跳轉到

Phase D Round 1A v2 — arXiv Baseline Scan

掃描時間：2026-06-20 01:32

arXiv 候選數：1000 篇（日期：2023-06-01 → 2026-06-20）

Top N：100（其中 0 篇被 exclude 懲罰後仍進榜）

Profile：llm-graphrag-goat

排序方式（v2）

語意相似度（cosine on nomic-embed）再乘上 exclude 三級懲罰係數（取命中的最重一級）：

Tier 命中內容 係數
1 典範衝突 global GraphRAG / community detection / Leiden ×0.3
2 方法論不採用 LoRA・QLoRA・parameter-efficient / continual pretraining ×0.5
3 其他排除 chatbot UI / hiring / careers 等 off-topic ×0.7

各階段狀態：

  • ❌ Stage 4 規則加權（venue tier、citation、recency）
  • ❌ Stage 5 Claude Haiku 評分（R / Q / S 三維度）
  • ❌ Stage 6 Claude Sonnet 深度分析（引用建議、差異化論述）
  • ❌ Stage 7 VPN 下載清單（IEEE/Springer 等付費期刊）

高 similarity ≠ 必然相關（embedding 可能誤判）。請當作 初步篩選，後續會由 Claude 評分修正。

Top 100 候選論文

1. OMD-GraphRAG: Enhancing GraphRAG with Ontology-Guided Extraction, Multi-Dimensional Clustering and Dual-Channel Fusion

🀄 中文彙整

一句話：OMD-GraphRAG 在開源 GraphRAG 基礎上，用本體論引導抽取、多維度社群分群與雙通道檢索融合三招，提升複雜推理與多跳問答的表現。

研究問題：現有 GraphRAG 框架在知識抽取精度、社群報告完整性與檢索效能上仍有不足，如何強化以改善複雜推理、multi-hop 查詢與領域問答？

方法： - 本體論引導知識抽取：用預定義 Schema 指導 LLM 精準辨識領域實體與關係。 - 多維度社群分群：透過對齊補全（alignment completion）、屬性分群、以及多跳關係分群三種策略提升社群完整性。 - 雙通道圖檢索融合：混合「圖檢索」與「社群檢索」兩條通道，平衡問答準確度與效能。 - 在 MultiHop-RAG benchmark 上評測，與 LightRAG 等主流開源方案比較 F1。

主要發現：綜合 F1 優於主流開源方案（如 LightRAG），尤其在推理型（inference）與時序型（temporal）查詢上提升明顯。

對我的關聯：高度相關，直接命中你的 GraphRAG／multi-hop 主題，有幾個可借鏡點： - 多跳關係分群正是針對 multi-hop 推理在社群結構層面做的改良，值得看其如何在分群階段就保留跨節點推理路徑。 - 雙通道（圖 + 社群）檢索融合是檢索層的設計思路，可對照你對知識圖譜增強檢索的架構。 - 本體論／Schema 引導抽取對「提升圖譜品質→改善下游 multi-hop」這條因果鏈是常見但實用的手法。

保留評估：這是工程整合型（三個既有概念的組合）而非方法論突破，且僅在單一 benchmark（MultiHop-RAG）上驗證、主要對手是 LightRAG，泛化性與消融深度待確認。若你要看的是新穎理論貢獻，價值偏中；若你要找可落地的 GraphRAG 強化工程套路與社群分群細節，值得細讀方法章節。

Abstract

Retrieval-Augmented Generation (RAG) systems face significant challenges in complex reasoning, multi-hop queries, and domain-specific QA. While existing GraphRAG frameworks have made progress in structural knowledge organization, they still have limitations in knowledge extraction precision, community report integrity, and retrieval performance. This paper proposes OMD-GraphRAG, an enhanced framework built upon open-source GraphRAG. The framework introduces three core innovations: (1) Ontology-G...

2. HELP: HyperNode Expansion and Logical Path-Guided Evidence Localization for Accurate and Efficient GraphRAG

🀄 中文彙整

一句話：HELP 提出一套兼顧準確度與效率的 GraphRAG 框架，透過「超節點擴展」串接知識三元組成推理路徑、並以預計算的圖-文關聯直接定位證據，在多跳 QA 上達到競爭性表現且最高加速 28.8 倍。

研究問題：現有 Graph-based RAG 在多跳推理時，常受困於昂貴的圖遍歷（如 random walk）與 LLM 生成摘要帶來的語意噪音，導致準確度與效率難以兼得。如何在不犧牲知識完整性的前提下，大幅降低檢索延遲？

方法：兩大核心策略—— 1. HyperNode Expansion（超節點擴展）：迭代式地將知識三元組串接成連貫的推理路徑，並抽象成「超節點（HyperNode）」，以捕捉複雜的結構依賴、確保檢索準確度。 2. Logical Path-Guided Evidence Localization（邏輯路徑導向證據定位）：利用預先計算好的「圖-文（graph-text）關聯」，把推理路徑直接映射回語料庫定位證據，避免昂貴的 random walk 與摘要語意失真，以換取效率。

主要發現：在多個 simple 與 multi-hop QA 基準上取得具競爭力的表現，同時相較主流 Graph-based RAG baseline 達到最高 28.8× 的檢索加速。（摘要未提供具體準確度數字與資料集名稱，需看正文。）

對我的關聯：高度相關，值得細讀。 - 直接針對你關注的 GraphRAG／知識圖譜增強檢索與 multi-hop 推理，且兩個機制都有可借鏡點： - 「把三元組鏈成路徑再抽象為 HyperNode」是一種介於 triplet 與 community summary 之間的中間表示，可作為你設計多跳證據聚合單元的參考。 - 「預計算 graph-text correlation 直接做 evidence localization」是避開線上圖遍歷的工程取巧，對在意延遲的檢索 pipeline 有實務價值。 - 需確認的點（摘要未交代）：HyperNode 的建構成本與離線索引開銷、graph-text 關聯如何預計算、在哪些資料集（如 HotpotQA / 2WikiMultihopQA / MuSiQue）上比較、以及對比的 baseline（GraphRAG、HippoRAG 等）是哪些——這些決定了 28.8× 加速與準確度權衡的可信度。

Abstract

Large Language Models (LLMs) often struggle with inherent knowledge boundaries and hallucinations, limiting their reliability in knowledge-intensive tasks. While Retrieval-Augmented Generation (RAG) mitigates these issues, it frequently overlooks structural interdependencies essential for multi-hop reasoning. Graph-based RAG approaches attempt to bridge this gap, yet they typically face trade-offs between accuracy and efficiency due to challenges such as costly graph traversals and semantic nois...

3. STAR: Semantic-Tuned and Tail-Adaptive Retriever for Graph-Augmented Generation

🀄 中文彙整

一句話：STAR 是一個用於 GraphRAG 的輕量級檢索器，透過「語意微調」與「長尾路徑自適應」兩種學習機制，緩解知識圖譜檢索中的語意捷徑偏誤與長尾路徑偏誤，提升 multi-hop QA 表現。

研究問題：在 GraphRAG 中，輕量檢索器從知識圖譜抽取 multi-hop 推理所需資訊時，圖譜本身語意資訊稀疏會導致兩種偏誤——(1) Semantic Shortcut Bias（語意捷徑偏誤，模型走捷徑、語意建模不足）與 (2) Long-Tail Path Bias（長尾路徑偏誤，罕見路徑被忽略）。如何在保持輕量的前提下緩解這兩種偏誤？

方法：提出 STAR，整合兩個學習範式： - Token-level interaction learning：以 cross-attention 架構聯合建模 query 與 path，並搭配 hard path mining（困難路徑挖掘）機制 → 緩解語意捷徑偏誤。 - Path-weighted contrastive learning：設計帶有 tail-adaptive path weighting（尾部自適應路徑加權）的對比學習目標，優化訓練過程 → 緩解長尾路徑偏誤。

主要發現：在所有 benchmark 上一致優於 baseline，平均檢索效能提升 1.8%、LLM QA 效能提升 2.2%。（提供匿名程式碼連結）

對我的關聯：高度相關。 直接命中你的核心題目（GraphRAG／KG 增強檢索／multi-hop 推理）。可借鏡處： - 它把「圖譜語意稀疏」拆解成兩個可操作的偏誤（語意捷徑、長尾路徑）——這個問題框架本身就值得借用，可作為你分析自家檢索器失效模式的診斷視角。 - 路徑層級的對比學習 + 困難樣本挖掘是 path-based GraphRAG 檢索的具體技術手段，可直接對照／移植。 - cross-attention 做 query–path 細粒度交互（相對於 dual-encoder 的粗粒度匹配），對 multi-hop 路徑排序有參考價值。 - 提升幅度偏小（檢索 1.8%／QA 2.2%），是漸進式改良而非範式突破；若你只想快速掌握「path retriever 的偏誤問題與對比學習解法」，重點讀方法章節（兩個 learning paradigm）與其偏誤的實證分析即可，實驗細節可略。建議細讀。

Abstract

To augment Large Language Models (LLMs) for multi-hop question answering, a mainstream solution within Graph Retrieval Augmented Generation (GraphRAG) leverages lightweight retrievers to efficiently extract information from a given Knowledge Graph (KG). However, existing methods often overlook the inherent challenge of sparse semantic information in graphs. Specifically, our experiments reveal that these methods produce biased retrieval Semantic Shortcut Bias and Long-Tail Path Bias, leading to ...

4. RLM-on-KG: Heuristics First, LLMs When Needed: Adaptive Retrieval Control over Mention Graphs for Scattered Evidence

🀄 中文彙整

一句話：本文提出 RLM-on-KG，將 LLM 當作在「即時建構的 RDF 提及圖（mention graph）」上自主導航的檢索控制器，並系統性地回答「何時 LLM 控制會勝過規則式遍歷」這個問題——答案是取決於證據分散程度與工具呼叫能力。

研究問題：在知識圖譜探索中，以 LLM 作為自主導航控制器，相較於規則式（heuristic）遍歷，何時、在什麼條件下才真正有優勢？

方法： - 建構 entity-first、查詢時（query-time）的多跳探索系統，使用確定性圖建構＋固定工具集，與依賴離線 LLM 索引的傳統 GraphRAG pipeline 不同。 - 核心架構洞見：將候選發現（candidate discovery）與排序（ranking）分離——LLM 負責擴大探索廣度，最終證據選擇則交給純向量 re-ranking。 - 在 GraphRAG-Bench Novel（519 題）上比較三種設定：規則式 heuristic baseline、GraphRAG-local 變體、LLM 控制器（Gemini 2.0 Flash 與較強的 Claude Haiku 4.5）。 - 跨尺度驗證使用 MuSiQue；並按「黃金證據分散在幾個 chunk」分層分析。

主要發現： - 條件式優勢：LLM 控制的價值取決於證據分散度與工具呼叫成熟度，而非對 GraphRAG 的全面勝出（作者明確自我設限）。 - Gemini 2.0 Flash：對 heuristic +2.47 pp F1（p<0.0001 顯著），但對 GraphRAG-local 僅 +0.16 pp（不顯著）。 - 較強的 Claude Haiku 4.5：對 heuristic 提升至 +4.37 pp（p<0.001），對 GraphRAG-local 也達 +2.42 pp 顯著改善。 - 證據分散時收益最大：證據散落在 6–10 個 chunk 時 +3.21 pp，集中時僅 +1.85 pp。 - MuSiQue 上 LLM-over-heuristic 優勢可遷移，但在每題圖較小的情況下如預期衰減。 - 額外貢獻：探索軌跡（exploration traces）可作為結構化資料品質的壓力測試工具，產出涵蓋率、連通性、來源溯源、可查詢性的診斷。

對我的關聯：高度相關，值得細讀。 對「GraphRAG／知識圖譜增強檢索、multi-hop 推理」研究有多個直接可借鏡點： 1. 最有價值的可借鏡點是把「候選發現 vs. 排序」明確解耦——LLM 做廣度探索、純向量做最終 re-ranking。這是一個可直接套用到自己 GraphRAG pipeline 的架構原則。 2. 證據分散度分層分析（6–10 chunk 散落 vs. 集中）提供了一個評估 multi-hop 檢索何時真正有用的乾淨切片方法，可借來設計自己的 benchmark 分層與消融。 3. query-time 確定性圖建構 + 固定工具集，相對於 offline LLM indexing，是一條值得對照的設計分支，有助於釐清「LLM 該用在索引還是導航」。 4. 「LLM 控制 vs. heuristic 遍歷」的條件式框架與顯著性檢定（含 controller 強度的影響）對如何嚴謹報告 GraphRAG 增益、避免過度宣稱很有參考價值。 5. 提及圖以 RDF 編碼、探索軌跡作為資料品質診斷，對知識圖譜建構與評估也有外溢價值。

唯一須留意：核心主張是「LLM 控制 vs. heuristic」，並非宣稱全面勝過 GraphRAG（對 GraphRAG-local 的優勢只在較強 controller 下才顯著），引用時別誇大其結論。

Abstract

When does an LLM controller outperform rule-based traversal for knowledge graph exploration? We study this question through RLM-on-KG, a retrieval system that treats an LLM as an autonomous navigator over an RDF-encoded mention graph for grounded question answering. Unlike GraphRAG pipelines that rely on offline LLM indexing, RLM-on-KG performs entity-first, multi-hop exploration at query time using deterministic graph construction and a fixed tool set. Our central finding is a conditional advan...

5. Do We Still Need GraphRAG? Benchmarking RAG and GraphRAG for Agentic Search Systems

🀄 中文彙整

一句話：作者提出 RAGSearch 統一基準，公平比較 dense RAG 與 GraphRAG 作為「agentic search（多輪動態檢索）」底層檢索基礎設施的表現，回答「有了 agentic search 是否還需要昂貴的 GraphRAG」這個問題。

研究問題：當檢索從靜態／單次（one-shot）轉為 agentic 的多輪動態檢索與序列決策後，agentic search 透過互動引入的「隱式結構」，能否補償 GraphRAG 顯式圖結構的缺席，從而降低對昂貴 GraphRAG pipeline 的需求？

方法： - 建立 RAGSearch 統一基準，把 dense RAG 與代表性 GraphRAG 方法都當作「agentic search 之下的檢索基礎設施」來評測。 - 涵蓋 training-free 與 training-based（含 RL） 兩類 agentic 推理，橫跨多個 QA 基準。 - 為求公平與可重現：統一 LLM backbone、檢索預算（retrieval budget）、推理協定，並在完整測試集上報告。 - 評估面向不只答案正確率，還包括離線預處理成本、線上推理效率、穩定性。

主要發現： - Agentic search 大幅提升 dense RAG，並縮小其與 GraphRAG 的差距，在 RL-based 設定下尤其明顯。 - 但 GraphRAG 在複雜 multi-hop 推理上仍有優勢，且當離線成本被攤提後，其 agentic search 行為更穩定。 - 結論：顯式圖結構與 agentic search 是互補角色，論文據此給出現代 agentic RAG 系統的檢索設計實務建議。

對我的關聯：高度相關，值得細讀。 這篇直接命中你關注的 GraphRAG／知識圖譜增強檢索與 multi-hop 推理三個核心： - 它把「GraphRAG vs. dense RAG」這個常被混淆的比較，放到受控、可重現的統一框架下（固定 backbone／檢索預算／協定、跑完整測試集），方法論本身可借鏡——很多 GraphRAG 論文比較條件不一致，這裡可當作公平評測的範本。 - 明確指出 GraphRAG 的價值區間：在 multi-hop 複雜推理仍勝出，這對你判斷「何時該投入圖結構建構成本」很有參考價值。 - 提供 multi-hop 上顯式圖結構 vs. agentic 多輪檢索隱式結構的取捨證據與成本（offline/online/穩定性）量化，可直接用於你研究中的 baseline 選擇與設計論證。 - 可借鏡點：RAGSearch 基準本身、RL-based agentic 與 GraphRAG 結合的設定，以及「成本攤提後穩定性」這個常被忽略的評估維度。

建議讀法：重點看其評測協定設計、multi-hop 子集的拆解結果，以及 GraphRAG 在 agentic 設定下的穩定性分析。

Abstract

Retrieval-augmented generation (RAG) and its graph-based extensions (GraphRAG) are effective paradigms for improving large language model (LLM) reasoning by grounding generation in external knowledge. However, most existing RAG and GraphRAG systems operate under static or one-shot retrieval, where a fixed set of documents is provided to the LLM in a single pass. In contrast, recent agentic search systems enable dynamic, multi-round retrieval and sequential decision-making during inference, and h...

6. Beyond Chunk-Local Extraction: Cross-Chunk Graph Augmentation for GraphRAG

🀄 中文彙整

一句話：CrossAug 用 GNN 引導、在離線索引階段補上「跨 chunk 關係」，讓 GraphRAG 的知識圖譜不再只侷限於單一段落內的抽取。

研究問題：現有 GraphRAG 框架只在單一 chunk 內抽取實體與關係，導致證據橫跨多個段落的「跨 chunk 關係」系統性地從索引中缺失；而用 LLM 窮舉所有 chunk 組合來補回這些關係，會因組合爆炸而不可行。如何在可負擔成本下補全這類跨段落關係結構？

方法：CrossAug 是一個離線（query 前）的圖譜增強步驟，三個組件—— 1. 自監督圖譜破壞（graph corruption）：人工製造缺失邊作為訓練監督訊號； 2. 拓撲感知 GNN：對子圖打分，預測哪些區域「可能缺關係」（missingness scoring）； 3. 證據錨定的 LLM 補全：只對高分子圖區域呼叫 LLM 做關係補全，避免全量窮舉。

主要發現：在 3 個 LLM-based GraphRAG 框架、4 個 multi-hop 與長文件 QA benchmark 上測試，CrossAug 一致地提升表現，驗證了跨 chunk 圖譜增強對檢索式問答的效益。程式碼開源（github.com/DonFinliani/CrossAug）。

對我的關聯：高度相關，建議細讀。直接命中你關注的 GraphRAG／知識圖譜增強檢索與 multi-hop 推理交集。幾個可借鏡點：(1) 把「跨 chunk 關係缺失」明確界定為 GraphRAG 索引建構的系統性瓶頸，對 multi-hop 推理尤其關鍵（多跳答案往往就靠跨段落連結）；(2) 用 GNN 做 missingness scoring 來「選擇性補全」而非全量抽取，是控制 LLM 成本的實用工程模式，可移植到你自己的圖譜建構 pipeline;(3) 自監督 graph corruption 產生監督訊號的做法，省去人工標註，值得借用；(4) 它是「offline 索引增強」而非改檢索演算法，與大多數 query-time 方法正交，理論上可疊加。需細讀處：GNN 打分的 precision/recall、補全引入的雜訊與錯誤關係風險、以及相對於其他補邊基線的提升幅度是否顯著。

Abstract

GraphRAG extends retrieval-augmented generation by organizing corpora as explicit knowledge graphs, enabling graph-based retrieval for complex question answering. However, existing frameworks extract entities and relations within individual chunks, leaving cross-chunk relations -- those whose evidence spans multiple passages -- systematically absent from the index. Exhaustive LLM-based recovery of such relations is impractical due to the combinatorial explosion of chunk combinations. We present ...

7. WildGraphBench: Benchmarking GraphRAG with Wild-Source Corpora

🀄 中文彙整

一句話：提出 WildGraphBench，用維基百科文章的外部參考文獻當作「野生（長文、異質、大規模）語料」來評測 GraphRAG，揭露現有圖式檢索在彙整時偏重高層陳述、弱於細節摘要的問題。

研究問題：現有 GraphRAG 基準多用短而精選的段落作為外部知識，無法反映真實場景中「長上下文 + 大規模異質文件」的檢索難度。如何建立更貼近現實的評測，並檢驗 GraphRAG 管線在此條件下的真實能力與弱點？

方法： - 利用維基百科的結構特性——條目敘事由長且異質的外部參考文件支撐——來構建語料：以條目的外部參考作為檢索語料庫，以帶引用連結的陳述句作為 ground truth。 - 跨 12 個頂層主題抽樣文章，產出 1,100 題，分三種複雜度：單事實 QA、多事實 QA、章節級摘要。 - 在多個 baseline 上做實驗比較 GraphRAG 管線表現。

主要發現： - 當證據來自「中等數量」來源時，GraphRAG 對多事實彙整（multi-fact aggregation）有幫助。 - 但這種彙整範式傾向過度強調高層級陳述、犧牲細粒度細節，導致在摘要任務上表現較弱。 - 即現有 GraphRAG 的「聚合」優勢有明確邊界（來源數量、細節保真度）。

對我的關聯：關聯高，值得細讀。 - 對 GraphRAG／知識圖譜增強檢索研究是直接相關的評測工具與診斷：它定位了一個具體失效模式——層級式圖聚合會「往上抽象」而丟失細節，這對你思考圖結構設計（如何同時保留 fine-grained node 與 high-level summary node）有直接啟發。 - multi-hop / multi-fact 推理方面：其「多事實 QA」分層與「以引用連結陳述為 ground truth」的設計，是可借鏡的 multi-hop 證據可追溯標註法，且揭示了「來源數量」這個影響 multi-hop 彙整成敗的變因，可作為你實驗的對照維度。 - 可直接拿來當你方法的評測基準（有開源 repo），尤其用來檢驗你的方法是否能克服「摘要任務細節流失」這個已知弱點。 - 限制提醒：語料來源綁定維基百科 + 外部引用，領域分布與 ground truth 品質受維基編輯生態影響，遷移到專業/封閉領域時的代表性需自行評估。

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) organizes external knowledge as a hierarchical graph, enabling efficient retrieval and aggregation of scattered evidence across multiple documents. However, many existing benchmarks for GraphRAG rely on short, curated passages as external knowledge, failing to adequately evaluate systems in realistic settings involving long contexts and large-scale heterogeneous documents. To bridge this gap, we introduce WildGraphBench, a benchmark designed ...

8. GraphRAG on Consumer Hardware: Benchmarking Local LLMs for Healthcare EHR Schema Retrieval

🀄 中文彙整

一句話：在單張 8GB VRAM 消費級 GPU 上，系統性評測四個本地開源 LLM 跑 Microsoft GraphRAG 做醫療 EHR schema 檢索的可行性與表現差異。

研究問題：在隱私敏感、受法規約束（如醫療 EHR）的資源受限環境下，能否用本地部署的開源 LLM 可靠地運行 GraphRAG？不同模型在索引、知識圖譜建構、查詢延遲、答案品質與幻覺上的表現如何？

方法：以真實 EHR schema 文件為語料，實作 Microsoft GraphRAG pipeline，透過 Ollama 在單張消費級 GPU（8GB VRAM）上部署並比較四個模型——Llama 3.1 (8B)、Mistral (7B)、Qwen 2.5 (7B)、Phi-4-mini (3.8B)。在 global 與 local 兩種檢索模式下，評估索引效率、KG 建構、查詢延遲、答案品質與幻覺。

主要發現： - 能力門檻：約 7B 參數以下的模型無法可靠產生有效結構化輸出、跑不完 pipeline（Phi-4-mini 因 structured-output 錯誤失敗；Mistral 出現退化性重複）。 - 索引與答案品質脫鉤：Llama 3.1 建出最豐富的圖（1,172 個實體），但答案品質最佳的是 Qwen 2.5（3.⅗）。 - 檢索模式：local retrieval 在延遲、事實接地與較低幻覺上都穩定優於 global summarization。 - 結論：GraphRAG 在消費級硬體上可行，但模型選擇與檢索設計對穩健部署至關重要。

對我的關聯：中等偏低，偏工程／部署層面而非方法創新。這篇沒有提出新的 GraphRAG 演算法、圖結構或 multi-hop 推理機制，主要貢獻是「本地小模型跑現成 Microsoft GraphRAG」的 benchmark。對你的研究可借鏡之處有限但有幾點：(1) 它量化了「結構化輸出能力」是 GraphRAG/KG 抽取的隱性瓶頸——entity/relation 抽取需要可靠的 structured output，這對任何 KG 增強檢索的 pipeline 設計都是實務警訊；(2) global（社群摘要）vs local 檢索在事實接地與幻覺上的對比，可作為你評估 multi-hop / 全域推理權衡的參考資料點；(3) 索引豐富度與答案品質脫鉤的觀察，提醒「圖建得大 ≠ 推理好」。但若你關注的是 multi-hop 推理機制本身或圖檢索演算法創新，這篇參考價值不高，可略讀摘要與結論即可。

Abstract

Graph-based Retrieval Augmented Generation (GraphRAG) extends retrieval-augmented generation to support structured reasoning over complex corpora, but its reliability under resource-constrained, privacy-sensitive deployments remains unclear. In healthcare, where Electronic Health Record (EHR) data is complex and strictly regulated, reliance on cloud-based large language models (LLMs) introduces challenges in cost, latency, and compliance. In this work, we present a systematic evaluation of Graph...

9. Toward Robust GraphRAG: Mitigating Retrieval Drift and Hallucination from Imperfect Knowledge Graphs

🀄 中文彙整

一句話：CS-RAG 透過在「檢索階段」處理不完美知識圖譜（而非修圖譜），緩解 GraphRAG 在 multi-hop 推理時的檢索漂移與幻覺問題。

研究問題：現有 GraphRAG 多半預設 LLM 建構的知識圖譜能為證據鏈提供可靠結構支撐，但這假設常不成立。作者問：當 KG 同時存在「虛假雜訊（spurious noise）」與「資訊不完整（incomplete information）」時，如何讓檢索仍保持穩健？

方法：提出 CS-RAG 框架，核心三步： 1. 查詢規劃：把 query 拆成有序、可執行的原子約束（atomic constraints）序列。 2. 細粒度檢索：以 anchor-aware 與 relation-aware 方式檢索，把證據限縮在每一跳（hop）的預期語意內，抑制漂移到「看似合理但無支撐」的 triple。 3. 充分性檢查（sufficiency check）：判斷檢索到的證據是否足以安全地產生變數綁定（variable binding）以供後續傳播；若結構支撐不足，則啟動「文本回退（textual recovery）」，避免在不足的圖結構上硬性續接而產生幻覺。 重點是繞過 KG 修復，直接在檢索時容錯。

主要發現：在三個 multi-hop QA benchmark 上，CS-RAG 對「圖譜建構器（builder）選擇」較不敏感，且在受控注入 KG 缺陷（雜訊／不完整）下仍維持穩定表現。並透過實證分析確認了兩種 KG 問題模式各自對應的失效機制（noise→retrieval drift；incomplete→retrieval hallucination）。

對我的關聯：高度相關。對你的 GraphRAG／multi-hop 研究有幾個直接可借鏡點： - 問題框架化：把 KG 缺陷明確二分為 spurious noise 與 incomplete information，並各自對應到 retrieval drift 與 retrieval hallucination——這套診斷語彙可直接用於你分析自家 pipeline 的失效模式。 - 方法論轉向：主張「檢索時容錯」優於「事前修圖譜」，是值得對照的設計哲學（與 KG completion／refinement 路線形成對比實驗）。 - 可遷移技巧：query→有序原子約束的規劃、anchor/relation-aware 的 per-hop 限縮、以及 sufficiency check + 文本回退的混合（結構不足時退回非結構化文本）——尤其「sufficiency gate」與「structural→textual fallback」是可直接移植到你 multi-hop 檢索器的模組。 - 有公開程式碼（GitHub: myz12138/CS-RAG），便於重現與當 baseline。 建議值得細讀，重點看實證分析章節（兩種失效機制的證據）與 sufficiency check 的具體判準。

Abstract

Graph Retrieval-Augmented Generation (GraphRAG) has become a common approach for multi-hop reasoning by using knowledge graphs (KGs) as structured retrieval indexes. However, most existing GraphRAG methods implicitly assume that LLM-constructed KGs provide structural support for evidence chaining. In this paper, we show that this assumption does not always hold in practice through an empirical analysis, and identify two recurring KG issue modes often overlooked by current retrievers: spurious no...

10. Use Graph When It Needs: Efficiently and Adaptively Integrating Retrieval-Augmented Generation with Graphs

🀄 中文彙整

一句話：提出 EA-GraphRAG，依查詢語法複雜度動態決定走一般 RAG、GraphRAG 或兩者融合，兼顧準確率與延遲。

研究問題：GraphRAG 在實務上常反而輸給 vanilla RAG（準確率下降、延遲過高），根因是「不論查詢難易都一律套用圖檢索」。如何依查詢複雜度自適應地選擇檢索範式？

方法： - 語法特徵建構器：解析每條查詢，抽取一組結構性（句法）特徵。 - 輕量複雜度評分器：將特徵映射為連續的複雜度分數。 - 分數驅動路由策略：低分→走 dense RAG；高分→走 graph-based 檢索；邊界（borderline）案例→用「複雜度感知的 reciprocal rank fusion（RRF）」融合兩者結果。 - 在 2 個 single-hop + 2 個 multi-hop QA benchmark 上評估。

主要發現：在同時含簡單與複雜查詢的混合情境下，EA-GraphRAG 同時提升準確率、降低延遲，達到 SOTA；驗證了「並非所有查詢都該用圖」的核心假設。

對我的關聯：高關聯。直接命中你關注的 GraphRAG／multi-hop 主題，且提供幾個可借鏡點： 1. 自適應路由的視角——把「是否啟用知識圖譜」當成可學習/可評分的決策，而非預設全用，對控制 multi-hop 系統的成本/延遲很實用。 2. 以查詢複雜度區分 single-hop vs multi-hop——可作為你設計 router 或 benchmark 切分的特徵工程參考（語法特徵→複雜度分數）。 3. RRF 融合處理邊界查詢——一個簡單可複現的 RAG/GraphRAG 結果融合 baseline。 - 須留意/待查證：論文未在摘要說明「複雜度評分器」的訓練/標註方式、語法特徵具體清單，以及採用的圖譜建構與 benchmark 名稱（如 HotpotQA、2Wiki 等）——這些是判斷可複現性與是否細讀的關鍵，值得進正文確認。

Abstract

Large language models (LLMs) often struggle with knowledge-intensive tasks due to hallucinations and outdated parametric knowledge. While Retrieval-Augmented Generation (RAG) addresses this by integrating external corpora, its effectiveness is limited by fragmented information in unstructured domain documents. Graph-augmented RAG (GraphRAG) emerged to enhance contextual reasoning through structured knowledge graphs, yet paradoxically underperforms vanilla RAG in real-world scenarios, exhibiting ...

11. LiteSemRAG: Lightweight LLM-Free Semantic-Aware Graph Retrieval for Robust RAG

Abstract

Graph-based Retrieval-Augmented Generation (RAG) has shown great potential for improving multi-level reasoning and structured evidence aggregation. However, existing graph-based RAG frameworks heavily rely on exploiting large language models (LLMs) during indexing and querying, leading to high token consumption, computational cost and latency overhead. In this paper, we propose LiteSemRAG, a lightweight, fully LLM-free, semantic-aware graph retrieval framework. LiteSemRAG constructs a heterogene...

12. LogosKG: Hardware-Optimized Scalable and Interpretable Knowledge Graph Retrieval

Abstract

Knowledge graphs (KGs) are increasingly integrated with large language models (LLMs) to provide structured, verifiable reasoning. A core operation in this integration is multi-hop retrieval, yet existing systems struggle to balance efficiency, scalability, and interpretability. We introduce LogosKG, a novel, hardware-aligned framework that enables scalable and interpretable k-hop retrieval on large KGs by building on symbolic KG formulations and executing traversal as hardware-efficient operatio...

13. EHRAG: Bridging Semantic Gaps in Lightweight GraphRAG via Hybrid Hypergraph Construction and Retrieval

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) enhances LLMs by structuring corpus into graphs to facilitate multi-hop reasoning. While recent lightweight approaches reduce indexing costs by leveraging Named Entity Recognition (NER), they rely strictly on structural co-occurrence, failing to capture latent semantic connections between disjoint entities. To address this, we propose EHRAG, a lightweight RAG framework that constructs a hypergraph capturing both structure and semantic level r...

14. BubbleRAG: Evidence-Driven Retrieval-Augmented Generation for Black-Box Knowledge Graphs

Abstract

Large Language Models (LLMs) exhibit hallucinations in knowledge-intensive tasks. Graph-based retrieval augmented generation (RAG) has emerged as a promising solution, yet existing approaches suffer from fundamental recall and precision limitations when operating over black-box knowledge graphs -- graphs whose schema and structure are unknown in advance. We identify three core challenges that cause recall loss (semantic instantiation uncertainty and structural path uncertainty) and precision los...

15. N2N-GQA: Noise-to-Narrative for Graph-Based Table-Text Question Answering Using LLMs

Abstract

Multi-hop question answering over hybrid table-text data requires retrieving and reasoning across multiple evidence pieces from large corpora, but standard Retrieval-Augmented Generation (RAG) pipelines process documents as flat ranked lists, causing retrieval noise to obscure reasoning chains. We introduce N2N-GQA. To our knowledge, it is the first zeroshot framework for open-domain hybrid table-text QA that constructs dynamic evidence graphs from noisy retrieval outputs. Our key insight is tha...

16. XGRAG: A Graph-Native Framework for Explaining KG-based Retrieval-Augmented Generation

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) extends traditional RAG by using knowledge graphs (KGs) to give large language models (LLMs) a structured, semantically coherent context, yielding more grounded answers. However, GraphRAG reasoning process remains a black-box, limiting our ability to understand how specific pieces of structured knowledge influence the final output. Existing explainability (XAI) methods for RAG systems, designed for text-based retrieval, are limited to interpr...

17. GraphRAG for Engineering Diagrams: ChatP&ID Enables LLM Interaction with P&IDs

Abstract

Large Language Models (LLMs) combined with Retrieval-Augmented Generation (RAG) and knowledge graphs offer new opportunities for interacting with engineering diagrams such as Piping and Instrumentation Diagrams (P&IDs). However, directly processing raw images or smart P&ID files with LLMs is often costly, inefficient, and prone to hallucinations. This work introduces ChatP&ID, an agentic framework that enables grounded and cost-effective natural-language interaction with P&IDs using Graph Retrie...

18. S-Path-RAG: Semantic-Aware Shortest-Path Retrieval Augmented Generation for Multi-Hop Knowledge Graph Question Answering

Abstract

We present S-Path-RAG, a semantic-aware shortest-path Retrieval-Augmented Generation framework designed to improve multi-hop question answering over large knowledge graphs. S-Path-RAG departs from one-shot, text-heavy retrieval by enumerating bounded-length, semantically weighted candidate paths using a hybrid weighted $k$-shortest, beam, and constrained random-walk strategy, learning a differentiable path scorer together with a contrastive path encoder and lightweight verifier, and injecting a ...

19. ConRAG: Consensus-Driven Multi-View Retrieval for Multi-Hop Question Answering

Abstract

Retrieval-augmented generation (RAG) has emerged as a promising paradigm for enhancing large language models (LLMs) on multi-hop question answering (QA), which requires reasoning over evidence from multiple documents. Current multi-hop RAG methods generally focus on either query-side task decomposition or corpus-side knowledge graph construction. Despite their progress, these methods still struggle to achieve satisfactory performance on complex multi-hop QA tasks. To this end, we propose ConRAG,...

20. Domain-Specific Knowledge Graphs in RAG-Enhanced Healthcare LLMs

Abstract

Large Language Models (LLMs) generate fluent answers but can struggle with trustworthy, domain-specific reasoning. We evaluate whether domain knowledge graphs (KGs) improve Retrieval-Augmented Generation (RAG) for healthcare by constructing three PubMed-derived graphs: $\mathbb{G}_1$ (T2DM), $\mathbb{G}_2$ (Alzheimer's disease), and $\mathbb{G}_3$ (AD+T2DM). We design two probes: Probe 1 targets merged AD T2DM knowledge, while Probe 2 targets the intersection of $\mathbb{G}_1$ and $\mathbb{G}_2$...

21. IndexRAG: Bridging Facts for Cross-Document Reasoning at Index Time

Abstract

Multi-hop question answering (QA) requires reasoning across multiple documents, yet existing retrieval-augmented generation (RAG) approaches address this either through graph-based methods requiring additional online processing or iterative multi-step reasoning. We present IndexRAG, a novel approach that shifts cross-document reasoning from online inference to offline indexing. IndexRAG identifies bridge entities shared across documents and generates bridging facts as independently retrievable u...

22. StratRAG: A Multi-Hop Retrieval Evaluation Dataset for Retrieval-Augmented Generation Systems

Abstract

We introduce StratRAG, an open-source retrieval evaluation dataset for benchmarking Retrieval-Augmented Generation (RAG) systems on multi-hop reasoning tasks under realistic, noisy document-pool conditions. Derived from HotpotQA (distractor setting), StratRAG comprises 2,200 examples across three question types -- bridge, comparison, and yes-no -- each paired with a pool of 15 candidate documents containing exactly 2 gold documents and 13 topically related distractors. We benchmark three retriev...

23. A phenotype-driven and evidence-governed framework for knowledge graph enrichment and hypotheses discovery in population data

Abstract

Current knowledge graph (KG) construction methods are confirmatory, focusing on recovering known relationships rather than identifying novel or context-dependent nodes. This paper proposes a phenotype-driven and evidence-governed framework that shifts the paradigm toward structured hypothesis discovery and controlled KG expansion. The approach integrates graph neural networks (GNNs) for phenotype discovery, causal inference, probabilistic reasoning and large language models (LLMs) for hypothesis...

24. KGiRAG: An Iterative GraphRAG Approach for Responding Sensemaking Queries

Abstract

Recent literature highlights the potential of graph-based approaches within large language model (LLM) retrieval-augmented generation (RAG) pipelines for answering queries of varying complexity, particularly those that fall outside the LLM's prior knowledge. However, LLMs are prone to hallucination and often face technical limitations in handling contexts large enough to ground complex queries effectively. To address these challenges, we propose a novel iterative, feedback-driven GraphRAG archit...

25. A Human-in-the-Loop, LLM-Centered Architecture for Knowledge-Graph Question Answering

Abstract

Large Language Models (LLMs) excel at language understanding but remain limited in knowledge-intensive domains due to hallucinations, outdated information, and limited explainability. Text-based retrieval-augmented generation (RAG) helps ground model outputs in external sources but struggles with multi-hop reasoning. Knowledge Graphs (KGs), in contrast, support precise, explainable querying, yet require a knowledge of query languages. This work introduces an interactive framework in which LLMs g...

26. SAGE: Structure Aware Graph Expansion for Retrieval of Heterogeneous Data

Abstract

Retrieval-augmented question answering over heterogeneous corpora requires connected evidence across text, tables, and graph nodes. While entity-level knowledge graphs support structured access, they are costly to construct and maintain, and inefficient to traverse at query time. In contrast, standard retriever-reader pipelines use flat similarity search over independently chunked text, missing multi-hop evidence chains across modalities. We propose SAGE (Structure Aware Graph Expansion) framewo...

27. TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning

Abstract

This paper presents an agentic multimodal retrieval-augmented generation (RAG) framework for domain-specific literature reasoning, instantiated on a curated corpus of several thousand papers in intelligent tires, vehicle dynamics, vehicle control, sensing, estimation, and machine learning. Unlike conventional single-pass RAG systems, the proposed architecture uses an autonomous, evidence-gated pipeline that classifies query intent, generates separate text and visual query rewrites, performs hybr...

28. Ex-GraphRAG: Interpretable Evidence Routing for Graph-Augmented LLMs

Abstract

GraphRAG conditions language models on subgraphs retrieved from knowledge graphs, encoded via message-passing GNNs. Because these encoders entangle node contributions through iterated neighborhood aggregation, there is no closed-form way to determine how much each retrieved entity influenced the encoder's output, and therefore no way to faithfully audit what structural evidence actually reached the model. We introduce Ex-GraphRAG, which replaces the GNN encoder with a Multivariate Graph Neural A...

29. ContextRAG: Extraction-Free Hierarchical Graph Construction for Retrieval-Augmented Generation

Abstract

Graph-structured retrieval-augmented generation (RAG) systems can improve answer quality on multi-hop questions, but many current systems rely on large language models (LLMs) to extract entities, relations, and summaries during indexing. These calls add token and wall-clock costs that grow with corpus size. We present ContextRAG, a graph RAG system whose graph topology is constructed without LLM-based entity or relation extraction. ContextRAG derives a fuzzy concept graph over chunk embeddings u...

30. LLMs+Graphs: Toward Graph-Native, Synergistic AI Systems

Abstract

Large Language Models (LLMs) have advanced rapidly, but their limitations in structured and multi-hop reasoning underscore the need for graph-native, synergistic artificial intelligence (AI) systems. Graph-structured data underpins critical applications across social, biological, financial, transportation, web, and knowledge domains, making it essential to understand how LLMs can leverage graph computation for grounded, context-rich inference. Three complementary synergies are emerging: LLMs aug...

31. SCOUT-RAG: Scalable and Cost-Efficient Unifying Traversal for Agentic Graph-RAG over Distributed Domains

Abstract

Graph-RAG improves LLM reasoning using structured knowledge, yet conventional designs rely on a centralized knowledge graph. In distributed and access-restricted settings (e.g., hospitals or multinational organizations), retrieval must select relevant domains and appropriate traversal depth without global graph visibility or exhaustive querying. To address this challenge, we introduce \textbf{SCOUT-RAG} (\textit{\underline{S}calable and \underline{CO}st-efficient \underline{U}nifying \underline{...

32. From Flat to Structural: Enhancing Automated Short Answer Grading with GraphRAG

Abstract

Automated short answer grading (ASAG) is critical for scaling educational assessment, yet large language models (LLMs) often struggle with hallucinations and strict rubric adherence due to their reliance on generalized pre-training. While Rretrieval-Augmented Generation (RAG) mitigates these issues, standard "flat" vector retrieval mechanisms treat knowledge as isolated fragments, failing to capture the structural relationships and multi-hop reasoning essential for complex educational content. T...

33. Query-Efficient Agentic Graph Extraction Attacks on GraphRAG Systems

Abstract

Graph-based retrieval-augmented generation (GraphRAG) systems construct knowledge graphs over document collections to support multi-hop reasoning. While prior work shows that GraphRAG responses may leak retrieved subgraphs, the feasibility of query-efficient reconstruction of the hidden graph structure remains unexplored under realistic query budgets. We study a budget-constrained black-box setting where an adversary adaptively queries the system to steal its latent entity-relation graph. We pro...

34. BridgeRAG: Training-Free Bridge-Conditioned Retrieval for Multi-Hop Question Answering

Abstract

Multi-hop retrieval is not a single-step relevance problem: later-hop evidence should be ranked by its utility conditioned on retrieved bridge evidence, not by similarity to the original query alone. We present BridgeRAG, a training-free, graph-free retrieval method for retrieval-augmented generation (RAG) over multi-hop questions that operationalizes this view with a tripartite scorer s(q,b,c) over (question, bridge, candidate). BridgeRAG separates coverage from scoring: dual-entity ANN expansi...

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) mitigates hallucinations in Large Language Models (LLMs) by grounding them in structured knowledge. However, current GraphRAG methods are constrained by a prevailing \textit{build-then-reason} paradigm, which relies on a static, pre-constructed Knowledge Graph (KG). This paradigm faces two critical challenges. First, the KG's inherent incompleteness often breaks reasoning paths. Second, the graph's low signal-to-noise ratio introduces distrac...

36. UltRAG: a Universal Simple Scalable Recipe for Knowledge Graph RAG

Abstract

Large language models (LLMs) frequently generate confident yet factually incorrect content when used for language generation (a phenomenon often known as hallucination). Retrieval augmented generation (RAG) tries to reduce factual errors by identifying information in a knowledge corpus and putting it in the context window of the model. While this approach is well-established for document-structured data, it is non-trivial to adapt it for Knowledge Graphs (KGs), especially for queries that requir...

37. RPO-RAG: Aligning Small LLMs with Relation-aware Preference Optimization for Knowledge Graph Question Answering

Abstract

Large Language Models (LLMs) have recently demonstrated remarkable reasoning abilities, yet hallucinate on knowledge-intensive tasks. Retrieval-augmented generation (RAG) mitigates this issue by grounding answers in external sources, e.g., knowledge graphs (KGs). However, existing KG-based RAG approaches rely on semantics-unaware path sampling and are weakly aligned with KG reasoning objectives, which limits further accuracy gains. They also feed retrieved paths directly into the reasoner withou...

38. FlowRAG: Synergizing Explicit Reasoning via Frequency-Aware Multi-Granularity Graph Flow

Abstract

Graph-based retrieval-augmented generation (GraphRAG) is effective for knowledge-intensive and multi-hop query tasks; however, many existing methods primarily seed entity-based graphs and rely on implicit semantic relevance propagation. This often (i) under-retrieves when user queries are abstract and semantically sparse at the entity level, and (ii) suffers from brittle multi-hop reasoning, where noisy activations can derail entity-to-entity transitions and corrupt the inferred relation chain, ...

39. STEM: Structure-Tracing Evidence Mining for Knowledge Graphs-Driven Retrieval-Augmented Generation

Abstract

Knowledge Graph-based Question Answering (KGQA) plays a pivotal role in complex reasoning tasks but remains constrained by two persistent challenges: the structural heterogeneity of Knowledge Graphs(KGs) often leads to semantic mismatch during retrieval, while existing reasoning path retrieval methods lack a global structural perspective. To address these issues, we propose Structure-Tracing Evidence Mining (STEM), a novel framework that reframes multi-hop reasoning as a schema-guided graph sear...

40. PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents

Abstract

We introduce PersonalAI 2.0 (PAI-2), a novel framework, designed to enhance large language model (LLM) based systems through integration of external knowledge graphs (KG). The proposed approach addresses key limitations of existing Graph Retrieval-Augmented Generation (GraphRAG) methods by incorporating a dynamic, multistage query processing pipeline. The central point of PAI-2 design is its ability to perform adaptive, iterative information search, guided by extracted entities, matched graph ve...

41. HyperRAG: Reasoning N-ary Facts over Hypergraphs for Retrieval Augmented Generation

Abstract

Graph-based retrieval-augmented generation (RAG) methods, typically built on knowledge graphs (KGs) with binary relational facts, have shown promise in multi-hop open-domain QA. However, their rigid retrieval schemes and dense similarity search often introduce irrelevant context, increase computational overhead, and limit relational expressiveness. In contrast, n-ary hypergraphs encode higher-order relational facts that capture richer inter-entity dependencies and enable shallower, more efficien...

42. Position: How can Graphs Help Large Language Models?

Abstract

With the rapid advancement of large language models (LLMs), classic graph learning tasks have greatly benefited from LLMs, including improved encoding of textual features, more efficient construction of graphs from text, and enhanced reasoning over knowledge graphs. In this paper, we ask a complementary question: How can graphs help LLMs? We address this question from three perspectives: 1) graphs provide an up-to-date knowledge source that helps reduce LLM hallucinations, 2) graph-based prompti...

43. MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation

Abstract

Retrieval-Augmented Generation (RAG) has become an essential method for mitigating hallucinations in Large Language Models (LLMs) by leveraging external knowledge. Although effective for simple queries, traditional RAG struggles with large-scale, unstructured corpora where information is highly fragmented. Graph-based RAG (GraphRAG) incorporates knowledge graphs to capture structural relationships, enabling more comprehensive retrieval for complex reasoning. However, existing GraphRAG methods re...

44. Agentic GraphRAG: Navigating Unstructured Financial Data with Collaborative AI

Abstract

We present a collaborative agentic GraphRAG framework for expert analysis of commercial registry data. Public registries are often formally accessible, yet difficult to use in practice because they combine structured records with large volumes of unstructured legal text. This limits conventional keyword and vector-only retrieval, especially for multi-hop, temporal, and entity-centric investigations. Our approach builds a Neo4j knowledge graph through a three-phase pipeline: (i) deterministic ing...

45. Large Language Models Meet Biomedical Knowledge Graphs for Mechanistically Grounded Therapeutic Prioritization

Abstract

Drug repurposing is often framed as a candidate identification task, but existing approaches provide limited guidance for distinguishing biologically plausible candidates from historically well-connected ones. Here we introduce DrugKLM, a hybrid framework that integrates biomedical knowledge graph structure with large language model-based mechanistic reasoning to enable mechanistically grounded therapeutic prioritization. Across benchmark datasets, DrugKLM outperforms knowledge graph-only and la...

46. Overview of the MedHopQA track at BioCreative IX: track description, participation and evaluation of systems for multi-hop medical question answering

Abstract

Multi-hop question answering (QA) remains a significant challenge in the biomedical domain, requiring systems to integrate information across multiple sources to answer complex questions. To address this problem, the BioCreative IX MedHopQA shared task was designed to benchmark in multi-hop reasoning for large language models (LLMs). We developed a novel dataset of 1,000 challenging QA pairs spanning diseases, genes, and chemicals, with particular emphasis on rare diseases. Each question was con...

47. Beyond Explicit Edges: Robust Reasoning over Noisy and Sparse Knowledge Graphs

Abstract

GraphRAG is increasingly adopted for converting unstructured corpora into graph structures to enable multi-hop reasoning. However, standard graph algorithms rely heavily on static connectivity and explicit edges, often failing in real-world scenarios where knowledge graphs (KGs) are noisy, sparse, or incomplete. To address this limitation, we introduce INSES (Intelligent Navigation and Similarity Enhanced Search), a dynamic framework designed to reason beyond explicit edges. INSES couples LLM-gu...

48. Incorporating contextual information into KGWAS for interpretable GWAS discovery

Abstract

Genome-Wide Association Studies (GWAS) identify associations between genetic variants and disease; however, moving beyond associations to causal mechanisms is critical for therapeutic target prioritization. The recently proposed Knowledge Graph GWAS (KGWAS) framework addresses this challenge by linking genetic variants to downstream gene-gene interactions via a knowledge graph (KG), thereby improving detection power and providing mechanistic insights. However, the original KGWAS implementation r...

49. The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QA

Abstract

Graph-RAG systems achieve strong multi-hop question answering by indexing documents into knowledge graphs, but strong retrieval does not guarantee strong answers. Evaluating KET-RAG, a leading Graph-RAG system, on three multi-hop QA benchmarks (HotpotQA, MuSiQue, 2WikiMultiHopQA), we find that 77% to 91% of questions have the gold answer in the retrieved context, yet accuracy is only 35% to 78%, and 73% to 84% of errors are reasoning failures. We propose two augmentations: (i) SPARQL chain-of-th...

50. GraphER: An Efficient Graph-Based Enrichment and Reranking Method for Retrieval-Augmented Generation

Abstract

Retrieval-augmented generation (RAG) systems that rely on semantic search often fail to retrieve the complete set of evidence for complex queries, particularly when information is distributed across multiple sources. Existing approaches either rely on iterative agentic retrieval, which can be inefficient, or maintain additional structures such as knowledge graphs, which introduce storage and maintenance overhead. In this paper, we propose GraphER, a graph-based enrichment and reranking framework...

51. SCPRM: A Schema-aware Cumulative Process Reward Model for Knowledge Graph Question Answering

Abstract

Large language models excel at complex reasoning, yet evaluating their intermediate steps remains challenging. Although process reward models provide step-wise supervision, they often suffer from a risk compensation effect, where incorrect steps are offset by later correct ones, assigning high rewards to flawed reasoning paths. This issue is further exacerbated in knowledge graph (KG) reasoning, as there may exist multiple paths between the start and end entities in the KGs, and a risky step can...

52. SRAG: RAG with Structured Data Improves Vector Retrieval

Abstract

Retrieval Augmented Generation (RAG) provides the necessary informational grounding to LLMs in the form of chunks retrieved from a vector database or through web search. RAG could also use knowledge graph triples as a means of providing factual information to an LLM. However, the retrieval is only based on representational similarity between a question and the contents. The performance of RAG depends on the numeric vector representations of the query and the chunks. To improve these representati...

53. Retrieving Minimal and Sufficient Reasoning Subgraphs with Graph Foundation Models for Path-aware GraphRAG

Abstract

Graph-based retrieval-augmented generation (GraphRAG) exploits structured knowledge to support knowledge-intensive reasoning. However, most existing methods treat graphs as intermediate artifacts, and the few subgraph-based retrieval methods depend on heuristic rules coupled with domain-specific distributions. They fail in typical cold-start scenarios where data in target domains is scarce, thus yielding reasoning contexts that are either informationally incomplete or structurally redundant. In ...

54. Bounded Path Context: A Controlled Study of Visible Path History in LLM-Based Knowledge Graph Question Answering

Abstract

LLM-based knowledge-graph question answering (KGQA) delegates graph traversal to language models, turning each question into a sequence of local relation-selection decisions repeated across beams and hops. A common but untested default is to serialize the complete partial path into every routing prompt, even though the controller already maintains this path as exact symbolic state. Bounded Path Context (BPC) decouples these two roles: the controller retains full paths in symbolic memory for answ...

55. HyperGraphPro: Progress-Aware Reinforcement Learning for Structure-Guided Hypergraph RAG

Abstract

Graph Retrieval-Augmented Generation (GraphRAG) has emerged as a promising paradigm that organizes external knowledge into structured graphs of entities and relations, enabling large language models (LLMs) to perform complex reasoning beyond text-chunk retrieval. Recent advances have integrated reinforcement learning (RL) into agentic GraphRAG approaches, enabling iterative interactions with knowledge graphs during training. However, existing RL-based methods suffer from two key limitations: (1)...

56. Pruning Minimal Reasoning Graphs for Efficient Retrieval-Augmented Generation

Abstract

Retrieval-augmented generation (RAG) is now standard for knowledge-intensive LLM tasks, but most systems still treat every query as fresh, repeatedly re-retrieving long passages and re-reasoning from scratch, inflating tokens, latency, and cost. We present AutoPrunedRetriever, a graph-style RAG system that persists the minimal reasoning subgraph built for earlier questions and incrementally extends it for later ones. AutoPrunedRetriever stores entities and relations in a compact, ID-indexed code...

57. Deep GraphRAG: A Balanced Approach to Hierarchical Retrieval and Adaptive Integration

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) frameworks face a trade-off between the comprehensiveness of global search and the efficiency of local search. Existing methods are often challenged by navigating large-scale hierarchical graphs, optimizing retrieval paths, and balancing exploration-exploitation dynamics, frequently lacking robust multi-stage re-ranking. To overcome these deficits, we propose Deep GraphRAG, a framework designed for a balanced approach to hierarchical retrieva...

58. A2RAG: Adaptive Agentic Graph Retrieval for Cost-Aware and Reliable Reasoning

Abstract

Graph Retrieval-Augmented Generation (Graph-RAG) enhances multihop question answering by organizing corpora into knowledge graphs and routing evidence through relational structure. However, practical deployments face two persistent bottlenecks: (i) mixed-difficulty workloads where one-size-fits-all retrieval either wastes cost on easy queries or fails on hard multihop cases, and (ii) extraction loss, where graph abstraction omits fine-grained qualifiers that remain only in source text. We presen...

59. KEPo: Knowledge Evolution Poison on Graph-based Retrieval-Augmented Generation

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) constructs the Knowledge Graph (KG) from external databases to enhance the timeliness and accuracy of Large Language Model (LLM) generations. However, this reliance on external data introduces new attack surfaces. Attackers can inject poisoned texts into databases to manipulate LLMs into producing harmful target responses for attacker-chosen queries. Existing research primarily focuses on attacking conventional RAG systems. However, such meth...

60. Codebase-Memory: Tree-Sitter-Based Knowledge Graphs for LLM Code Exploration via MCP

Abstract

Large Language Model (LLM) coding agents typically explore codebases through repeated file-reading and grep-searching, consuming thousands of tokens per query without structural understanding. We present Codebase-Memory, an open-source system that constructs a persistent, Tree-Sitter-based knowledge graph via the Model Context Protocol (MCP), parsing 66 languages through a multi-phase pipeline with parallel worker pools, call-graph traversal, impact analysis, and community discovery. Evaluated a...

61. Paper Circle: An Open-source Multi-agent Research Discovery and Analysis Framework

Abstract

The rapid growth of scientific literature has made it increasingly difficult for researchers to efficiently discover, evaluate, and synthesize relevant work. Recent advances in multi-agent large language models (LLMs) have demonstrated strong potential for understanding user intent and are being trained to utilize various tools. In this paper, we introduce Paper Circle, a multi-agent research discovery and analysis system designed to reduce the effort required to find, assess, organize, and unde...

62. MedHopQA: A Disease-Centered Multi-Hop Reasoning Benchmark and Evaluation Framework for LLM-Based Biomedical Question Answering

Abstract

Evaluating large language models (LLMs) in the biomedical domain requires benchmarks that can distinguish reasoning from pattern matching and remain discriminative as model capabilities improve. Existing biomedical question answering (QA) benchmarks are limited in this respect. Multiple-choice formats can allow models to succeed through answer elimination rather than inference, while widely circulated exam-style datasets are increasingly vulnerable to performance saturation and training data con...

63. TIEG-Youpu Solution for NeurIPS 2022 WikiKG90Mv2-LSC

Abstract

WikiKG90Mv2 in NeurIPS 2022 is a large encyclopedic knowledge graph. Embedding knowledge graphs into continuous vector spaces is important for many practical applications, such as knowledge acquisition, question answering, and recommendation systems. Compared to existing knowledge graphs, WikiKG90Mv2 is a large scale knowledge graph, which is composed of more than 90 millions of entities. Both efficiency and accuracy should be considered when building graph embedding models for knowledge graph a...

64. Plasma GraphRAG: Physics-Grounded Parameter Selection for Gyrokinetic Simulations

Abstract

Accurate parameter selection is fundamental to gyrokinetic plasma simulations, yet current practices rely heavily on manual literature reviews, leading to inefficiencies and inconsistencies. We introduce Plasma GraphRAG, a novel framework that integrates Graph Retrieval-Augmented Generation (GraphRAG) with large language models (LLMs) for automated, physics-grounded parameter range identification. By constructing a domain-specific knowledge graph from curated plasma literature and enabling struc...

65. Open Biomedical Knowledge Graphs at Scale: Construction, Federation, and AI Agent Access with Samyama Graph Database

Abstract

Biomedical knowledge is fragmented across siloed databases -- Reactome for pathways, STRING for protein interactions, ClinicalTrials.gov for study registries, DrugBank for drug vocabularies, DGIdb for drug-gene interactions, SIDER for side effects. We present three open-source biomedical knowledge graphs -- Pathways KG (118,686 nodes, 834,785 edges from 5 sources), Clinical Trials KG (7,774,446 nodes, 26,973,997 edges from 5 sources), and Drug Interactions KG (32,726 nodes, 191,970 edges from 3 ...

66. KNIGHT: Knowledge Graph-Driven Multiple-Choice Question Generation with Adaptive Hardness Calibration

Abstract

With the rise of large language models (LLMs), they have become instrumental in applications such as Retrieval-Augmented Generation (RAG). Yet evaluating these systems remains bottlenecked by the time and cost of building specialized assessment datasets. We introduce KNIGHT, an LLM-based, knowledge-graph-driven framework for generating multiple-choice question (MCQ) datasets from external sources. KNIGHT constructs a topic-specific knowledge graph, a structured and parsimonious summary of entiti...

67. Towards Robust Retrieval-Augmented Generation Based on Knowledge Graph: A Comparative Analysis

Abstract

Retrieval-Augmented Generation (RAG) was introduced to enhance the capabilities of Large Language Models (LLMs) beyond their encoded prior knowledge. This is achieved by providing LLMs with an external source of knowledge, which helps reduce factual hallucinations and enables access to new information not available during pretraining. However, inconsistent retrieved information can negatively affect LLM responses. The Retrieval-Augmented Generation Benchmark (RGB) was introduced to evaluate the ...

68. Enhancing Metacognitive AI: Knowledge-Graph Population with Graph-Theoretic LLM Enrichment

Abstract

Metacognition-the ability to monitor one's own knowledge state, spot gaps, and autonomously fill them--remains largely absent from modern AI. Here, we present MetaKGEnrich, a fully automated pipeline that endows large language model (LLM) applications with self-directed knowledge repair. The system (i) builds knowledge graphs from a seed query, (ii) detects sparse regions via seven graph metrics, (iii) has GPT-4o generate targeted questions, (iv) retrieves web evidence with Tavily and ingests it...

69. GROUNDEDKG-RAG: Grounded Knowledge Graph Index for Long-document Question Answering

Abstract

Retrieval-augmented generation (RAG) systems have been widely adopted in contemporary large language models (LLMs) due to their ability to improve generation quality while reducing the required input context length. In this work, we focus on RAG systems for long-document question answering. Current approaches suffer from a heavy reliance on LLM descriptions resulting in high resource consumption and latency, repetitive content across hierarchical levels, and hallucinations due to no or limited g...

70. GraphScout: Empowering Large Language Models with Intrinsic Exploration Ability for Agentic Graph Reasoning

Abstract

Knowledge graphs provide structured and reliable information for many real-world applications, motivating increasing interest in combining large language models (LLMs) with graph-based retrieval to improve factual grounding. Recent Graph-based Retrieval-Augmented Generation (GraphRAG) methods therefore introduce iterative interaction between LLMs and knowledge graphs to enhance reasoning capability. However, existing approaches typically depend on manually designed guidance and interact with kno...

71. GAAMA: Graph Augmented Associative Memory for Agents

Abstract

AI agents that interact with users across multiple sessions require persistent long-term memory to maintain coherent, personalized behavior. Current approaches either rely on flat retrieval-augmented generation (RAG), which loses structural relationships among memories, or use entity-centric knowledge graphs that suffer from mega-hub effects in conversational data, diluting graph-based relevance propagation. We propose GAAMA, a graph-augmented associative memory for agents that constructs a conc...

72. Frugal Knowledge Graph Construction with Local LLMs: A Zero-Shot Pipeline, Self-Consistency and Wisdom of Artificial Crowds

Abstract

This paper presents an empirical study of a multi-model zero-shot pipeline for knowledge graph construction and exploitation, executed entirely through local inference on consumer-grade hardware. We propose a reproducible evaluation framework integrating two external benchmarks (DocRED, HotpotQA), WebQuestionsSP-style synthetic data, and the RAGAS evaluation framework in an automated pipeline. On 500 document-level relations, our system achieves an F1 of 0.70 $\pm$ 0.041 in zero-shot, compared t...

73. Mixture of Demonstrations for Textual Graph Understanding and Question Answering

Abstract

Textual graph-based retrieval-augmented generation (GraphRAG) has emerged as a powerful paradigm for enhancing large language models (LLMs) in domain-specific question answering. While existing approaches primarily focus on zero-shot GraphRAG, selecting high-quality demonstrations is crucial for improving reasoning and answer accuracy. Furthermore, recent studies have shown that retrieved subgraphs often contain irrelevant information, which can degrade reasoning performance. In this paper, we p...

74. LEC-KG: An LLM-Embedding Collaborative Framework for Domain-Specific Knowledge Graph Construction -- A Case Study on SDGs

Abstract

Constructing domain-specific knowledge graphs from unstructured text remains challenging due to heterogeneous entity mentions, long-tail relation distributions, and the absence of standardized schemas. We present LEC-KG, a bidirectional collaborative framework that integrates the semantic understanding of Large Language Models (LLMs) with the structural reasoning of Knowledge Graph Embeddings (KGE). Our approach features three key components: (1) hierarchical coarse-to-fine relation extraction t...

75. ChronoMedKG: A Temporally-Grounded Biomedical Knowledge Graph and Benchmark for Clinical Reasoning

Abstract

Biomedical knowledge graphs (KGs) treat disease associations as static facts, but temporal information is crucial for clinical reasoning, e.g., a symptom diagnostic of one disease at age 3 may imply a different disease at age 13. Existing KGs such as PrimeKG, Hetionet, and iKraph do not encode when a finding becomes clinically relevant over the course of a disease. This limits their usefulness for longitudinal clinical reasoning and retrieval augmentation. We introduce ChronoMedKG, a temporal ...

76. UniD$^3$: A Knowledge Graph-Enhanced RAG Framework for Drug-Disease Discovery and Reasoning

Abstract

Systematic characterization of drug-disease relationships is essential for drug discovery and repurposing, yet is hindered by the heterogeneity and rapid growth of biomedical literature. Existing datasets rely on labor-intensive curation and are often incomplete, while LLM-only approaches suffer from hallucination and weak evidence grounding. We introduce UniD$^3$, a unified framework that integrates Large Language Models with Knowledge Graph-enhanced Retrieval-Augmented Generation (KG-RAG) to e...

77. Text-Graph Synergy: A Bidirectional Verification and Completion Framework for RAG

Abstract

Retrieval-Augmented Generation (RAG) has become a core paradigm for enhancing factual grounding and multi-hop reasoning in Large Language Models (LLMs). Traditional text-based RAG often retrieves logically irrelevant pseudo-evidence, while graph-based RAG is frequently hindered by search-time pruning, which may discard potentially valid reasoning paths. Existing hybrid approaches primarily adopt simple evidence concatenation or unidirectional enhancement, which fails to address the fundamental "...

78. PrimeKG-CL: A Continual Graph Learning Benchmark on Evolving Biomedical Knowledge Graphs

Abstract

Biomedical knowledge graphs underwrite drug repurposing and clinical decision support, yet the upstream ontologies they depend on update on independent cycles that add millions of edges and deprecate hundreds of thousands more between releases. Yet existing continual graph learning has been studied almost exclusively on synthetic random splits of static, generic KGs, a regime that cannot reproduce the asynchronous, structured evolution real biomedical KGs undergo. To this end, we introduce Prime...

79. ActiShade: Activating Overshadowed Knowledge to Guide Multi-Hop Reasoning in Large Language Models

Abstract

In multi-hop reasoning, multi-round retrieval-augmented generation (RAG) methods typically rely on LLM-generated content as the retrieval query. However, these approaches are inherently vulnerable to knowledge overshadowing - a phenomenon where critical information is overshadowed during generation. As a result, the LLM-generated content may be incomplete or inaccurate, leading to irrelevant retrieval and causing error accumulation during the iteration process. To address this challenge, we prop...

80. GraphSteal: Structural Knowledge Stealing from Graph RAG via Traversal Reconstruction

Abstract

Retrieval-Augmented Generation (RAG) enhances LLMs by grounding generation in query-relevant external evidence. Beyond unstructured text corpora, Graph RAG integrates knowledge graphs into the retrieval pipeline, enabling LLMs to access entities, relations, and multi-hop dependencies encoded in structured knowledge. However, the same structured knowledge that empowers Graph RAG also creates a new privacy attack surface. We demonstrate that Graph RAG systems can be turned into structural oracles:...

81. LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling

Abstract

Search agent benchmarks exemplified by BrowseComp have rapidly saturated over the past year, with the strongest models surpassing 90% accuracy. Since these benchmarks are predominantly human-authored, annotators lack a global perspective on entity statistics and cannot systematically maximize search space size and structural complexity. This creates a difficulty ceiling that is hard to break. To address this, we introduce LoHoSearch (Long-Horizon Search Agents), a challenging benchmark comprisin...

82. CacheRAG: A Semantic Caching System for Retrieval-Augmented Generation in Knowledge Graph Question Answering

Abstract

The integration of Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) has significantly advanced Knowledge Graph Question Answering (KGQA). However, existing LLM-driven KGQA systems act as stateless planners, generating retrieval plans in isolation without exploiting historical query patterns: analogous to a database system that optimizes every query from scratch without a plan cache. This fundamental design flaw leads to schema hallucinations and limited retrieval coverage. ...

83. GraphRAG-IRL: Personalized Recommendation with Graph-Grounded Inverse Reinforcement Learning and LLM Re-ranking

Abstract

Personalized recommendation requires models that capture sequential user preferences while remaining robust to sparse feedback and semantic ambiguity. Recent work has explored large language models (LLMs) as recommenders and re-rankers, but pure prompt-based ranking often suffers from poor calibration, sensitivity to candidate ordering, and popularity bias. These limitations make LLMs useful semantic reasoners, but unreliable as standalone ranking engines. We present \textbf{GraphRAG-IRL}, a h...

84. PathISE: Learning Informative Path Supervision for Knowledge Graph Question Answering

Abstract

Knowledge Graph Question Answering (KGQA) aims to answer user questions by reasoning over Knowledge Graphs (KGs). Recent KGQA methods mainly follow the retrieval-augmented generation paradigm to ground Large Language Models~(LLMs) with structured knowledge from KGs. However, training effective models to retrieve question-relevant evidence from KGs typically requires high-quality intermediate supervision signals, such as question-relevant paths or subgraphs, which are time- and resource-intensive...

85. HOME-KGQA: A Benchmark Dataset for Multimodal Knowledge Graph Question Answering on Household Daily Activities

Abstract

Large Language Models (LLMs) provide flexible natural language processing capabilities, while knowledge graphs (KGs) offer explicit and structured knowledge. Integrating these two in a complementary manner enables the development of reliable and verifiable AI systems. In particular, knowledge graph question answering (KGQA) has attracted attention as a means to reduce LLM hallucinations and to leverage knowledge beyond the training data. However, existing KGQA benchmark datasets are biased towar...

86. Integrating Graphs, Large Language Models, and Agents: Reasoning and Retrieval

Abstract

Generative AI, particularly Large Language Models, increasingly integrates graph-based representations to enhance reasoning, retrieval, and structured decision-making. Despite rapid advances, there remains limited clarity regarding when, why, where, and what types of graph-LLM integrations are most appropriate across applications. This survey provides a concise, structured overview of the design choices underlying the integration of graphs with LLMs. We categorize existing methods based on their...

87. FastInsight: Fast and Insightful Retrieval via Fusion Operators for Graph RAG

Abstract

Existing Graph RAG methods aiming for insightful retrieval on corpus graphs typically rely on time-intensive processes that interleave Large Language Model (LLM) reasoning. To enable time-efficient insightful retrieval, we propose FastInsight. We first introduce a graph retrieval taxonomy that categorizes existing methods into three fundamental operations: vector search, graph search, and model-based search. Through this taxonomy, we identify two critical limitations in current approaches: the t...

88. GRASP: Graph Agentic Search over Propositions for Multi-hop Question Answering

Abstract

Agentic retrieval improves multi-hop question answering by giving language models autonomy to iteratively gather evidence. Recent work augments these systems with knowledge graphs for structured traversal, but this combination introduces significant cost: expensive graph construction at index time and compounding token usage at inference time. We introduce Graph Agentic Search over Propositions (GRASP), an agentic system that simultaneously optimizes for high accuracy and minimal token usage in ...

89. PrivGemo: Privacy-Preserving Dual-Tower Graph Retrieval for Empowering LLM Reasoning with Memory Augmentation

Abstract

Knowledge graphs (KGs) provide structured evidence that can ground large language model (LLM) reasoning for knowledge-intensive question answering. However, many practical KGs are private, and sending retrieved triples or exploration traces to closed-source LLM APIs introduces leakage risk. Existing privacy treatments focus on masking entity names, but they still face four limitations: structural leakage under semantic masking, uncontrollable remote interaction, fragile multi-hop and multi-entit...

90. ReGraM: Region-First Knowledge Graph Reasoning for Medical Question Answering

Abstract

Recent studies in medical question answering (Medical QA) have actively explored the integration of large language models (LLMs) with biomedical knowledge graphs (KGs) to improve factual accuracy. However, most existing approaches still rely on traversing the entire KG or performing large-scale retrieval, which introduces substantial noise and leads to unstable multi-hop reasoning. We argue that the core challenge lies not in expanding access to knowledge, but in identifying and reasoning over t...

91. MHGraphBench: Knowledge Graph-Grounded Benchmarking of Mental Health Knowledge in Large Language Models

Abstract

Large language models (LLMs) are increasingly used in the mental health domain, yet it remains unclear how well they capture related biomedical knowledge and how reliably they apply it to clinically salient structured judgments. Here, we present a knowledge-graph (KG)-grounded benchmark for assessing LLMs on mental-health entity recognition, relation judgment, and two-hop reasoning. The benchmark is derived from PrimeKG and comprises nine task families with KG-supported answers and controlled ne...

92. Breaking the Static Graph: Context-Aware Traversal for Robust Retrieval-Augmented Generation

Abstract

Recent advances in Retrieval-Augmented Generation (RAG) have shifted from simple vector similarity to structure-aware approaches like HippoRAG, which leverage Knowledge Graphs (KGs) and Personalized PageRank (PPR) to capture multi-hop dependencies. However, these methods suffer from a "Static Graph Fallacy": they rely on fixed transition probabilities determined during indexing. This rigidity ignores the query-dependent nature of edge relevance, causing semantic drift where random walks are dive...

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) advances flat document retrieval by structuring knowledge as relational graphs, enabling more coherent and effective reasoning. However, applying it to specific domains like legal reasoning faces critical challenges. (i) Legal corpora are heterogeneous, containing multi-granular knowledge from cases, articles and interpretations. A flat knowledge graph cannot adequately differentiate between factual details, applied rules, and abstract princi...

94. RAGA: Reading-And-Graph-building-Agent for Autonomous Knowledge Graph Construction and Retrieval-Augmented Generation

Abstract

Existing LLM-driven knowledge graph (KG) construction methods predominantly employ stateless batch processing pipelines, exhibiting structural deficiencies in cross-chunk semantic relation capture, entity disambiguation, and construction process interpretability. These limitations undermine KG quality, retrieval precision, and deployment trust in high-stakes domains. We propose RAGA (Reading And Graph-building Agent), an LLM-based autonomous KG construction and retrieval fusion framework. RAGA...

95. Reliable Graph-RAG for Codebases: AST-Derived Graphs vs LLM-Extracted Knowledge Graphs

Abstract

Retrieval-Augmented Generation for software engineering often relies on vector similarity search, which captures topical similarity but can fail on multi-hop architectural reasoning such as controller to service to repository chains, interface-driven wiring, and inheritance. This paper benchmarks three retrieval pipelines on Java codebases (Shopizer, with additional runs on ThingsBoard and OpenMRS Core): (A) vector-only No-Graph RAG, (B) an LLM-generated knowledge graph RAG (LLM-KB), and (C) a d...

96. HugRAG: Hierarchical Causal Knowledge Graph Design for RAG

Abstract

Retrieval augmented generation (RAG) has enhanced large language models by enabling access to external knowledge, with graph-based RAG emerging as a powerful paradigm for structured retrieval and reasoning. However, existing graph-based methods often over-rely on surface-level node matching and lack explicit causal modeling, leading to unfaithful or spurious answers. Prior attempts to incorporate causality are typically limited to local or single-document contexts and also suffer from informatio...

97. GATHER: Convergence-Centric Hyper-Entity Retrieval for Zero-Shot Cell-Type Annotation

Abstract

Zero-shot single-cell cell-type annotation aims to determine a cell's type from a given set of expressed genes without any training. Existing knowledge-graph-based RAG approaches retrieve evidence by expanding from source entities and relying on iterative LLM reasoning. However, in this setting each query contains tens to hundreds of genes, where no single gene is decisive and the label emerges only from their collective co-occurrence. Such hyper-entity queries fundamentally challenge local, ent...

98. UnWeaving the knots of GraphRAG -- turns out VectorRAG is almost enough

Abstract

One of the key problems in Retrieval-augmented generation (RAG) systems is that chunk-based retrieval pipelines represent the source chunks as atomic objects, mixing the information contained within such a chunk into a single vector. These vector representations are then fundamentally treated as isolated, independent and self-sufficient, with no attempt to represent possible relations between them. Such an approach has no dedicated mechanisms for handling multi-hop questions. Graph-based RAG sys...

99. SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory

Abstract

Long-term memory is becoming a central bottleneck for language agents. Exsting RAG and GraphRAG systems largely treat memory graphs as static retrieval middleware, which limits their ability to recover complete evidence chains from partial cues, exploit reusable graph-structrual roles, and improve the memory itself through downstream feedback. We introduce SAGE, a Self-evolving Agentic Graph-memory Engine that models graph memory as a dynamic long-term memory substrate. SAGE couples two roles: a...

100. SkillGraph: Graph Foundation Priors for LLM Agent Tool Sequence Recommendation

Abstract

LLM agents must select tools from large API libraries and order them correctly. Existing methods use semantic similarity for both retrieval and ordering, but ordering depends on inter-tool data dependencies that are absent from tool descriptions. As a result, semantic-only methods can produce negative Kendall-$τ$ in structured workflow domains. We introduce SkillGraph, a directed weighted execution-transition graph mined from 49,831 successful LLM agent trajectories, which encodes workflow-prece...