Pith. sign in

REVIEW 3 major objections 4 minor 42 references

PAGE-RAG: Evidence-Grounded Adaptive Graph Retrieval for Long-Document Question Answering

T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read PAGE-RAG treats graphs as lossy projections, not knowledge replacements, and gates answers on an explicit evidence-sufficiency check — yielding ~90% lenient accuracy with perfect refusal on all 24 unanswerable questions.

desk verdict Honest, well-scoped GraphRAG paper with a genuinely useful abstention mechanism, but the central answer-or-refuse decision rests on an unspecified evidence-sufficiency check that must be disclosed before the reliability claim is fully credible. read the letter →

arxiv 2607.19301 v1 pith:GVZHRA6I submitted 2026-07-21 cs.IR

Xingyu Chen , Junxiu An , Jun Guo , Li Wang This is my paper · ORCID
classification cs.IR
keywords GraphRAGretrieval-augmentedgenerationlong-documentquestionansweringknowledgegraphsprojection-awareretrievaladaptiveroutingevidence-boundedabstention
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 tries to establish that automatically constructed knowledge graphs should be treated as lossy semantic skeletons over source text, not as independent knowledge bases, and that reliable long-document QA requires three design choices: keep an always-on textual evidence floor, route queries to graph structure only when the query needs it, and make the final answer-or-abstain decision depend on an explicit evidence-sufficiency check. It claims this combination yields competitive answer quality while making knowledge-boundary violations rare: across two book-length corpora, PAGE-RAG reaches lenient accuracy of 92.6% and 87.5% and refuses all 24 unanswerable questions, where a strong passage baseline without the evidence gate correctly refuses only 4 of 24. It further claims a large efficiency gain, using roughly 1,956 query tokens per question on a global sensemaking benchmark versus 136,509 for a heavyweight graph-community system, while adding a balanced point on the quality–efficiency frontier rather than claiming universal dominance. A sympathetic reader should care because the paper reframes GraphRAG's goal: not just more answers, but answers that respect the boundary of available evidence, with abstention treated as a correct action.

What carries the argument

The central machinery is the projection-aware hybrid repository R(D)=(T(D), G(D), P, b): a textual evidence floor of citable chunks, a governed graph skeleton with communities and summaries, document profiles that condition extraction, and evidence bindings mapping every graph element to supporting spans. Around it sit the query-adaptive routing policy Π(u,O) — which composes retrieval operators (textual, neighborhood, path, community, rerank) under budget B — and the evidence-bounded generator with the sufficiency check s(q,E)∈{0,1}, which turns 'enough evidence?' into a binary answer-or-refuse decision. The binding keeps graph-guided answers traceable; the sufficiency check makes abstentio

What would settle it

Construct a test set of answerable questions with the key evidence removed but a thematically related distractor left in the retrieved package; if PAGE-RAG answers with citations when its sufficiency check says 'sufficient' on such a package — or refuses on genuinely sufficient evidence — then the gate is not tracking evidence sufficiency. Concretely: on the two book datasets, hide the chapter that actually contains the answer while leaving a paragraph that mentions the same characters, and observe whether the system abstains or re-retrieves instead of answering.

Watch

Extended reading notes

Core claim

PAGE-RAG's central claim: an automatically constructed graph is a lossy projection of a long document, so it should navigate and organize evidence, not replace it. The system pairs a citable textual floor T(D) with a governed graph skeleton G(D) via evidence bindings b, and a routing policy activates textual, neighborhood, path, and community-summary operators under token budgets. At generation, an evidence-sufficiency check s(q,E) decides between a cited answer and refusal. Empirically, this yields 92.6% and 87.5% lenient accuracy with perfect refusal on all 24 unanswerable questions; removing the evidence gate raises apparent accuracy to 95.1% but collapses correct refusal to 4/24. On a gl

Load-bearing premise

Everything hinges on the evidence-sufficiency check s(q,E) — the paper's central answer-or-refuse gate — but the manuscript does not describe how it is computed, trained, or calibrated, so the claimed reliable abstention rests on an unverified component.

Editorial extensions

If this is right

  • Across two book-length datasets, a system can maintain competitive accuracy (92.6%/87.5% lenient) while correctly refusing every unanswerable question, so answer quality and boundary reliability are not mutually exclusive.
  • An explicit evidence-sufficiency gate, not a prompt instruction, is what enforces reliable abstention: removing it turns 22 of 24 unanswerable questions into substantive out-of-evidence answers even though apparent accuracy rises.
  • The textual floor is load-bearing: dropping it cuts lenient accuracy from 90.1% to 58.2%, implying graph-only retrieval cannot carry long-document QA alone.
  • Query-adaptive routing concentrates graph-path computation on queries that need structure, improving accuracy exactly where structural evidence helps (94.1%→89.7% when structural guidance is removed on routed questions).
  • Heavy community-summary style GraphRAG is a high-quality, high-cost extreme; a projection-aware system can approach it on global sensemaking while using ~1/70th of the query tokens.

Reading between the lines

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

  • The answer-or-refuse gate, if hardened, could double as a permission-aware access-control layer for private RAG deployments, abstaining when evidence exists but the user's tier does not authorize it — a direction the paper only gestures at.
  • The routing policy's token savings (1,956 vs 136,509 per question) suggest that many 'global' queries can be served by a targeted subset of community summaries; a testable extension is measuring how summary count and precision trade off as the budget shrinks.
  • Because document profiles condition both extraction and routing, the framework predicts that a single fixed entity-relation schema underperforms profile-conditioned schemas on heterogeneous corpora; this could be tested by running the same pipeline across mixed fiction/nonfiction collections.
  • The 22/24 collapse when the evidence gate is removed implies that accuracy-only leaderboards reward systems that answer beyond their evidence; an editorial consequence is that closed-corpus benchmarks should report correct-refusal alongside accuracy to expose boundary violations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes PAGE-RAG, a GraphRAG framework for long-document question answering that combines a textual passage index with a graph-based 'semantic skeleton,' routes queries adaptively among retrieval operators, and makes the answer-versus-refuse decision depend on an explicit evidence-sufficiency check. Evaluated on two book-length datasets and the UltraDomain-Mix benchmark, the system is reported to achieve competitive answer quality while using far fewer query tokens than a community-summary GraphRAG baseline, and to abstain correctly on all 24 unanswerable book questions. Ablations attribute the refusal behavior to the evidence constraint and the answer quality to the textual retrieval floor.

Significance. The paper addresses a real and under-emphasized problem: automatically constructed graphs are lossy projections of source documents, and GraphRAG systems should be able to abstain when evidence is insufficient. The authors are appropriately honest that PAGE-RAG is a balanced operating point rather than a universal improvement, and the positive results are accompanied by code release and ablations that support the role of the textual floor and the evidence constraint. If the evidence-sufficiency mechanism can be specified and the datasets made auditable, the contribution would be useful to the GraphRAG community. As it stands, the central reliability claim is not yet reproducible from the manuscript.

major comments (3)
  1. [Evidence-Bounded Generation (Section 4)] The mapping y = a if s(q,E)=1 else refusal is the load-bearing mechanism for the paper's central reliability claim, but the manuscript never states how s(q,E) is computed. There is no classifier, no prompt, no threshold, no calibration data, and no error analysis. The 12/12 correct refusals on each book and the contrast in Table 3 between 24/24 and 4/24 refusals depend entirely on this unspecified binary check. Without a precise specification and an evaluation of s on a range of sufficient and insufficient evidence sets, the 'evidence-bounded generation' contribution cannot be reproduced, audited, or distinguished from a model-specific artifact. This is an internal reproducibility gap, not a matter of taste. Please specify the decision procedure and provide calibration/error analysis.
  2. [Experimental Setup: Datasets] The two book-length datasets (Simulacra and Simulation, One Hundred Years of Solitude) appear to be self-constructed, but no construction protocol is described: how were the answerable and unanswerable questions written, what defined 'unanswerable' for the annotators, how was the gold answer determined, and was there any inter-annotator agreement? The claim of 'correctly refuses all 24 unanswerable questions' rests on 24 labeled instances whose provenance and labels are not auditable. The authors should release the questions and evidence annotations, describe the annotation instructions, and ideally report confidence intervals for the refusal rate and accuracy metrics.
  3. [Query-Adaptive Retrieval Routing (Section 2) and Table 2] The routing policy Π(u,O) and the evidence budgets B=(Bt,Bn,Bp,Bc) are central to the efficiency claims, but the paper does not specify how they are set: are they hand-tuned, prompt-based, or learned? No sensitivity analysis is reported. The dramatic token reduction in Table 2 (1,956 vs 136,509 query tokens per question against Microsoft GraphRAG) could be an artifact of an aggressive budget choice rather than a property of the routing framework. Please report the concrete values used for the budgets and thresholds, and include a sensitivity study showing how BBS and token cost vary when these budgets are changed.
minor comments (4)
  1. [Evaluation metrics] 'Strict' and 'lenient' accuracy are used in Table 1 but never defined. Please define both in Section 5.
  2. [Figure 2] The caption mentions a 'PAGE-RAG-anchored four-state quality' index without defining it. The four states and how the pairwise judgments are converted into this index should be stated.
  3. [Table 1 formatting] Table 1 repeats the dataset name in every row; using row spans or grouping would improve readability. Consider also adding the number of questions in the table caption.
  4. [References] Several references are to arXiv preprints with 2026 dates that may not be stable or peer reviewed. Consider citing published versions where available.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: claims are empirical and externally benchmarked; the only flag is the unspecified evidence-sufficiency check, which is a reproducibility gap, not a circular derivation.

full rationale

PAGE-RAG is an empirical systems paper. The three framework components (projection-aware hybrid repository, query-adaptive retrieval routing, evidence-bounded generation) are presented as design choices and tested against external baselines (BDR-RAG, LightRAG, Microsoft GraphRAG) on two book-length datasets and UltraDomain-Mix, with ablations reported. The central claims are not derived from the inputs by construction; they are measured trade-offs among accuracy, refusal, latency, and token cost. The ablation evidence is particularly important: removing the evidence constraint changes correct refusal from 24/24 to 4/24 (Table 3), so the boundary behavior is empirically attributed to the evidence-bounded component rather than assumed in the formal setup. The only self-citation (ref [2]) appears in a general introductory sentence and is not load-bearing. The evidence-sufficiency check s(q,E) in Section 4 ('Evidence-Bounded Generation') is not specified: the paper provides no model, prompt, thresholds, training data, or calibration evidence. This is a genuine reproducibility gap and a risk to the central reliability claim, but the paper does not define s in terms of the unanswerable labels nor report fitting it to the test set. A missing implementation is not, on the available evidence, a demonstrated case of a prediction reducing to its inputs. Therefore no significant circularity; score 1 reflects the peripheral self-citation and the unspecified boundary mechanism, not a circular derivation.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new entities or forces; it is an engineering framework. Its reliance is on domain assumptions about graph lossiness and on unverified implementations of the router and the sufficiency check. The free parameters are unquantified budgets and routing thresholds that are likely tuned per dataset.

free parameters (2)
  • Evidence budgets B_t, B_n, B_p, B_c = not reported
    Budgets controlling passage tokens, graph neighborhood scope, path count, and community summaries are introduced in Section 4; no values or tuning procedure are given. They likely affect the quality–efficiency trade-off and may be per-dataset.
  • Routing policy thresholds / prompt for query profile u=(α,β,γ,η) = not reported
    The query profile classification is LLM-based but the prompt, temperature, or any decision thresholds are unspecified. The router's accuracy is load-bearing for the adaptive-retrieval claim.
assumptions (3)
  • domain assumption Automatically constructed graphs are lossy projections of source text.
    This is the central premise motivating the always-on textual floor. It is plausible and supported by cited refs [21–23], but it is an empirical claim about graph extraction quality, not a proven theorem.
  • domain assumption The evidence sufficiency function s(q,E) can be reliably implemented and calibrated.
    Section 4 defines s(q,E) as a binary function but gives no implementation. The entire answer-or-refuse mechanism depends on this unverified assumption.
  • domain assumption The query-adaptive routing policy Π accurately identifies query evidence needs.
    The routing policy is described abstractly (Section 4, 'Query-Adaptive Retrieval Routing'); its accuracy is not measured or discussed, yet it determines which retrieval operators are activated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PAGE-RAG: Evidence-Grounded Adaptive Graph Retrieval for Long-Document Question Answering." pith.science (2026). https://pith.science/paper/GVZHRA6I

@misc{pith2026260719301,
  author       = {Pith},
  title        = {Pith review of: PAGE-RAG: Evidence-Grounded Adaptive Graph Retrieval for Long-Document Question Answering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GVZHRA6I}},
  note         = {Machine review of arXiv:2607.19301}
}
read the original abstract

GraphRAG improves long-document question answering by introducing structured representations beyond conventional retrieval. However, automatically constructed graphs are inherently incomplete projections of source documents, and treating them as independent knowledge sources may lead to unreliable retrieval and generation. We propose PAGE-RAG, a projection-aware adaptive graph retrieval framework for reliable long-document question answering. PAGE-RAG views graph structures as semantic skeletons that organize and navigate document knowledge, rather than replacing the original knowledge source. Based on this perspective, PAGE-RAG introduces a task-adaptive retrieval routing strategy that dynamically selects appropriate retrieval behaviors according to query requirements. Furthermore, PAGE-RAG incorporates strict knowledge boundary control, ensuring that generated responses remain grounded within available evidence and abstaining from unsupported information beyond the accessible knowledge scope. Experiments demonstrate that PAGE-RAG achieves competitive answer quality while improving retrieval efficiency and knowledge reliability, highlighting the importance of projection-aware graph modeling, adaptive retrieval, and explicit knowledge boundary control for trustworthy GraphRAG systems. The source code is publicly available at https://github.com/CXY0112/PAGE-RAG.

Figures

Figures reproduced from arXiv: 2607.19301 by the authors.

Figure 1
Figure 1. PAGE-RAG framework. The build stage constructs a hybrid repository with both [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Empirical quality–efficiency frontier across book-length QA and UltraDomain [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 10 linked inside Pith

  1. [1]

    Yue, A survey of large language model agents for question answering, arXiv preprint arXiv:2503.19213 (2025)

    M. Yue, A survey of large language model agents for question answering, arXiv preprint arXiv:2503.19213 (2025)

    arXiv 2025

  2. [2]

    X. Chen, J. An, J. Guo, L. Wang, J. Guo, Kg -augmented executable cot for mathematical coding, Neural Networks 202 (2026) 109006. doi:https://doi.org/10.1016/j.neunet.2026.109006. URL https://www.sciencedirect.com/science/article/pii/ S0893608026004673

    arXiv 2026

  3. [3]

    Masterman, S

    T. Masterman, S. Besen, M. Sawtell, A. Chao, The landscape of emerg- ing ai agent architectures for reasoning, planning, and tool calling: A survey, arXiv preprint arXiv:2404.11584 (2024)

    arXiv 2024

  4. [4]

    N. F. Liu, K. Lin, J. Hewitt, A. Paranjape, M. Bevilacqua, F. Petroni, P. Liang, Lost in the middle: How language models use long contexts, 17 Transactions of the association for computational linguistics 12 (2024) 157–173

    2024

  5. [5]

    Press, N

    O. Press, N. A. Smith, M. Lewis, Train short, test long: Attention with linear biases enables input length extrapolation, arXiv preprint arXiv:2108.12409 (2021)

    arXiv 2021

  6. [6]

    S. Chen, S. Wong, L. Chen, Y. Tian, Extending context window of large language models via positional interpolation, arXiv preprint arXiv:2306.15595 (2023)

    arXiv 2023

  7. [7]

    R. Xu, Z. Qi, Z. Guo, C. Wang, H. Wang, Y. Zhang, W. Xu, Knowledge conflicts for llms: A survey, in: Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024, pp. 8541– 8565

    2024

  8. [8]

    Lewis, E

    P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Küttler, M. Lewis, W.-t. Yih, T. Rocktäschel, et al., Retrieval- augmented generation for knowledge-intensive nlp tasks, Advances in neural information processing systems 33 (2020) 9459–9474

    2020

Show all 42 references
  1. [9]

    Nogueira, K

    R. Nogueira, K. Cho, Passage re-ranking with bert, arXiv preprint arXiv:1901.04085 (2019)

    1901 arXiv

  2. [10]

    W. Sun, L. Yan, X. Ma, S. Wang, P. Ren, Z. Chen, D. Yin, Z. Ren, Is chatgpt good at search? investigating large language models as re- ranking agents, in: Proceedings of the 2023 conference on empirical methods in natural language processing, 2023, pp. 14918–14937

    2023

  3. [11]

    Wang, Retrieval-augmented generation for large language models: A survey, arXiv preprint arXiv:2312.10997 (2023)

    Y.Gao,Y.Xiong,X.Gao,K.Jia,J.Pan,Y.Bi,Y.Dai,J.Sun,M.Wang, H. Wang, Retrieval-augmented generation for large language models: A survey, arXiv preprint arXiv:2312.10997 (2023)

    2023 arXiv

  4. [12]

    Barnett, S

    S. Barnett, S. Kurniawan, S. Thudumu, Z. Brannelly, M. Abdelrazek, Seven failure points when engineering a retrieval augmented generation system, in: Proceedings of the IEEE/ACM 3rd International Conference on AI Engineering-Software Engineering for AI, 2024, pp. 194–199

    2024

  5. [13]

    X.He,Y.Tian,Y.Sun,N.V.Chawla,T.Laurent,Y.LeCun,X.Bresson, B. Hooi, G-retriever: Retrieval-augmented generation for textual graph 18 understanding and question answering, Advances in Neural Information Processing Systems 37 (2024) 132876–132907

    2024

  6. [14]

    Y. Yao, P. Wang, B. Tian, S. Cheng, Z. Li, S. Deng, H. Chen, N. Zhang, Editing large language models: Problems, methods, and opportunities, in: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, 2023, pp. 10222–10240

    2023

  7. [15]

    H. Han, L. Ma, Y. Wang, H. Shomer, Y. Lei, Z. Qi, K. Guo, Z. Hua, B. Long, H. Liu, et al., Rag vs. graphrag: A systematic evaluation and key insights, arXiv preprint arXiv:2502.11371 (2025)

    2025

  8. [16]

    D. Edge, H. Trinh, N. Cheng, J. Bradley, A. Chao, A. Mody, S. Truitt, J. Larson, From local to global: A graph RAG approach to query-focused summarization (2024). arXiv:2404.16130

    2024 arXiv

  9. [17]

    V. A. Traag, L. Waltman, N. J. Van Eck, From louvain to leiden: guaran- teeing well-connected communities, Scientific reports 9 (1) (2019) 5233

    2019

  10. [18]

    Song,Z.Xu,M

    J. Song,Z.Xu,M. He,J.Feng,B.Shen,Graph retrievalaugmentedlarge language models for facial phenotype associated rare genetic disease, NPJ digital medicine 8 (1) (2025) 543

    2025

  11. [19]

    Z. Zhu, X. Yuan, M. Galkin, L.-P. Xhonneux, M. Zhang, M. Gazeau, J.Tang,A*net:Ascalablepath-basedreasoningapproachforknowledge graphs, Advances in neural information processing systems 36 (2023) 59323–59336

    2023

  12. [20]

    W. Tao, X. Li, Y. Lan, W. Qian, Tagrag: Tag-guided hierarchical knowl- edge graph retrieval-augmented generation, in: Findings of the Associa- tion for Computational Linguistics: ACL 2026, 2026, pp. 6434–6456

    2026

  13. [21]

    B. Chen, Z. Guo, Z. Yang, Y. Chen, J. Chen, Z. Liu, C. Shi, C. Yang, Pathrag: Pruning graph-based retrieval augmented generation with re- lational paths, in: Proceedings of the AAAI conference on artificial in- telligence, Vol. 40, 2026, pp. 30183–30191

    2026

  14. [22]

    B. Peng, Y. Zhu, Y. Liu, X. Bo, H. Shi, C. Hong, Y. Zhang, S. Tang, Graph retrieval-augmented generation: A survey, ACM Transactions on Information Systems 44 (2) (2025) 1–52. 19

    2025

  15. [23]

    Zhang, S

    Q. Zhang, S. Chen, Y. Bei, Z. Yuan, H. Zhou, Z. Hong, H. Chen, Y. Xiao, C. Zhou, J. Dong, et al., A survey of graph retrieval- augmented generation for customized large language models, arXiv preprint arXiv:2501.13958 (2025)

    2025

  16. [24]

    K. S. Shin, The reasoning trap: An information-theoretic bound on closed-system multi-step llm reasoning, arXiv preprint arXiv:2605.01704 (2026)

    2026 arXiv

  17. [25]

    Petroni, T

    F. Petroni, T. Rocktäschel, S. Riedel, P. Lewis, A. Bakhtin, Y. Wu, A. Miller, Language models as knowledge bases?, in: Proceedings of the 2019 conference on empirical methods in natural language processing andthe9thinternationaljointconferenceonnaturallanguageprocessing (EMNL...

    2019

  18. [26]

    K. Meng, D. Bau, A. Andonian, Y. Belinkov, Locating and editing fac- tual associations in gpt, Advances in neural information processing sys- tems 35 (2022) 17359–17372

    2022

  19. [27]

    Y. Bai, Z. Liu, H. Yan, Z. Xu, Y. Wan, C. Chen, Z. Wang, X. Yuan, Y. Huang, G. Dou, et al., Inference-time control for trustworthy large language models (2026)

    2026

  20. [28]

    J. Mao, C. Gan, P. Kohli, J. B. Tenenbaum, J. Wu, The neuro-symbolic concept learner: Interpreting scenes, words, and sentences from natural supervision, arXiv preprint arXiv:1904.12584 (2019)

    1904 arXiv

  21. [29]

    Hitzler, A

    P. Hitzler, A. Eberhart, M. Ebrahimi, M. K. Sarker, L. Zhou, Neuro- symbolic approaches in artificial intelligence, National Science Review 9 (6) (2022) nwac035

    2022

  22. [30]

    W. Chen, X. Ma, X. Wang, W. W. Cohen, Program of thoughts prompt- ing: Disentangling computation from reasoning for numerical reasoning tasks, arXiv preprint arXiv:2211.12588 (2022)

    2022 arXiv

  23. [31]

    Aglionby, S

    G. Aglionby, S. Teufel, Faithful knowledge graph explanations in com- monsense question answering, in: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, pp. 10811– 10817. 20

    2022

  24. [32]

    10764–10799

    L.Gao,A.Madaan,S.Zhou,U.Alon,P.Liu,Y.Yang,J.Callan,G.Neu- big, Pal: Program-aided language models, in: International conference on machine learning, PMLR, 2023, pp. 10764–10799

    2023

  25. [33]

    L. Pan, A. Albalak, X. Wang, W. Wang, Logic-lm: Empowering large language models with symbolic solvers for faithful logical reasoning, in: Findings of the Association for Computational Linguistics: EMNLP 2023, 2023, pp. 3806–3824

    2023

  26. [34]

    Lewis, E

    P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Küttler, M. Lewis, W.-t. Yih, T. Rocktäschel, S. Riedel, D. Kiela, Retrieval-augmented generation for knowledge-intensive nlp tasks, in: H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, H. Lin (Eds.), Advan...

    2020

  27. [35]

    Karpukhin, B

    V. Karpukhin, B. Oguz, S. Min, P. Lewis, L. Wu, S. Edunov, D. Chen, W.-t. Yih, Dense passage retrieval for open-domain question answering, in: Proceedings of the 2020 conference on empirical methods in natural language processing (EMNLP), 2020, pp. 6769–6781

    2020

  28. [36]

    Thakur, N

    N. Thakur, N. Reimers, A. Rücklé, A. Srivastava, I. Gurevych, BEIR: A heterogeneous benchmark for zero-shot evaluation of information re- trieval models, in: Thirty-fifth Conference on Neural Information Pro- cessing Systems Datasets and Benchmarks Track, 2021

    2021

  29. [37]

    Sarthi, S

    P. Sarthi, S. Abdullah, A. Tuli, S. Khanna, A. Goldie, C. Manning, Raptor: Recursive abstractive processing for tree-organized retrieval, in: International Conference on Learning Representations, Vol. 2024, 2024, pp. 32628–32649

    2024

  30. [38]

    Z. Guo, L. Xia, Y. Yu, T. Ao, C. Huang, LightRAG: Simple and fast retrieval-augmented generation, in: C. Christodoulopoulos, T. Chakraborty, C. Rose, V. Peng (Eds.), Findings of the Associ- ation for Computational Linguistics: EMNLP 2025, Association for Computational Linguist...

    2025 doi

  31. [39]

    B. J. Gutiérrez, Y. Shu, Y. Gu, M. Yasunaga, Y. Su, Hipporag: Neuro- biologically inspired long-term memory for large language models, Ad- vances in neural information processing systems 37 (2024) 59532–59569

    2024

  32. [40]

    Liang, Z

    L. Liang, Z. Bo, Z. Gui, Z. Zhu, L. Zhong, P. Zhao, M. Sun, Z. Zhang, J. Zhou, W. Chen, et al., Kag: Boosting llms in professional domains via knowledge augmented generation, in: Companion Proceedings of the ACM on Web Conference 2025, 2025, pp. 334–343

    2025

  33. [41]

    Baudrillard, Simulacra and simulation, University of Michigan press, 1994

    J. Baudrillard, Simulacra and simulation, University of Michigan press, 1994

    1994

  34. [42]

    G. G. Márquez, One hundred years of solitude, in: Medicine and litera- ture, volume two, CRC Press, 2018, pp. 255–272. 22

    2018

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.