Organizations: Shanghai Jiao Tong University · Singapore University of Technology and Design · Monash University · Tencent Youtu Lab · Nanyang Technological University · Griffith University
Large language models (LLMs) have shown strong capabilities, with retrieval-augmented generation (RAG) supporting complex multi-hop reasoning by retrieving evidence distributed across documents. Graph-based approaches exploit connections among evidence, and hypergraph-based retrieval further preserves higher-order entity associations within documents and connects documents through shared entities. However, existing hypergraph retrievers often rely on predefined structural expansion or diffusion, which may miss query-dependent interactions needed to identify relevant evidence. They also leave connections among retrieved evidence implicit, requiring LLMs to reconstruct these connections before reasoning. Therefore, we propose HyperReCo, a framework for retrieving and connecting evidence with a hypergraph neural network (HyperGNN). We represent each document as a hyperedge over its extracted entities, with shared entities connecting the hyperedges. Through hypergraph message passing with joint supervision over documents and entities, the HyperGNN learns query-dependent interactions to retrieve complementary evidence. We further introduce Gradient-Guided Hyper-Path Decoding (GGHD), which uses gradient attribution to interpret the learned interactions and translate them into explicit hyper-paths that help LLMs combine complementary facts for multi-hop reasoning. Experiments on six benchmarks show that HyperReCo achieves the best retrieval performance among the compared methods on all three multi-hop QA datasets, together with strong downstream QA performance. Case studies and further analyses demonstrate the utility of decoded hyper-paths for connecting retrieved evidence.
Figures & tables
Figure 1: Illustration of the evidence routing gap. HyperReCo augments the retrieved evidence with a decoded hyper-path linking retrieved documents.
Figure 2: Overview of HyperReCo . (a) The query-conditioned HyperGNN learns structural relevance over hypergraph for evidence retrieval. (b) Gradient-Guided Hyper-Path Decoding (GGHD) uses gradient attribution and Q-PCSF to translate the learned structural relevance into explicit grouped hyper-paths. (c) The retrieved documents and decoded hyper-paths are provided to the LLM for answer generation.
Method
HotpotQA
MuSiQue
2Wiki
G-bench (Novel)
G-bench (Medical)
G-bench (CS)
EM
F1
EM
F1
EM
F1
ACC
ACC
ACC
Non-structure Methods
None (GPT-4o-mini) ( OpenAI, 2024 )
28.6
41.0
11.2
36.3
30.2
36.3
51.4
67.1
70.7
BM25 ( Robertson and Walker, 1994 )
52.0
63.4
20.3
28.8
47.9
51.2
56.5
68.7
71.7
ColBERTv2 ( Santhanam et al., 2022 )
43.4
57.7
15.5
26.4
33.4
43.3
56.2
71.8
71.9
Qwen3-Emb (8B) ( Zhang et al., 2025 )
53.4
67.6
31.9
44.1
57.2
63.2
56.2
70.4
73.5
Table 1: QA reasoning performance comparison. GPT-4o-mini is used as the LLM for reasoning. The best result is in bold.
Method
HotpotQA
MuSiQue
2Wiki
G-bench (Novel)
G-bench (Medical)
R@2
R@5
R@2
R@5
R@2
R@5
Recall
Recall
Non-structure Methods
BM25 ( Robertson and Walker, 1994 )
55.4
72.2
32.3
41.2
51.8
61.9
82.1
87.9
ColBERTv2 ( Santhanam et al., 2022 )
64.7
79.3
37.9
49.2
59.2
68.2
82.4
89.5
Qwen3-Emb (8B) ( Zhang et al., 2025 )
74.1
88.8
46.8
62.1
66.2
74.1
82.6
92.7
NV-Embed-v2 (7B) ( Lee et al., 2025 )
84.1
94.4
52.7
69.5
69.1
76.5
92.5
96.5
Table 2: Retrieval performance comparison. Recall@ k (R@ k ) is used for multi-hop QA datasets, and evidence recall (Recall) is used for G-bench ( Xiang et al., 2026 ) . The best result is in bold.
Question
When did Lord George Scott’s father die?
Gold Answer
5 November 1914
Supporting Docs.
[“Lord George Scott”, “William Montagu Douglas Scott, 6th Duke of Buccleuch”]
Retrieved Docs.
[D1: “Lord George Scott” , D2: “George Lyon, 5th Lord Glamis”, D3: “William Montagu Douglas Scott, 6th Duke of Buccleuch” , D4: “Charlotte Montagu Douglas Scott, Duchess of Buccleuch”, D5: “Walter Montagu Douglas Scott, 8th Duke of Buccleuch”]
Evidence Input
D1, D2, D3, D4, D5 w/o Hyper-paths
Answer
The passage does not provide the date of Lord George Scott’s father’s death. ✗ EM =0 , F1 =0
Evidence Input
D1, D2, D3, D4, D5 HyperReCo
Table 3: Qualitative example of hyper-path-guided reasoning
Figure 6
Method
Tokens / 10K ↓
Time (s) ↓
EM ↑
GraphRAG
76M
5.02
37.7
LightRAG
55M
8.39
4.8
GFM-RAG
48M
3.36
52.1
HGRAG
3.3M
3.08
53.4
HyperReCo
3.3M
3.55
57.6
Table 4: Efficiency comparison. Token cost is measured per 10K documents. Time and EM are macro-averaged over HotpotQA, MuSiQue, and 2Wiki.
Appendix figures & tables9 assets
Supplementary material from the paper’s appendix.
Appendix
Dataset
#Questions
#Documents
#Entities
HotpotQA
2,000
19,527
130,484
MuSiQue
2,000
22,417
126,603
2Wiki
2,000
11,629
69,199
Total
6,000
53,573
326,286
Appendix
Table 5: Statistics of the training datasets.
Dataset
# Query
# Document
HotpotQA ( Yang et al., 2018 )
1,000
9,221
MuSiQue ( Trivedi et al., 2022 )
1,000
11,656
2Wiki ( Ho et al., 2020 )
1,000
6,119
G-bench (Novel) ( Xiang et al., 2026 )
2,010
461
G-bench (Medical) ( Xiang et al., 2026 )
2,062
2,406
G-bench (CS) ( Xiao et al., 2025 )
1,018
24,534
Appendix
Table 6: Statistics of the evaluation datasets.
Figure 5: The prompt template used in HyperReCo for hyper-path-guided LLM reasoning.
Embedding Model
Method
HotpotQA
MuSiQue
2Wiki
R@2
R@5
R@2
R@5
R@2
R@5
NV-Embed-v2 (7B) ( Lee et al., 2025 )
Embedding
84.1
94.4
52.7
69.5
69.1
76.5
HGRAG
77.8
94.0
53.7
73.1
74.9
91.4
HyperReCo
87.1
96.8
57.2
74.5
81.2
97.8
Qwen3-Emb (8B) ( Zhang et al., 2025 )
Embedding
74.1
88.8
46.8
62.1
66.2
74.1
HGRAG
72.2
88.0
47.2
61.1
66.3
74.1
Appendix
Table 7: Retrieval performance with different embedding models. We report document Recall@2 (R@2) and Recall@5 (R@5). “Embedding” denotes direct dense retrieval using the corresponding embedding model.
HotpotQA
MuSiQue
2Wiki
Avg.
Method
Time (s) ↓
EM ↑
Time (s) ↓
EM ↑
Time (s) ↓
EM ↑
Time (s) ↓
EM ↑
GraphRAG
4.10
51.4
6.18
27.0
4.79
34.7
5.02
37.7
LightRAG
8.71
9.9
8.56
2.0
7.89
2.5
8.39
4.8
GFM-RAG
3.83
56.2
3.27
30.2
2.99
69.8
3.36
52.1
HGRAG
2.92
55.9
3.33
36.2
2.99
68.2
3.08
53.4
HyperReCo
3.34
60.7
4.01
39.9
3.30
72.2
3.55
57.6
Appendix
Table 8: Detailed inference efficiency across three multi-hop QA datasets. Avg. denotes the macro-average across HotpotQA, MuSiQue, and 2Wiki.
Method
HotpotQA
MuSiQue
2Wiki
Mean
HyperReCo
60.7
39.9
72.2
57.6
w/o paths
57.7
35.6
71.6
55.0
w/o Q-PCSF
59.8
38.5
70.8
56.4
G-paths
59.9
38.8
72.0
56.9
S-forest
60.2
39.4
71.5
57.0
Appendix
Table 9: Detailed EM (%) results for different hyper-path decoding strategies. Mean denotes the macro average over HotpotQA, MuSiQue, and 2Wiki.
Figure 6: The prompt template used for LLM-as-a-judge evaluation of hyper-path helpfulness.
Method
Helpful-path ratio (%) ↑
Mean paths per question ↓
GGHD
59.2
1.68
w/o Q-PCSF
28.4
5.39
G-paths
43.1
2.97
S-forest
50.6
2.11
Appendix
Table 10: LLM-judged hyper-path utility and compactness. Helpful-path ratios are averaged over questions. GGHD achieves the highest helpful-path ratio with the fewest paths per question.
Question
When did Lord George Scott’s father die?
Gold Answer
5 November 1914
Supporting Docs.
[“Lord George Scott”, “William Montagu Douglas Scott, 6th Duke of Buccleuch”]
Retrieved Docs.
[D1: “Lord George Scott” , D2: “George Lyon, 5th Lord Glamis”, D3: “William Montagu Douglas Scott, 6th Duke of Buccleuch” , D4: “Charlotte Montagu Douglas Scott, Duchess of Buccleuch”, D5: “Walter Montagu Douglas Scott, 8th Duke of Buccleuch”] (Recall@5 = 1.0)
w/o Hyper-paths
LLM Input
Question: When did Lord George Scott’s father die? Retrieved Docs: D1, D2, D3, D4, D5
Reasoning
The passage provides information about Lord George Scott and his family, specifically mentioning that he was the son of William Montagu Douglas Scott, 6th Duke of Buccleuch. However, it does not provide a specific date of death for Lord George Scott’s father. The only date mentioned in the passage is the death date of Lord George Scott himself, which is 23 February 1947. Since the passage does not include any information about the death of William Montagu Douglas Scott, 6th Duke of Buccleuch, I cannot determine when he died based on the provided text.
Appendix
Table 11: Qualitative case study of cross-document evidence composition with and without hyper-path guidance. Both settings retrieve all supporting documents (Recall@5 = 1.0), yet only the hyper-path-guided setting produces the correct answer.