MAPS: Macro-Adaptive Parallel Reasoning for Regulating Search Momentum in Retrieval-Augmented Generation
Authors: Xinze Li, Dingwen Zhang, Zhenghao Liu, Yukun Yan, Shuo Wang, Yu Gu, Ge Yu, Maosong Sun
Organizations: School of Computer Science and Engineering, Northeastern University, Shenyang, China · Department of Computer Science and Technology, Tsinghua University, Beijing, China
Retrieval-Augmented Generation (RAG) improves the performance of Large Language Models (LLMs) by retrieving external knowledge. Recently, some RAG methods interleave reasoning and retrieval, enabling LLMs to progressively generate sub-queries and retrieve external knowledge. However, these methods may suffer from redundant or ineffective sub-queries and insufficiently diverse query exploration. Moreover, due to the tightly interleaved nature of reasoning and retrieval, it remains unclear how the evolving reasoning process influences query intent. To characterize the evolution of query intent during reasoning, we introduce search momentum, which captures the accumulated retrieval tendency induced by historical changes in generated sub-queries along the reasoning trajectory. Through momentum-based analysis, we find that the query generation process tends to continuously evolve along existing retrieval trajectories. Based on this observation, we propose MAPS, a Macro-Adaptive Parallel Reasoning framework for regulating search momentum and mitigating query-intent drift. MAPS first retains the retrieved knowledge and a refined reasoning state, thereby reducing the influence of historical reasoning trajectories on subsequent sub-query generation. Subsequently, MAPS extends the conventional sequential reasoning process into a parallel reasoning paradigm. At each iteration, MAPS identifies complementary retrieval directions, generates corresponding sub-queries for parallel retrieval, and consolidates the retrieved evidence to update the knowledge and reasoning states. Experiments on six question-answering benchmarks demonstrate that MAPS consistently outperforms representative RAG methods across different backbone models. All code and resources are available at https://github.com/NEUIR/MAPS.
Figures & tables
Dataset
β1
S1
β2
S2
β3
S3
β4
S4
Avg. St
2WikiQA
0.0000
0.4051
0.6352
0.5698
0.7552
0.6483
0.8761
0.6955
0.5797
MuSiQue
0.0000
0.3105
0.5400
0.5196
0.7214
0.6489
0.8252
0.6913
0.5426
HotpotQA
0.0000
0.4427
0.6161
0.6840
0.8151
0.7731
0.8836
0.7750
0.6687
Pooled-MHQA
0.0000
0.3875
0.5959
0.5651
0.7559
0.6774
0.8630
0.7279
0.5895
Table 1: Fitted Momentum Coefficients and Directional Alignment Between Search Momentum and Subsequent Query Intent Deviations. St denotes the mean directional alignment score St across trajectories in the independent test set. βt controls the retention of previously accumulated retrieval tendencies when constructing search momentum. The Pooled-MHQA fits the momentum coefficients on the mixed training data from the three datasets and evaluates directional alignment on the corresponding test sets. Detailed experimental settings and results are provided in Appendix A.1 .
Method
NQ
PopQA
2WikiQA
HotpotQA
MuSiQue
GPQA
Avg.
Qwen3-32B
Vanilla LLM
25.0
19.4
27.3
21.3
7.1
44.4
24.1
Vanilla RAG
49.0
52.1
30.0
38.3
9.9
44.2
37.2
Self-RAG
39.2
37.9
29.8
31.5
17.2
45.1
33.4
IRCOT
38.2
37.6
38.8
33.2
22.3
54.2
37.4
Search-o1
53.1
56.5
43.7
40.0
23.7
46.4
43.9
Table 2: Overall Performance of MAPS across All Benchmarks. The best results are in bold and the second-best are underlined .
Method
NQ
PopQA
2WikiQA
HotpotQA
MuSiQue
GPQA
Avg.
Qwen3-32B
MAPS
63.4
64.8
59.4
48.8
37.6
62.3
56.1
w/o Retrieval Direction
63.0
64.3
55.8
46.9
35.6
46.0
51.9
w/o Parallel Reasoning
60.1
61.2
54.1
45.2
33.1
45.3
49.8
w/o Disentanglement
61.0
58.8
53.1
48.3
34.3
45.1
50.1
Qwen3.6-35B-A3B
Table 3: Ablation Study. The best results are in bold and the second-best are underlined .
Figure 1: Evolution of Search Momentum during Iterative Reasoning. Figures 1(a) and 1(b) visualize the trajectories of search momentum through t-SNE. Figure 1(c) shows the cumulative directional offset of search momentum with respect to the original query across reasoning iterations.
Figure 2: Evolution of Query Similarity and Query-Evidence Similarity during Iterative Reasoning.
Figure 3: Termination Distribution and Answer Performance across Different Methods.
Appendix figures & tables7 assets
Supplementary material from the paper’s appendix.
Appendix
Dataset
β1
S1
β2
S2
β3
S3
β4
S4
Avg.
2WikiQA
0.0000
0.5264
0.5408
0.8206
0.7472
0.8689
0.8698
0.8929
0.7772
MuSiQue
0.0000
0.5120
0.5272
0.8297
0.7404
0.8844
0.8394
0.9050
0.7828
HotpotQA
0.0000
0.5851
0.5256
0.8438
0.7259
0.9037
0.8557
0.9131
0.8115
Pooled MHQA
0.0000
0.5331
0.5328
0.8276
0.7408
0.8803
0.8570
0.9006
0.7854
Appendix
Table 4: Fitted Momentum Coefficients and Directional Alignment between Search Momentum and Subsequent Query Intent Deviations for MAPS.
Dataset
β1
S1
β2
S2
β3
S3
β4
S4
Avg.
2WikiQA
0.0000
0.5036
0.5469
0.8102
0.7543
0.8548
0.8527
0.8533
0.7555
MuSiQue
0.0000
0.4855
0.5369
0.7824
0.7538
0.8329
0.8629
0.8545
0.7388
HotpotQA
0.0000
0.5435
0.5467
0.8086
0.7607
0.8236
0.8651
0.8596
0.7588
Pooled MHQA
0.0000
0.5065
0.5422
0.7978
0.7549
0.8396
0.8589
0.8547
0.7497
Appendix
Table 5: Fitted Momentum Coefficients and Held-out Alignment Scores for MAPS w/o Disentanglement on Three Multi-hop Datasets and Pooled MHQA.
Figure 4: T-SNE Projection of E5 Embeddings for the Case-study Trajectories in Table 6 . The black diamond marks the original question, and gold stars mark the four gold documents. Orange points trace Search-o1’s sequential queries; blue points mark MAPS’s parallel queries, with letters identifying branches. Blue outlines indicate the range covered by MAPS’s queries in each iteration.
Table 6: Case Study Comparing Search-o1 and MAPS. Red highlights misleading observations, off-topic evidence associated with query drift, and the incorrect final output; blue marks correctly identified key entities and the correct final answer.
Figure 5: Prompt Template Used in MAPS for Generating Retrieval Directions.
Figure 6: Prompt Template Used in MAPS for Parallel Sub-Query Generation.
Figure 7: Prompt Template Used in MAPS for Refined Reasoning.
Large reasoning models such as DeepSeek-R1 and OpenAI o1 generate extended chains of thought spanning thousands of tokens, yet their integration with retrieval-augmented generation (RAG) remains fundamentally misaligned. Current RAG systems optimize for providing context before reasoning begins, while reasoning models require evidence injection during multi-step inference chains. We introduce ReaLM-Retrieve, a reasoning-aware retrieval framework that addresses this mismatch through three key innovations: (1) a step-level uncertainty detector that identifies knowledge gaps at reasoning-step granularity rather than token or sentence level; (2) a retrieval intervention policy that learns when external evidence maximally benefits ongoing reasoning; and (3) an efficiency-optimized integration mechanism that reduces per-retrieval overhead by 3.2x compared to naive integration. Experiments on MuSiQue, HotpotQA, and 2WikiMultiHopQA demonstrate that ReaLM-Retrieve achieves on average 10.1% absolute improvement in answer F1 over standard RAG (range: 9.0-11.8% across the three benchmarks) while reducing retrieval calls by 47% compared to fixed-interval approaches like IRCoT (all improvements significant at p<0.01, paired bootstrap). On the challenging MuSiQue benchmark requiring 2-4 hop reasoning, our method achieves 71.2% F1 with an average of only 1.8 retrieval calls per question. Analysis shows that ReaLM-Retrieve also improves retrieval quality itself, achieving 81.3% Recall@5 with consistently higher precision and MRR than fixed-interval baselines on supporting evidence, establishing new state-of-the-art efficiency-accuracy trade-offs for reasoning-intensive retrieval tasks.
Dongxin Guo, Jikun Wu, Siu Ming Yiu
The University of Hong Kong Hong Kong, China · Brain Investing Limited Hong Kong, China · Stellaris AI Limited Hong Kong, China
Retrieval-Augmented Generation (RAG) has become a standard approach for enhancing large language models (LLMs) with external knowledge, mitigating hallucinations, and improving factuality. However, existing systems rely on generating natural language queries at each hop and maintaining a strict architectural separation between retriever and generator, preventing them from leveraging the full representational capacity of the LLM. We propose LAnR (Latent Abstraction for RAG), a unified framework in which a single LLM jointly performs encoding, retrieval, and generation entirely within its own latent space. Rather than generating textual queries, LAnR produces dense retrieval vectors from the hidden states of a designated [PRED] token and uses them to match against encoded document representations from the same model. Furthermore, LAnR adaptively decides when sufficient evidence has been retrieved using a lightweight MLP control head over those same hidden states, eliminating both the separate retriever and explicit token-level stopping reasoning. This design is motivated by our empirical observation that answer token entropy reliably signals retrieval sufficiency. Extensive experiments on six QA benchmarks spanning single-hop and multi-hop settings demonstrate that LAnR outperforms existing RAG methods, while achieving improved inference efficiency through reduced number of retrieval calls and tighter model integration.
Retrieval-Augmented Generation (RAG) expands the knowledge of Large Language Models (LLMs), yet current static retrieval methods struggle with complex, multi-hop problems. While recent dynamic retrieval strategies offer improvements, they face two key challenges: 1) irrelevant retrieved noise can misdirect the reasoning process, and 2) processing full documents incurs prohibitive computational and latency costs. To address these issues, we propose OThink-SRR1, a framework that enhances large models with an iterative Search-Refine-Reason process trained via reinforcement learning. Its core Refine stage distills retrieved documents into concise, relevant facts before reasoning. We introduce GRPO-IR, an end-to-end reinforcement learning algorithm that rewards accurate evidence identification while penalizing excessive retrievals, thus training the model to be both focused and efficient. Experiments on four multi-hop QA benchmarks show our approach achieves superior accuracy over strong baselines while using fewer retrieval steps and tokens. This positions OThink-SRR1 as a potent foundational model for information-seeking agents.