Pith. sign in

REVIEW 5 major objections 6 minor 61 references

Enhancing Large Language Models with Reward-guided Tree Search for Knowledge Graph Question and Answering

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A training-free tree-search framework is claimed to make LLMs retrieve better reasoning paths from knowledge graphs than prior search strategies.

desk verdict A coherent training-free KGQA method with plausible gains, but the SOTA claim rests on unreleased subsets, test-set-tuned hyperparameters, and single-run scores. read the letter →

arxiv 2505.12476 v1 pith:OJQA5UUO submitted 2025-05-18 cs.CL cs.AI

Xiao Long , Liansheng Zhuang , Chen Shen , Shaotian Yan , Yifei Li , Shafei Wang This is my paper · ORCID
classification cs.CLcs.AI
keywords knowledgegraphquestionansweringlargelanguagemodelsretrieval-augmentedgenerationMonteCarlotreesearchself-criticmechanismdecompositionreasoningpathretrievalFreebase
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that a training-free, reward-guided tree search over a knowledge graph retrieves better reasoning paths than the beam search, greedy search, or best-of-N planning used by earlier LLM-based KGQA methods. It argues that previous approaches explore new paths without exploiting paths already found, and that complex compositional questions need to be broken into sub-questions before retrieval. Its proposed framework, RTSoG, decomposes each question, runs a self-critic Monte Carlo tree search that scores paths with a reward model, and feeds the highest-weight paths into the LLM in a stack. According to the reported experiments, RTSoG achieves the best exact-match accuracy on WebQSP, ComplexWebQuestions, GrailQA, and WebQuestions with every LLM tested, with gains of roughly 7 to 16 points over the strongest prior method. If those numbers hold, the practical payoff is a way to ground LLM answers in a knowledge graph without any fine-tuning.

What carries the argument

The load-bearing mechanism is the Self-Critic Monte Carlo Tree Search (SC-MCTS). Starting from the question's linked topic entity, it repeatedly selects a node by UCT, expands by asking the LLM to filter the entity's adjacent relations and then pick the highest-reward tail entity for each surviving relation, and evaluates the new node with an LLM value model that combines a local relation reward and a global path reward as $Q = \alpha S^r_{filt} + (1-\alpha) S^P_{cand}$. The novel step is the self-critic: after each expansion the same LLM is asked whether the accumulated path already answers the question, and an affirmative answer emits an End-of-Search signal that makes the node a leaf. After $H$ iterations the top-$K$ nodes by $Q$ supply weighted reasoning paths, which are sorted descending and passed through a reasoning-path stack; earlier high-weight paths act as historical evidence when judging later ones, and the surviving paths condition the final answer.

What would settle it

Run RTSoG and the strongest prior baseline several times on the exact same publicly released test subsets with matched hyperparameters and fixed seeds; if the exact-match margins on WebQSP and GrailQA shrink below about two points or overlap across runs, the claimed state-of-the-art improvement does not hold.

Watch

Extended reading notes

Core claim

The paper's central discovery, on its own terms, is that a self-critic Monte Carlo tree search guided by an LLM reward model is a better graph-retrieval engine than the search strategies in prior LLM-based GraphRAG systems. In the reported experiments RTSoG reaches exact-match 94.3 on WebQSP, 82.8 on ComplexWebQuestions, 93.4 on GrailQA, and 83.5 on the WebQuestions sample, outperforming the strongest prior method by 7.0, 7.8, 8.7, and 16.4 points respectively, and outperforming it with all four tested policy models. The attribution is specific: question decomposition helps the search focus; the value-model reward combines a local relation filter score with a global path score; the self-critic End-of-Search signal stops expansion at the correct node; and the weighted path stack removes noisy paths before final generation. Ablations reported in the paper attribute 7.4 and 9.8 point losses on WebQSP and CWQ to removing the self-critic, and 5.2 and 7.2 point losses to removing the path stack.

Load-bearing premise

The reported margins rest on the assumption that the 1,000- and 1,500-question test subsets are the same slices used by the cited baselines and that one roll of the nondeterministic model API is a fair comparison.

Editorial extensions

If this is right

  • A training-free LLM pipeline can outperform fine-tuned semantic parsers on GrailQA and WebQuestions, not just other prompting methods, when retrieval is guided by a reward-modeled tree search.
  • Question decomposition before graph search is a large contributor: removing it drops performance substantially on both WebQSP and GrailQA in the paper's sensitivity analysis.
  • The self-critic stop signal is a core component: replacing SC-MCTS with plain MCTS loses 7.4 EM points on WebQSP and 9.8 on CWQ in the reported ablations.
  • Weighted use of retrieved paths matters: the reasoning-path stack adds 5.2 points on WebQSP and 7.2 on GrailQA over feeding the paths in without the stack.
  • Retrieval quality can compensate for model scale: RTSoG with a smaller open LLM matches or exceeds larger closed LLMs used with earlier retrieval strategies.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Not tested here but directly suggested: the same decomposition-plus-SC-MCTS loop should transfer to other knowledge graphs and to structured sources beyond graphs, since no component in the search is specific to Freebase.
  • A useful follow-up is to separate how much of the gain comes from the search structure versus the LLM's self-evaluation, by pairing a weak policy model with a strong value model and vice versa; the paper uses one LLM for both roles.
  • The self-critic signal amounts to an early-stopping rule learned by prompting, so an interesting extension is to make the stop decision trainable or calibrated, especially for low-resource domains where the policy model's self-assessment is unreliable.
  • Because the headline margins depend on a single run over test subsets, a variance-controlled replication across repeated API calls would reveal whether the 7-16 point leads are stable enough for deployment decisions.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes Reward-guided Tree Search on Graph (RTSoG), a training-free framework for knowledge graph question answering (KGQA). RTSoG decomposes the input question into sub-questions, performs a Self-Critic Monte Carlo Tree Search (SC-MCTS) over the knowledge graph using the same LLM as policy model and value model, selects the top-K weighted reasoning paths, and pushes them through a reasoning-path stack before generating the final answer. The authors report exact-match accuracy on WebQSP, CWQ, GrailQA, and WebQuestions, claiming state-of-the-art results with RTSoG and, for example, 8.7% and 7.0% improvements on GrailQA and WebQSP, respectively.

Significance. If the empirical claims hold, RTSoG would be a strong training-free GraphRAG baseline, showing that MCTS-style exploration/exploitation combined with path weighting and a self-critic mechanism can improve LLM KGQA performance across multiple datasets and policy models. The algorithmic description is reasonably clear, the ablations for the self-critic mechanism and the reasoning-path stack are informative, and the efficiency analysis is a useful addition. However, the central claim is entirely empirical, and the evaluation as presented does not currently support the claimed state-of-the-art margins: no code, data, sample identifiers, seeds, or multiple-run statistics are provided, and hyperparameters are selected on the same test sets used for the final numbers. The paper's strengths are its clear method formulation and the internal ablations, but the missing evaluation infrastructure is a major obstacle to verification.

major comments (5)
  1. [§V-A, Table II and Tables III–IV] The GrailQA and WebQuestions results are reported on 1,000- and 1,500-sample subsets that the paper says were 'randomly selected ... as previous works,' but no sample identifiers or release are provided, so there is no way to verify that the cited ToG and PoG numbers were computed on the same questions. Because every headline gain is an absolute EM difference on exactly these subsets, the claimed state-of-the-art margins are unsupported unless the exact sample sets are released and the baselines are rerun on them, or sample IDs are provided for independent verification.
  2. [§V-C, Fig. 2, Tables III–IV] The sensitivity analysis selects n, H, b, and K by sweeping one parameter at a time on the WebQSP and GrailQA test sets, and alpha is fixed after those same experiments; the final Tables III and IV are then produced on those same test sets. This is test-set tuning, and the prompting baselines ToG and PoG were not given an equivalent selection step. Please introduce a validation split for hyperparameter selection, or report the results across the full sweep to show that the SOTA margins are not an artifact of choosing the best test-set point.
  3. [§V-A Implementation Details, Tables III–IV] The evaluation is a single run per configuration under nondeterministic LLM APIs: Qwen2.5-14b is run at temperature 0.7, the OpenAI models are called through APIs with no seeds, and no confidence intervals or multiple-run statistics are reported. EM differences of 5–8 points can easily fall within run-to-run variance of LLM prompting and search. Please report means and standard deviations over at least three runs and a significance test for the main comparisons.
  4. [§V-B vs Table III] The text states that RTSoG with GPT-4 shows 'an increases of 7.0% and 7.8% on WebQSP and CWQ' over PoG with GPT-4, but Table III reports WebQSP 94.3 vs 87.3, a difference of 5.5 points, while the '+7.0' figure appears only in the abstract and conclusion. The comparison basis (PoG vs EPREM vs absolute vs relative) needs to be stated consistently and the table and text reconciled.
  5. [§IV-B, §V-D] The same LLM serves as question decomposer, policy model, value model, self-critic, and final answer generator, so the reward signals and EoS signals come from the same system being evaluated. This does not invalidate the method, but it makes the contribution of graph-grounded search to the reported gains unclear; the ablation in Fig 3(b) removes the whole SC-MCTS stage, not the reward model, and the path-stack ablation in Table V keeps the same LLM in both arms. An additional control that uses the same policy LLM with an alternative context-selection method (e.g., PoG or random paths) would more cleanly isolate the benefit of reward-guided search.
minor comments (6)
  1. [§II heading] The heading 'RELATED WROK' contains a typo and should read 'RELATED WORK'.
  2. [§IV-B, Eq. (11)] Equation (11) is described as a weighted average, but the displayed formula is the average of children's values; please define the initial Q and N for leaf nodes and clarify whether Q(s_i) in Eq. (2) is a cumulative score or an average score.
  3. [Algorithm 1, line 24] The condition 'if πθ(S,q,subq,P_rng,m)' should state explicitly what output of the LLM counts as acceptance, since the pseudocode otherwise leaves the behavior of the stack ambiguous.
  4. [Table VI] Table VI uses the variable 'k' for the tree width, but the paper uses 'b' for width and 'K' for the number of paths; please rename to avoid confusion.
  5. [§V-C] In the paragraph varying the number of iterations H, the text says 'when k is fixed' but H is the parameter being varied; this should be corrected to refer to the actual fixed hyperparameters.
  6. [§IV-B, Eq. (8)] Equation (8) writes EoS(s_j) = πθ(subq, s_j), but the preceding text says the self-critic receives the currently expanded historical reasoning path; the equation should include the reasoning path to match the description.

Circularity Check

0 steps flagged · score 0.0 of 10

No construction-level circularity; the empirical claims are anchored to external benchmarks and no reported result is derived from a fitted parameter by construction.

full rationale

RTSoG's contribution is an algorithmic pipeline (question decomposition, SC-MCTS retrieval, weighted path stacking), and its support is empirical EM accuracy on four public KGQA benchmarks. No component of the pipeline is defined in terms of the benchmark answers, and no reported EM is obtained by plugging a fitted parameter into an equation that forces the result. The reward and self-critic signals are LLM judgments rather than fitted quantities; using the same LLM as policy, value, and answer model makes the design self-referential, but it does not make the benchmark scores equivalent to the method's inputs. The closest concern is that hyperparameters H, b, K, and alpha are selected on the WebQSP and GrailQA test sets in Section V-C and the same test sets then produce Tables III and IV; that is test-set selection, an evaluation-validity risk, not a construction-level circularity. The GrailQA/WebQuestions subset identities are also not released, which undermines comparability checks but does not constitute a circular derivation. Self-citations such as EPREM [26] serve as baselines rather than load-bearing premises. Therefore the paper has no significant circularity under the stated definition.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claim is an empirical benchmark result, so most ledger entries are evaluation assumptions and search hyperparameters rather than mathematical axioms. No new physical or conceptual entities are postulated; the reasoning path stack and EoS signal are algorithmic components, not independent entities. The free parameters are tuned on the same test sets used for reporting, which is the main credibility risk.

free parameters (6)
  • H (number of MCTS iterations) = 24 for WebQSP, CWQ, WebQuestions; 18 for GrailQA
    Chosen via sensitivity analysis on the test sets (Fig 2b); higher H improves EM but costs compute.
  • b (tree width) = 7
    Selected from {3,5,7,9} in sensitivity analysis on WebQSP and GrailQA test sets (Fig 2c); larger b introduces noise.
  • K (number of weighted reasoning paths) = 10
    Chosen from sensitivity analysis (Fig 2d); too few paths lose important paths and too many add noise.
  • alpha (reward combination weight) = 0.33
    Given in implementation details with no ablation or justification; controls the balance between local relation reward and global path reward.
  • n (number of sub-questions) = 3
    Sensitivity analysis (Fig 2a) shows n=3 is best on WebQSP and GrailQA; chosen on the same test sets used for final results.
  • c (UCT exploration constant) = not reported
    Appears in Eq. 2 but its value is never stated in the implementation details; a hidden free parameter that affects the exploration-exploitation balance.
assumptions (4)
  • domain assumption Topic entities T_q and answer entities A_q are labeled and linked to entities in the knowledge graph for all test questions (Section III).
    This follows prior KGQA works and avoids the entity-linking problem, but does not hold in fully open-domain deployments.
  • domain assumption The LLM used as value model and self-critic produces well-calibrated reward scores (Eq. 7, Eq. 9) and reliable End-of-Search decisions (Eq. 8) that guide the search to correct paths.
    Since policy and value are the same LLM, the method's retrieved-context quality depends on these judgments being accurate; if they are poorly calibrated, search will select wrong paths.
  • domain assumption The Freebase subgraphs for the four datasets are sufficiently complete to contain the correct reasoning paths.
    If the KG lacks needed edges, even perfect search cannot retrieve correct paths; the paper assumes the benchmark KGs are adequate.
  • standard math The UCT formula Q/N + c*sqrt(ln N(p)/N(s_i)) is a valid bandit-style exploration-exploitation criterion for reasoning path selection.
    Standard MCTS assumption, widely accepted and not proved in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Enhancing Large Language Models with Reward-guided Tree Search for Knowledge Graph Question and Answering." pith.science (2026). https://pith.science/paper/OJQA5UUO

@misc{pith2026250512476,
  author       = {Pith},
  title        = {Pith review of: Enhancing Large Language Models with Reward-guided Tree Search for Knowledge Graph Question and Answering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OJQA5UUO}},
  note         = {Machine review of arXiv:2505.12476}
}
read the original abstract

Recently, large language models (LLMs) have demonstrated impressive performance in Knowledge Graph Question Answering (KGQA) tasks, which aim to find answers based on knowledge graphs (KGs) for natural language questions. Existing LLMs-based KGQA methods typically follow the Graph Retrieval-Augmented Generation (GraphRAG) paradigm, which first retrieves reasoning paths from the large KGs, and then generates the answers based on them. However, these methods emphasize the exploration of new optimal reasoning paths in KGs while ignoring the exploitation of historical reasoning paths, which may lead to sub-optimal reasoning paths. Additionally, the complex semantics contained in questions may lead to the retrieval of inaccurate reasoning paths. To address these issues, this paper proposes a novel and training-free framework for KGQA tasks called Reward-guided Tree Search on Graph (RTSoG). RTSoG decomposes an original question into a series of simpler and well-defined sub-questions to handle the complex semantics. Then, a Self-Critic Monte Carlo Tree Search (SC-MCTS) guided by a reward model is introduced to iteratively retrieve weighted reasoning paths as contextual knowledge. Finally, it stacks the weighted reasoning paths according to their weights to generate the final answers. Extensive experiments on four datasets demonstrate the effectiveness of RTSoG. Notably, it achieves 8.7\% and 7.0\% performance improvement over the state-of-the-art method on the GrailQA and the WebQSP respectively.

Figures

Figures reproduced from arXiv: 2505.12476 by the authors.

Figure 1
Figure 1. Overview of the proposed training-free RTSoG framework, which contains the three stages: Question Decomposition, Weighted Reasoning Paths [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The impact of four important hyper-parameters: the number of subquestions [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Ablation study on the self-critic mechanism and SC-MCTS in RTSoG. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: A typical case to analyze the difference between PoG and RTSoG in [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: A typical case to to analyze the impact of SC-MCTS in weighted [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 14 canonical work pages

  1. [1]

    Free- base: a collaboratively created graph database for structuring human knowledge,

    K. Bollacker, C. Evans, P. Paritosh, T. Sturge, and J. Taylor, “Free- base: a collaboratively created graph database for structuring human knowledge,” inProceedings of the 2008 ACM SIGMOD international conference on Management of data, 2008, pp. 1247–1250

    2008

  2. [2]

    Wikidata: a free collaborative knowl- edgebase,

    D. Vrande ˇci´c and M. Kr ¨otzsch, “Wikidata: a free collaborative knowl- edgebase,”Communications of the ACM, vol. 57, no. 10, pp. 78–85, 2014

    2014

  3. [3]

    Dbpedia: A nucleus for a web of open data,

    S. Auer, C. Bizer, G. Kobilarov, J. Lehmann, R. Cyganiak, and Z. Ives, “Dbpedia: A nucleus for a web of open data,” ininternational semantic web conference. Springer, 2007, pp. 722–735

    work page 2007

  4. [4]

    Complex knowledge base question answering: A survey,

    Y . Lan, G. He, J. Jiang, J. Jiang, W. X. Zhao, and J.-R. Wen, “Complex knowledge base question answering: A survey,”IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 11, pp. 11 196–11 215, 2022

    work page 2022

  5. [5]

    Hello gpt-4o,

    OpenAI, “Hello gpt-4o,” 2024. [Online]. Available: https://openai.com/ index/hello-gpt-4o/

    2024

  6. [6]

    Introducing chatgpt,

    ——, “Introducing chatgpt,” 2022. [Online]. Available: https://openai. com/index/chatgpt/

    work page 2022

  7. [7]

    Qwen2. 5 technical report,

    A. Yang, B. Yang, B. Zhang, B. Hui, B. Zheng, B. Yu, C. Li, D. Liu, F. Huang, H. Weiet al., “Qwen2. 5 technical report,”arXiv preprint arXiv:2412.15115, 2024

    arXiv 2024

  8. [8]

    Chain-of-thought prompting elicits reasoning in large language models,

    J. Wei, X. Wang, D. Schuurmans, M. Bosma, F. Xia, E. Chi, Q. V . Le, D. Zhouet al., “Chain-of-thought prompting elicits reasoning in large language models,”Advances in neural information processing systems, vol. 35, pp. 24 824–24 837, 2022

    2022

Show all 61 references
  1. [9]

    Tree of thoughts: Deliberate problem solving with large language models,

    S. Yao, D. Yu, J. Zhao, I. Shafran, T. Griffiths, Y . Cao, and K. Narasimhan, “Tree of thoughts: Deliberate problem solving with large language models,”Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  2. [10]

    Tree-of-traversals: A zero- shot reasoning algorithm for augmenting black-box language models with knowledge graphs,

    E. Markowitz, A. Ramakrishna, J. Dhamala, N. Mehrabi, C. Peris, R. Gupta, K.-W. Chang, and A. Galstyan, “Tree-of-traversals: A zero- shot reasoning algorithm for augmenting black-box language models with knowledge graphs,”arXiv preprint arXiv:2407.21358, 2024

    2024 arXiv

  3. [11]

    Think-on-graph 2.0: Deep and faithful large language model reasoning with knowledge-guided retrieval augmented generation,

    S. Ma, C. Xu, X. Jiang, M. Li, H. Qu, C. Yang, J. Mao, and J. Guo, “Think-on-graph 2.0: Deep and faithful large language model reasoning with knowledge-guided retrieval augmented generation,”arXiv preprint arXiv:2407.10805, 2024

    2024 arXiv

  4. [12]

    Plan-on-graph: Self-correcting adaptive planning of large language model on knowledge graphs,

    L. Chen, P. Tong, Z. Jin, Y . Sun, J. Ye, and H. Xiong, “Plan-on-graph: Self-correcting adaptive planning of large language model on knowledge graphs,”arXiv preprint arXiv:2410.23875, 2024

    2024 arXiv

  5. [13]

    Think-on-graph: Deep and responsible reasoning of large lan- guage model with knowledge graph,

    J. Sun, C. Xu, L. Tang, S. Wang, C. Lin, Y . Gong, H.-Y . Shum, and J. Guo, “Think-on-graph: Deep and responsible reasoning of large lan- guage model with knowledge graph,”arXiv preprint arXiv:2307.07697, 2023

    2023 arXiv

  6. [14]

    Graph retrieval-augmented generation: A survey,

    B. Peng, Y . Zhu, Y . Liu, X. Bo, H. Shi, C. Hong, Y . Zhang, and S. Tang, “Graph retrieval-augmented generation: A survey,”arXiv preprint arXiv:2408.08921, 2024

    2024 arXiv

  7. [15]

    Large-scale semantic parsing without question-answer pairs,

    S. Reddy, M. Lapata, and M. Steedman, “Large-scale semantic parsing without question-answer pairs,”Transactions of the Association for Computational Linguistics, vol. 2, pp. 377–392, 2014

    2014

  8. [16]

    Structgpt: A general framework for large language model to reason over structured data,

    J. Jiang, K. Zhou, Z. Dong, K. Ye, W. X. Zhao, and J.-R. Wen, “Structgpt: A general framework for large language model to reason over structured data,”arXiv preprint arXiv:2305.09645, 2023

    2023 arXiv

  9. [17]

    Rng-kbqa: Generation augmented iterative ranking for knowledge base question answering,

    X. Ye, S. Yavuz, K. Hashimoto, Y . Zhou, and C. Xiong, “Rng-kbqa: Generation augmented iterative ranking for knowledge base question answering,”arXiv preprint arXiv:2109.08678, 2021

    2021 arXiv

  10. [18]

    Arcaneqa: Dynamic program induction and contextu- alized encoding for knowledge base question answering,

    Y . Gu and Y . Su, “Arcaneqa: Dynamic program induction and contextu- alized encoding for knowledge base question answering,”arXiv preprint arXiv:2204.08109, 2022. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 11

    2022 arXiv

  11. [19]

    Query2box: Reasoning over knowl- edge graphs in vector space using box embeddings,

    H. Ren, W. Hu, and J. Leskovec, “Query2box: Reasoning over knowl- edge graphs in vector space using box embeddings,”arXiv preprint arXiv:2002.05969, 2020

    2002 arXiv

  12. [20]

    Neural-based mixture probabilistic query embedding for answering fol queries on knowledge graphs,

    X. Long, L. Zhuang, L. Aodi, S. Wang, and H. Li, “Neural-based mixture probabilistic query embedding for answering fol queries on knowledge graphs,” inProceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, pp. 3001–3013

    2022

  13. [21]

    Subgraph retrieval enhanced model for multi-hop knowledge base question answering,

    J. Zhang, X. Zhang, J. Yu, J. Tang, J. Tang, C. Li, and H. Chen, “Subgraph retrieval enhanced model for multi-hop knowledge base question answering,”arXiv preprint arXiv:2202.13296, 2022

    2022 arXiv

  14. [22]

    Improving multi-hop knowledge base question answering by learning intermediate supervision signals,

    G. He, Y . Lan, J. Jiang, W. X. Zhao, and J.-R. Wen, “Improving multi-hop knowledge base question answering by learning intermediate supervision signals,” inProceedings of the 14th ACM international conference on web search and data mining, 2021, pp. 553–561

    2021

  15. [23]

    Pullnet: Open domain question answering with iterative retrieval on knowledge bases and text,

    H. Sun, T. Bedrax-Weiss, and W. W. Cohen, “Pullnet: Open domain question answering with iterative retrieval on knowledge bases and text,” arXiv preprint arXiv:1904.09537, 2019

    1904 arXiv

  16. [24]

    Gnn-rag: Graph neural retrieval for large language model reasoning,

    C. Mavromatis and G. Karypis, “Gnn-rag: Graph neural retrieval for large language model reasoning,”arXiv preprint arXiv:2405.20139, 2024

    2024 arXiv

  17. [25]

    Reasoning on graphs: Faith- ful and interpretable large language model reasoning,

    L. Luo, Y .-F. Li, G. Haffari, and S. Pan, “Reasoning on graphs: Faith- ful and interpretable large language model reasoning,”arXiv preprint arXiv:2310.01061, 2023

    2023 arXiv

  18. [26]

    Eperm: An evidence path enhanced reasoning model for knowledge graph question and answering,

    X. Long, L. Zhuang, A. Li, M. Yao, and S. Wang, “Eperm: An evidence path enhanced reasoning model for knowledge graph question and answering,”arXiv preprint arXiv:2502.16171, 2025

    2025 arXiv

  19. [27]

    Unikgqa: Unified retrieval and reasoning for solving multi-hop question answering over knowledge graph,

    J. Jiang, K. Zhou, W. X. Zhao, and J.-R. Wen, “Unikgqa: Unified retrieval and reasoning for solving multi-hop question answering over knowledge graph,”arXiv preprint arXiv:2212.00959, 2022

    2022 arXiv

  20. [28]

    Self-consistency improves chain of thought reasoning in language models,

    X. Wang, J. Wei, D. Schuurmans, Q. Le, E. Chi, S. Narang, A. Chowdh- ery, and D. Zhou, “Self-consistency improves chain of thought reasoning in language models,”arXiv preprint arXiv:2203.11171, 2022

    2022 arXiv

  21. [29]

    Tree of thoughts: Deliberate problem solving with large language models,

    S. Yao, D. Yu, J. Zhao, I. Shafran, T. Griffiths, Y . Cao, and K. Narasimhan, “Tree of thoughts: Deliberate problem solving with large language models,”Advances in neural information processing systems, vol. 36, pp. 11 809–11 822, 2023

    2023

  22. [30]

    React: Synergizing reasoning and acting in language models,

    S. Yao, J. Zhao, D. Yu, N. Du, I. Shafran, K. Narasimhan, and Y . Cao, “React: Synergizing reasoning and acting in language models,” in International Conference on Learning Representations (ICLR), 2023

    2023

  23. [31]

    Q*: Improving multi-step reasoning for llms with deliberative planning,

    C. Wang, Y . Deng, Z. Lyu, L. Zeng, J. He, S. Yan, and B. An, “Q*: Improving multi-step reasoning for llms with deliberative planning,” arXiv preprint arXiv:2406.14283, 2024

    2024 arXiv

  24. [32]

    Mutual reasoning makes smaller llms stronger problem-solvers,

    Z. Qi, M. Ma, J. Xu, L. L. Zhang, F. Yang, and M. Yang, “Mutual reasoning makes smaller llms stronger problem-solvers,”arXiv preprint arXiv:2408.06195, 2024

    2024 arXiv

  25. [33]

    Llama-berry: Pairwise optimization for o1-like olympiad-level mathematical reasoning,

    D. Zhang, J. Wu, J. Lei, T. Che, J. Li, T. Xie, X. Huang, S. Zhang, M. Pavone, Y . Liet al., “Llama-berry: Pairwise optimization for o1-like olympiad-level mathematical reasoning,”arXiv preprint arXiv:2410.02884, 2024

    2024 arXiv

  26. [34]

    Large language monkeys: Scaling inference compute with repeated sampling,

    B. Brown, J. Juravsky, R. Ehrlich, R. Clark, Q. V . Le, C. R ´e, and A. Mirhoseini, “Large language monkeys: Scaling inference compute with repeated sampling,”arXiv preprint arXiv:2407.21787, 2024

    2024 arXiv

  27. [35]

    Scaling llm test-time compute optimally can be more effective than scaling model parameters,

    C. Snell, J. Lee, K. Xu, and A. Kumar, “Scaling llm test-time compute optimally can be more effective than scaling model parameters,”arXiv preprint arXiv:2408.03314, 2024

    2024 arXiv

  28. [36]

    Rest- mcts*: Llm self-training via process reward guided tree search,

    D. Zhang, S. Zhoubian, Z. Hu, Y . Yue, Y . Dong, and J. Tang, “Rest- mcts*: Llm self-training via process reward guided tree search,”arXiv preprint arXiv:2406.03816, 2024

    2024 arXiv

  29. [37]

    Mastering chess and shogi by self-play with a general reinforcement learning algorithm,

    D. Silver, T. Hubert, J. Schrittwieser, I. Antonoglou, M. Lai, A. Guez, M. Lanctot, L. Sifre, D. Kumaran, T. Graepelet al., “Mastering chess and shogi by self-play with a general reinforcement learning algorithm,” arXiv preprint arXiv:1712.01815, 2017

    2017 arXiv

  30. [38]

    Mastering the game of go with deep neural networks and tree search,

    D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. Van Den Driessche, J. Schrittwieser, I. Antonoglou, V . Panneershelvam, M. Lanctotet al., “Mastering the game of go with deep neural networks and tree search,”nature, vol. 529, no. 7587, pp. 484–489, 2016

    2016

  31. [39]

    Mastering atari, go, chess and shogi by planning with a learned model,

    J. Schrittwieser, I. Antonoglou, T. Hubert, K. Simonyan, L. Sifre, S. Schmitt, A. Guez, E. Lockhart, D. Hassabis, T. Graepelet al., “Mastering atari, go, chess and shogi by planning with a learned model,” Nature, vol. 588, no. 7839, pp. 604–609, 2020

    2020

  32. [40]

    Bandit based monte-carlo planning,

    L. Kocsis and C. Szepesv ´ari, “Bandit based monte-carlo planning,” in European conference on machine learning. Springer, 2006, pp. 282– 293

    2006

  33. [41]

    Learning entity and relation embeddings for knowledge graph completion,

    Y . Lin, Z. Liu, M. Sun, Y . Liu, and X. Zhu, “Learning entity and relation embeddings for knowledge graph completion,” inProceedings of the AAAI conference on artificial intelligence, vol. 29, no. 1, 2015

    2015

  34. [42]

    Kgdm: A diffusion model to capture multiple relation semantics for knowledge graph embedding,

    X. Long, L. Zhuang, A. Li, J. Wei, H. Li, and S. Wang, “Kgdm: A diffusion model to capture multiple relation semantics for knowledge graph embedding,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 8, 2024, pp. 8850–8858

    2024

  35. [43]

    Fact embedding through diffusion model for knowledge graph completion,

    X. Long, L. Zhuang, A. Li, H. Li, and S. Wang, “Fact embedding through diffusion model for knowledge graph completion,” inProceed- ings of the ACM Web Conference 2024, 2024, pp. 2020–2029

    2024

  36. [44]

    Let’s verify step by step,

    H. Lightman, V . Kosaraju, Y . Burda, H. Edwards, B. Baker, T. Lee, J. Leike, J. Schulman, I. Sutskever, and K. Cobbe, “Let’s verify step by step,”arXiv preprint arXiv:2305.20050, 2023

    2023 arXiv

  37. [45]

    The web as a knowledge-base for answering complex questions,

    A. Talmor and J. Berant, “The web as a knowledge-base for answering complex questions,”arXiv preprint arXiv:1803.06643, 2018

    2018 arXiv

  38. [46]

    The value of semantic parse labeling for knowledge base question answer- ing,

    W.-t. Yih, M. Richardson, C. Meek, M.-W. Chang, and J. Suh, “The value of semantic parse labeling for knowledge base question answer- ing,” inProceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 2016, pp. 201–206

    2016

  39. [47]

    Beyond iid: three levels of generalization for question answering on knowledge bases,

    Y . Gu, S. Kase, M. Vanni, B. Sadler, P. Liang, X. Yan, and Y . Su, “Beyond iid: three levels of generalization for question answering on knowledge bases,” inProceedings of the Web Conference 2021, 2021, pp. 3477–3488

    2021

  40. [48]

    Semantic parsing on freebase from question-answer pairs,

    J. Berant, A. Chou, R. Frostig, and P. Liang, “Semantic parsing on freebase from question-answer pairs,” inProceedings of the 2013 conference on empirical methods in natural language processing, 2013, pp. 1533–1544

    2013

  41. [49]

    Chain-of-knowledge: Grounding large language models via dynamic knowledge adapting over heterogeneous sources,

    X. Li, R. Zhao, Y . K. Chia, B. Ding, S. Joty, S. Poria, and L. Bing, “Chain-of-knowledge: Grounding large language models via dynamic knowledge adapting over heterogeneous sources,”arXiv preprint arXiv:2305.13269, 2023

    2023 arXiv

  42. [50]

    Knowledge-augmented language model prompting for zero-shot knowledge graph question answering,

    J. Baek, A. F. Aji, and A. Saffari, “Knowledge-augmented language model prompting for zero-shot knowledge graph question answering,” arXiv preprint arXiv:2306.04136, 2023

    2023 arXiv

  43. [51]

    Build the future of ai with meta llama 3,

    Meta, “Build the future of ai with meta llama 3,” 2024. [Online]. Available: https://llama.meta.com/llama3/

    2024

  44. [52]

    Hello gpt-4,

    OpenAI, “Hello gpt-4,” 2023. [Online]. Available: https://openai.com/ index/gpt-4/

    2023

  45. [53]

    Open domain question answering using early fusion of knowledge bases and text,

    H. Sun, B. Dhingra, M. Zaheer, K. Mazaitis, R. Salakhutdinov, and W. W. Cohen, “Open domain question answering using early fusion of knowledge bases and text,”arXiv preprint arXiv:1809.00782, 2018

    2018 arXiv

  46. [54]

    Enhancing complex question answering over knowledge graphs through evidence pattern retrieval,

    W. Ding, J. Li, L. Luo, and Y . Qu, “Enhancing complex question answering over knowledge graphs through evidence pattern retrieval,” inProceedings of the ACM on Web Conference 2024, 2024, pp. 2106– 2115

    2024

  47. [55]

    Decaf: Joint decoding of answers and logical forms for question answering over knowledge bases,

    D. Yu, S. Zhang, P. Ng, H. Zhu, A. H. Li, J. Wang, Y . Hu, W. Wang, Z. Wang, and B. Xiang, “Decaf: Joint decoding of answers and logical forms for question answering over knowledge bases,”arXiv preprint arXiv:2210.00063, 2022

    arXiv

  48. [56]

    Tiara: Multi-grained retrieval for robust question answering over large knowledge bases,

    Y . Shu, Z. Yu, Y . Li, B. F. Karlsson, T. Ma, Y . Qu, and C.-Y . Lin, “Tiara: Multi-grained retrieval for robust question answering over large knowledge bases,”arXiv preprint arXiv:2210.12925, 2022

    2022 arXiv

  49. [57]

    Fc-kbqa: A fine-to-coarse composition framework for knowledge base question answering,

    L. Zhang, J. Zhang, Y . Wang, S. Cao, X. Huang, C. Li, H. Chen, and J. Li, “Fc-kbqa: A fine-to-coarse composition framework for knowledge base question answering,”arXiv preprint arXiv:2306.14722, 2023

    2023 arXiv

  50. [58]

    Flexkbqa: A flexible llm-powered framework for few-shot knowledge base question answering,

    Z. Li, S. Fan, Y . Gu, X. Li, Z. Duan, B. Dong, N. Liu, and J. Wang, “Flexkbqa: A flexible llm-powered framework for few-shot knowledge base question answering,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 17, 2024, pp. 18 608–18 616

    2024

  51. [59]

    Distribution shifts are bottlenecks: Extensive evalua- tion for grounding language models to knowledge bases,

    Y . Shu and Z. Yu, “Distribution shifts are bottlenecks: Extensive evalua- tion for grounding language models to knowledge bases,” inProceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics: Student Research Workshop, 2024, pp. 71– 88

    2024

  52. [60]

    Interactive-kbqa: Multi-turn inter- actions for knowledge base question answering with large language models,

    G. Xiong, J. Bao, and W. Zhao, “Interactive-kbqa: Multi-turn inter- actions for knowledge base question answering with large language models,”arXiv preprint arXiv:2402.15131, 2024

    2024 arXiv

  53. [61]

    Few-shot in- context learning for knowledge base question answering,

    T. Li, X. Ma, A. Zhuang, Y . Gu, Y . Su, and W. Chen, “Few-shot in- context learning for knowledge base question answering,”arXiv preprint arXiv:2305.01750, 2023

    2023 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.