbors and source text spans. Across 8 TAG datasets, DANCE improves accuracy by 2.33% at an 8% condensation ratio, with 33.42% fewer tokens per condensed node than TAG-FGL baselines.} }
Endnote
%0 Conference Paper %T DANCE: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs %A Zekai Chen %A Haodong Lu %A Xunkai Li %A Henan Sun %A Jia Li %A Hongchao Qin %A Rong-Hua Li %A Guoren Wang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-chen26cr %I PMLR %P 15861--15884 %U https://proceedings.mlr.press/v306/chen26cr.html %V 306 %X Federated graph learning (FGL) enables collaborative training on graph data across multiple clients. With the rise of large language models (LLMs), textual attributes in FGL graphs are gaining attention. Text-attributed graph federated learning (TAG-FGL) improves FGL by explicitly leveraging LLMs to process and integrate these textual features. However, current TAG-FGL methods face three main challenges: (1) Overhead. LLMs for processing long texts incur high token and computation costs. To make TAG-FGL practical, we introduce graph condensation (GC) to reduce computation load, but this choice also brings new issues. (2) Suboptimal. To reduce LLM overhead, we introduce GC into TAG-FGL by compressing multi-hop texts/neighborhoods into a condensed core with fixed LLM surrogates (summaries/embeddings). However, this one-shot condensation is often not client-adaptive, leading to suboptimal performance. (3) Interpretability. LLM-based condensation further introduces a black-box bottleneck: summaries lack faithful attribution and clear grounding to specific source spans, making local inspection and auditing difficult. To address the above issues, we propose DANCE, a new TAG-FGL paradigm with GC. To improve suboptimal performance, DANCE performs round-wise, model-in-the-loop condensation refresh using the latest global model. To enhance interpretability, DANCE preserves provenance by storing locally inspectable evidence packs that trace predictions to selected neighbors and source text spans. Across 8 TAG datasets, DANCE improves accuracy by 2.33% at an 8% condensation ratio, with 33.42% fewer tokens per condensed node than TAG-FGL baselines.
APA
Chen, Z., Lu, H., Li, X., Sun, H., Li, J., Qin, H., Li, R. & Wang, G.. (2026). DANCE: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:15861-15884 Available from https://proceedings.mlr.press/v306/chen26cr.html.

Related Material