Research map / Large language models / Retrieval and knowledge
T-Retriever: Tree-based Hierarchical Retrieval Augmented Generation for Textual Graphs
Chunyu Wei, Huaiyu Qin, Siyuan He, Yunhai Wang, Yueguo Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| jointly optimize structure and semantics (S2-Entropy) instead of semantic-only (ablation) | T-Retriever · WebQSP | accuracy: 0.6319 -> 0.7715 | Improved |
| jointly optimize structure and semantics (S2-Entropy) instead of structural-only (ablation) | T-Retriever · WebQSP | accuracy: 0.7154 -> 0.7715 | Improved |
| structural entropy contributes more than semantic entropy alone (ablation) | T-Retriever · WebQSP | accuracy: 0.6319 (semantic) vs 0.7154 (structural) | Mixed |
| use T-Retriever instead of ArchRAG | T-Retriever · BookGraphs | accuracy: 0.7108 -> 0.7579 | Improved |
| use T-Retriever instead of ArchRAG | T-Retriever · SceneGraphs | accuracy: 0.8212 -> 0.8507 | Improved |
| use T-Retriever instead of fine-tuned G-Retriever w/ LoRA | T-Retriever · WebQSP | accuracy: 0.7186 -> 0.7715 | Improved |
| use semantics-first RAPTOR hierarchy ignoring topology | RAPTOR · WebQSP | accuracy: 0.7254 -> 0.7114 (vs GRAG w/ LoRA) | Hurt |
| add hierarchy (structure-first ArchRAG) over flat methods | ArchRAG · WebQSP | accuracy: 0.7254 -> 0.7533 | Improved |
| increase encoding tree depth L (ablation) | T-Retriever · SceneGraphs, WebQSP, BookGraphs | accuracy | Improved |
| retrieve more subgraphs (k beyond about 6) on larger graphs (ablation) | T-Retriever · larger graphs | accuracy | Mixed |
| set lambda around 1.0 to 1.5 (ablation) | T-Retriever · WebQSP | accuracy | Improved |
| replace G-Retriever retrieval with encoding-tree retrieval | T-Retriever · WebQSP | avg tokens: 721.0 -> 114.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- ArchRAG — ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation (AAAI 2026) · better on 3 of 3 shared results
- RAPTOR — RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval (ICLR 2024) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Simple is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation (ICLR 2025 · similarity 0.91)
- Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation (ICLR 2025 · similarity 0.90)
- A Systematic Exploration of Knowledge Graph Alignment with Large Language Models in Retrieval Augmented Generation (AAAI 2025 · similarity 0.90)
- HyperGraphRAG: Retrieval-Augmented Generation via Hypergraph-Structured Knowledge Representation (NeurIPS 2025 · similarity 0.89)
- Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need? (NeurIPS 2025 · similarity 0.89)
- GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
