- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper focuses on solving the wording gap problem in GraphRAG, which the entity mentioned in the query may be different in format from the entity in the KG. The authors propose BridgeKG-RAG, which augments the factual graph with typed bridge edges that capture wording relationships between entities. BridgeKG-RAG contains two stages. In the Bridge Construction stage, the framework first a pool of candidate entity pairs based on name similarity and document co-occurrence, and represents each pair with multi-dimensional similarity vector that estimates the similarity between entity names, contexts, role types, neighbors of the entities, relation overlaps, and their co-existence within the same section or documents. It then applies K-Means algorithm to cluster these vectors, and utilizes an LLM to interpret each cluster centroid to either reuse a type from a shared catalog (e.g., alias, abbreviation) or to define a new corpus-specific wording type. The resulting bridge types form a corpus skill. Finally, an LLM labels each candidate pair using its similarity vector and the skill as soft guidance, outputting a bridge type, direction, and confidence score, those with no meaningful relation or with a confidence score lower than the threshold will be discarded. In the traversal and retrieval stage, the system first determines whether the entities mentioned in the original query already include the one where the relevant evidence is organized. If yes, the LLM directly resolve the query with RAG, otherwise, otherwise, it follows the top-k highest-scoring bridges of predicted types to expand the query entities, retrieves query-relevant facts from the expanded neighborhood, and generates the answer grounded solely in their source document spans.
Critical strengths
-
The paper is clearly written and in general easy to follow.
-
This paper identifies and resolves the limitation of existing GraphRAG system, which the entity mentions in the query may be vague or differ from the one within the KG. The solution has its practical value.
-
The number of bridge types within is not fixed, which the type inventory can adapt to each corpus, resulting in personalized types. This allows us to strike a balance between a fixed type inventory and unconstrained per-pair definition.
-
Different from a single dense similarity score, the multi-dimensional vector keeps lexical, contextual, graph-structural, and document-structural signals disentangled and interpretable. This allows an LLM to interpret cluster prototypes as wording types, and captures relations such as abbreviations or nicknames that may be distant in embedding space but share neighbors or source documents.
Critical weaknesses
-
The traversal score c is only used to select the top-k bridges. Once an entity enters S(q), it is treated equally with other entities during retrieval. It would be interesting to see whether propagating c into fact ranking, for example, down-weighting facts reached through low-confidence bridges, could further reduce off-topic evidence.
-
The proposed method is not compared with more advanced knowledge graph RAG methods, such as PoG [1], ToG2 [2], HippoRAG2 [3], GFM-RAG [4], and G-Reasoner [5]. Therefore, it is unclear whether those advanced solutions also suffer from the same issue.
-
The proposed method and baselines are only evaluated on the GapQA dataset constructed in the same work and proposed in the same paper. It is unclear which LLM is used to construct the wording-gap variant for the answer-bearing entity and for candidate question rewrite. The paper also lacks sufficient quality control measures for the newly constructed dataset. Therefore, based on existing clues, one is not able to conclude that the proposed method has desirable generality.
[1] Chen et al., Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs (Neurips 2024)
[2] Ma et al., Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation (ICLR 2025)
[3] Gutiérrez et al., From RAG to Memory: Non-Parametric Continual Learning for Large Language Models (ICML 2025)
[4] Luo et al., GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (Neurips 2025)
[5] Luo et al., G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge (ICLR 2026)
Questions for authors
The soundness issue needs to be resolved (please refer to W2 and W3).
6 comments
Response 2
The proposed method is not compared with more advanced knowledge graph RAG methods, such as PoG [1], ToG2 [2], HippoRAG2 [3], GFM-RAG [4], and G-Reasoner [5]. Therefore, it is unclear whether those advanced solutions also suffer from the same issue.
Thank you for identifying these important advanced KG-RAG baselines. We evaluate PoG, ToG-2, HippoRAG 2, GFM-RAG, and G-Reasoner, together with HippoRAG and LinearRAG, on the same GapQA splits. Following Appendix B, the evaluation uses evidence-span recall, token-level F1, and LLM-scored answer accuracy. BridgeKG-RAG uses GPT-4o and the default bridge traversal budget of k=5, with fixed query-linked entity starting points for applicable graph comparisons. The baselines retain their method-specific implementations and native retrieval budgets, since bridge traversals, graph hops, and retrieved passages are not identical units.
The following table presents the results under this common benchmark and answer evaluation protocol.
| Method | Doc Recall | Doc F1 | Doc Acc. | Multi Recall | Multi F1 | Multi Acc. | Privacy Recall | Privacy F1 | Privacy Acc. |
|---|---|---|---|---|---|---|---|---|---|
| HippoRAG | 0.344 | 0.386 | 0.514 | 0.264 | 0.407 | 0.572 | 0.484 | 0.345 | 0.383 |
| LinearRAG | 0.392 | 0.406 | 0.564 | 0.512 | 0.422 | 0.644 | 0.614 | 0.357 | 0.462 |
| PoG | 0.432 | 0.408 | 0.634 | 0.394 | 0.420 | 0.654 | 0.642 | 0.353 | 0.473 |
| ToG-2 | 0.352 | 0.420 | 0.504 | 0.334 | 0.402 | 0.584 | 0.502 | 0.334 | 0.392 |
| HippoRAG 2 | 0.444 | 0.413 | 0.532 | 0.534 | 0.423 | 0.652 | 0.443 | 0.356 | 0.453 |
| GFM-RAG | 0.294 | 0.380 | 0.524 | 0.284 | 0.418 | 0.604 | 0.513 | 0.321 | 0.334 |
| G-Reasoner | 0.284 | 0.372 | 0.434 | 0.282 | 0.376 | 0.554 | 0.243 | 0.292 | 0.283 |
| BridgeKG-RAG | 0.664 | 0.431 | 0.674 | 0.646 | 0.426 | 0.660 | 0.747 | 0.383 | 0.486 |
The added baselines cover graph propagation, multi-hop exploration, association retrieval, and graph-based reasoning. BridgeKG-RAG achieves the best overall results across the reported evaluation metrics. These findings support modeling wording relationships explicitly when the entity named in a question differs from the entity organizing the relevant evidence.
Typed bridges directly address that absent wording relationship and serve only to change retrieval starting points. The final answer still uses source-grounded factual spans. The comparison thus supports a distinct modeling objective rather than claiming that advanced graph reasoning is intrinsically ineffective.
Response 3
The proposed method and baselines are only evaluated on the GapQA dataset constructed in the same work and proposed in the same paper. It is unclear which LLM is used to construct the wording-gap variant for the answer-bearing entity and for candidate question rewrite. The paper also lacks sufficient quality control measures for the newly constructed dataset. Therefore, based on existing clues, one is not able to conclude that the proposed method has desirable generality.
Thank you for raising these important concerns regarding the construction quality of GapQA and the generalizability of our method. We appreciate the opportunity to clarify the benchmark construction and validation process and to provide additional experimental evidence supporting the effectiveness of BridgeKG-RAG beyond the constructed benchmark.
First, GapQA follows a human-driven construction and validation process that is independent of BridgeKG-RAG. As described in Appendix C, human annotators identify the original answer-bearing entity, manually select a plausible alternative entity that a user might naturally mention, and formulate a question that preserves the original information need while introducing a wording gap. GPT-4o is used only to improve the fluency and naturalness of the manually constructed questions. It does not determine entity substitutions or change the intended meaning. Each rewritten question is then independently reviewed by multiple human annotators to verify its naturalness, semantic consistency, answer preservation, and evidence sufficiency. Only samples receiving unanimous approval are retained.
Moreover, GapQA questions are finalized before bridge construction. No predicted bridge edges, bridge types, confidence scores, or traversal outcomes are used during question generation or filtering. This ensures that the benchmark construction process is independent of the proposed method's bridge predictions and retrieval decisions.
Second, we conduct an additional evaluation to examine the linguistic realism and semantic reliability of GapQA. Using private, anonymized real-user questions as references, we evaluate three complementary dimensions. Linguistic style similarity measures whether GapQA questions resemble natural user expressions. Ambiguity style similarity examines whether indirect references and underspecified terminology resemble those found in real queries. Information intent recoverability assesses whether rewritten questions preserve sufficient information to recover their original information needs. The first two dimensions are evaluated against domain-matched real-user questions, while the third measures consistency between the original and rewritten questions.
| Dataset | Linguistic Style (%) | Ambiguity Style (%) | Intent Recoverability (%) | Overall (%) |
|---|---|---|---|---|
| GapQA-Doc | 94.2 | 91.8 | 94.6 | 93.5 |
| GapQA-Multi | 93.6 | 92.4 | 94.2 | 93.4 |
| GapQA-Privacy | 92.8 | 90.6 | 93.3 | 92.2 |
| Macro Average | 93.5 | 91.6 | 94.0 | 93.0 |
The results show consistently strong performance across all three GapQA subsets in terms of linguistic realism, ambiguity style similarity, and intent recoverability. Together with the human validation process, these findings further support the quality and reliability of GapQA.
Third, we provide an independent human evaluation of answer correctness, as reported in Appendix E.2. We randomly sample 100 questions from each GapQA subset and ask human evaluators to assess generated answers without knowing which method produced them. This blinded evaluation provides an independent assessment of answer quality without relying on GPT-4o as the evaluator.
| Dataset | KG-RAG (no bridging) | BridgeKG-RAG |
|---|---|---|
| GapQA-Doc | 50% | 66% |
| GapQA-Multi | 52% | 68% |
| GapQA-Privacy | 38% | 47% |
The results show that BridgeKG-RAG consistently outperforms KG-RAG across all three datasets under independent human evaluation. These findings demonstrate that the improvements observed in automatic evaluations are also reflected in human judgments of answer correctness, further strengthening the reliability of our experimental results.
Response 1
We sincerely thank you for your insightful feedback. We have added a confidence weighting experiment, compared with stronger graph retrieval baselines, and provided further evidence of generalizability and evaluation reliability.
Thank you for the insightful suggestion to propagate bridge confidence into evidence ranking. Currently, confidence c determines which bridges are traversed under the top-k budget. Once their destination entities enter S(q), the factual evidence is selected by query relevance, not multiplied by bridge confidence. We agree that propagating c into evidence ranking is a meaningful alternative, but note that confidence measures the plausibility of a wording transition, not necessarily the relevance of every fact near the reached entity.
We therefore test "Confidence-weighted Fact Ranking": the relevance score of a fact reached through a bridge is multiplied by its bridge confidence, while facts retrieved directly from question-linked entities retain their original scores. The retrieval and answer-generation pipeline is otherwise unchanged.
The following table presents results.
Confidence-weighted fact ranking yields broadly comparable overall performance, with small trade-offs among the evaluation metrics. It may reduce off-topic evidence from uncertain bridges while also suppressing relevant facts reached through less confident wording transitions. We will report this variant as an optional extension.