From Static Bibliometrics to Dynamic Knowledge Graphs: An LLM-Powered Framework for Modernizing Science, Technology, and Innovation (STI) Analytics

Muhsen Hammoud

REVIEW 4 major objections 3 minor 15 references

From Static Bibliometrics to Dynamic Knowledge Graphs: An LLM-Powered Framework for Modernizing Science, Technology, and Innovation (STI) Analytics

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

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

Pith's one-line read A framework that lets LLMs propose, not decide, what goes into science knowledge graphs, with multi-layer validation as the gate.

desk verdict A well-organized conceptual architecture for combining bibliometrics, dynamic knowledge graphs, and LLMs under validation discipline; honest about being unempirical, so it is a roadmap, not evidence that the approach works. read the letter →

arxiv 2607.21327 v1 pith:BOWNBWJX submitted 2026-07-23 cs.DL cs.AI

Muhsen Hammoud This is my paper · ORCID

classification cs.DLcs.AI

keywords STIanalyticsdynamicknowledgegraphslargelanguagemodelsbibliometricsvalidationfunnelprovenancescienceofsemanticaugmentation

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

This paper argues that modernizing science, technology, and innovation (STI) analytics requires more than better indicators or more powerful AI alone. It proposes a five-layer architecture that combines open bibliographic data, a versioned dynamic knowledge graph, constrained LLM-based semantic augmentation, and a multi-layer validation pipeline. The central claim is that LLMs should only generate candidate relations and labels, which become analytically admissible only after passing structural, evidentiary, comparative, and expert validation. The paper's contribution is theoretical and architectural: it positions validation as the mediating principle that lets STI analytics gain semantic richness and temporal responsiveness without sacrificing the evidentiary discipline of traditional scientometrics.

What carries the argument

The multi-layer validation funnel (Figure 5) is the load-bearing mechanism. It turns LLM outputs from probabilistic guesses into analytically admissible enrichments by routing every candidate triple through structural, evidentiary, comparative, and selective human-expert checks. Each accepted enrichment carries provenance metadata—source document, extraction method, timestamp, validation status, confidence—so the dynamic knowledge graph remains reconstructable at any past version (as-of query semantics).

What would settle it

Run the proposed validation funnel on a benchmark corpus (e.g., a sample of OpenAlex abstracts with known method–concept relations) and measure the acceptance rate and precision of the surviving enrichments. If the funnel admits a high rate of fabricated relations that pass structural and evidentiary checks, or if it rejects most genuinely novel relations, the central promise of 'semantic richness with epistemic discipline' fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that a credible modernization of STI analytics must integrate three traditions under a specific epistemic hierarchy: bibliometric baselines as validated indicators, dynamic knowledge graphs as relational and temporally versioned representations, and large language models as constrained generators of provisional semantic candidates. The key operational move is the validation funnel: structural checks against schema and ontology, evidentiary checks against source text and corroboration, comparative checks against established bibliometric signals, and selective expert review. Candidates that pass become versioned graph facts with full provenance; the rest are discar

Load-bearing premise

The framework assumes that the validation funnel can, in practice, filter out LLM hallucinations while preserving enough useful novel relations to justify the added cost and complexity.

Editorial extensions

If this is right

If the framework is adopted, STI analytics can detect emerging research themes and problem–method combinations before they accrue citations, reducing the temporal lag of conventional bibliometrics.
Science-to-technology translation pathways become traceable as typed edges (ENABLES_APPLICATION, USES_METHOD) with evidence spans, enabling policy-oriented gap analysis.
The distinction between baseline metadata, LLM candidates, validated enrichments, and derived analytics gives auditability that purely generative pipelines lack.
Versioned graph states make analytical results reproducible: any graph-derived trend or cluster can be replayed and inspected as of its source version.

Reading between the lines

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

The framework's success hinges on an empirical quantity it does not report: the acceptance rate of LLM candidates after validation. A useful extension would publish precision/recall curves across the four validation layers for different relation types and corpus genres.
The comparative validation layer implies a testable hypothesis: that validated semantic enrichments will correlate with—but lead—established bibliometric signals. An implementation could measure the lead time between a USES_METHOD edge appearing in the graph and a later citation burst.
A possible weak point the author leaves implicit is that expert review itself is a bottleneck and a source of subjectivity; a concrete extension would measure inter-rater agreement and the cost of expert time per retained enrichment.

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.

Desk Editor's Note

read the letter

Quick take: this is a serious, clearly written framework paper, but it is conceptual by its own admission. Don't let it be cited as a validated method. Do let it be read as a well-structured template for building such a system.

What is new: the assembly is more disciplined than most LLM-KG proposals. The five layers are coherent, and the symbolic-first, LLM-as-candidate-generator ordering is a genuinely useful positioning. The paper does real work in situating itself against KARMA, Tsaneva, ATOM, and other recent systems, and the comparison table is honest. It also gives a sensible four-dimensional validation agenda (extraction, structural, analytical, operational). The citation pattern looks clean and appropriate; no self-citation or obvious gaps.

Soft spots: the central thesis, that this hybrid path enables reliable modernization of STI analytics, is asserted rather than demonstrated. The validation funnel is the lynchpin, and the paper does not report acceptance rates, precision/recall, inter-rater reliability, or cost. Section 5.6 explicitly calls this a roadmap for future empirical studies, and the conclusion says the paper is intentionally conceptual. So this is an openly acknowledged limitation, not a hidden one. But it is still load-bearing: without at least a pilot study or a simulation of the funnel, reliable is not established. The stress-test note is fair; I do not think it overreaches. There are also some loose conceptual terms like semantic velocity that get defined only by analogy.

Who it is for: researchers building LLM-augmented scholarly knowledge graphs or STI analytics pipelines; also useful as a checklist for anyone tempted to put LLMs into science analytics without validation. It deserves serious peer review because the architecture is plausible, the literature coverage is strong, and the framework will be easier to build on than to reinvent. A referee should ask for a concrete pilot implementation, even small-scale, before the paper is seen as evidence of feasibility. But the paper should not be desk-rejected for being conceptual; the scope is stated clearly.

Referee Report

4 major / 3 minor

Summary. The paper proposes a five-layer conceptual framework for modern STI analytics: an open scholarly data backbone, a dynamic versioned knowledge graph, a constrained LLM-assisted semantic augmentation layer, a multi-layer validation pipeline, and an analytics layer. The central thesis is that LLMs should serve only as generators of provisional candidate enrichments, and that structural, evidentiary, comparative, and selective expert validation is what makes semantic augmentation analytically admissible. The paper positions itself as a synthesis of bibliometric baselines, knowledge-graph structure, and LLM capabilities, with provenance, versioning, and temporal decay as key governance mechanisms. It is explicitly conceptual: Section 5 presents a validation strategy and a staged future evaluation roadmap rather than empirical results.

Significance. The paper addresses a genuine and timely gap: the absence of a principled methodology for combining LLM-based semantic extraction with the symbolic rigor, temporal expressiveness, and auditability of dynamic knowledge graphs in STI analytics. Its strengths are conceptual clarity and methodological honesty: it clearly separates generated hypotheses from validated facts, insists on provenance and versioning, names concrete failure modes (hallucination, corpus bias, opacity), and provides a structured evaluation matrix in Figure 7 and a staged agenda in Section 5.6. If the framework performs as intended, it could provide a useful template for policy-relevant, semantically enriched STI analytics. However, the paper ships no implementation, code, data, or experiments, and its central claim that the hybrid path 'enables reliable modernization' is not yet supported by evidence. The value is as a framework proposal, not as a demonstrated system.

major comments (4)

[§1 Thesis; §5; §7] The Thesis states that the hybrid path 'enables reliable modernization of STI analytics,' but Section 7 explicitly disclaims a fully implemented or empirically validated system, and Section 5 is framed as a 'roadmap for future empirical studies' rather than as completed evaluation. The term 'reliable' is load-bearing for the paper's contribution and is not demonstrated by any data, baseline comparison, or prototype. I recommend reframing the Thesis as a testable design claim or conditional hypothesis, and/or adding a small proof-of-concept pilot on one STI domain reporting extraction precision/recall, acceptance rates, and at least one analytical task comparison.
[§4.4, Figure 5] The validation funnel is asserted to be effective, but no operating point is given. The paper reports no acceptance rates, precision/recall figures, inter-rater reliability for expert review, or cost/throughput estimates. Section 4.4 states that substantial attrition is 'a design feature,' yet without a quantified trade-off it is unclear whether the funnel filters hallucinations or destroys recall of novel relations. Since the framework's reliability claim rests on this funnel, the manuscript should either provide a minimal empirical characterization (even on a small annotated corpus) or explicitly discuss the expected operating range and the failure modes at both extremes.
[§4.4.3] Comparative validation compares new graph relations against established bibliometric or network signals and flags candidates that deviate strongly from baselines. This creates a potential hindsight bias: genuinely novel relations that have no prior citation or co-occurrence signal would be systematically penalized. The paper does not discuss this tension or specify how comparative validation avoids suppressing the very emergence signals the framework aims to detect. Please clarify whether this layer is non-blocking, optional, or accompanied by a mechanism to preserve weak-baseline candidates with strong textual evidence.
[§4.6] The framework relies on temporal decay functions with 'domain-specific half-lives' as part of its time-aware ranking, but provides no guidance on selecting or validating these half-lives. Since temporal responsiveness is a core claimed advantage over static bibliometrics, the absence of any sensitivity-analysis design or principled default leaves an important free parameter unconstrained. At minimum, the paper should state how half-lives would be estimated from data or set by expert judgment, and how decay interacts with versioning and reproducibility.

minor comments (3)

[§2.3, Reference [11]] Reference [11] lists only 'Bian, H.' but the text cites 'Bian et al., 2025.' This is inconsistent; please correct the reference or citation.
[§4.3, Module B] The JSON output for conceptual cluster labeling includes a 'confidence' score (e.g., 0.88) for a synthesized label and description. The meaning of this score is unclear; it is not a relation-after-validation confidence and should be defined or removed.
[§5.1] The extraction validity section calls for expert-annotated gold-standard corpora but does not specify how annotation disagreements are resolved or whether annotation guidelines will be made available. Adding a brief note on annotation protocol would strengthen the roadmap.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a conceptual framework proposal with an explicit, self-contained validation roadmap and no fitted-input predictions or self-citation chains.

full rationale

The paper proposes a conceptual five-layer architecture and explicitly disclaims empirical implementation (Section 7: 'The paper has remained intentionally conceptual. It does not claim to provide a fully implemented or empirically validated system.'). The derivation chain it offers is architectural: bibliometric baselines + dynamic KGs + constrained LLM augmentation + multi-layer validation. None of the claimed outputs is a numerical prediction fitted from a parameter that is then renamed as an output. The validation section (Section 5) is framed as a 'roadmap for future empirical studies' (Section 5.6) and lists metrics (precision, recall, acceptance rates, lead time) appropriate for later empirical testing, but no results are reported that could be retrofitted to inputs. There are no load-bearing self-citations: the single author cites external work throughout, and no central claim depends on the author's own prior results. The framework's notion of 'analytically admissible' is defined by its own validation layers, but this is a definitional design choice for a proposed system, not a hidden reduction: the paper does not claim to have measured that the funnel rejects hallucinations at a specific rate. The absence of empirical validation is an evidence gap, not circularity. Score 0.

Assumptions & free parameters 4 free parameters · 7 assumptions · 0 invented entities

As a conceptual framework paper, there are no fitted constants or empirical calibrations. The ledger instead records the hand-chosen or unspecified settings and domain assumptions that a real implementation would need: validation thresholds, decay half-lives, a hand-authored relation ontology, and LLM configuration. The axioms capture the framework's reliance on open-data adequacy, the effectiveness of validation at scale, expert reliability, and the stipulated status of validation as the mediating principle. No new physical or theoretical entities are introduced.

free parameters (4)

Validation acceptance thresholds (confidence cutoffs, corroboration counts, expert-routing criteria)
Section 4.3 shows example confidence values (0.92, 0.88, 0.76) and Section 4.4 routes low-confidence candidates to experts, but no numeric acceptance thresholds are specified. Any implementation would have to choose or fit these values, directly deciding how much LLM output enters the graph.
Decay half-lives for edge-weight decay functions
Section 4.6 introduces optional exponential decay with 'domain-specific half-lives' for time-aware ranking. No half-life values are given; they would be chosen per domain and would materially affect trend and recency analyses.
Relation ontology R (closed set of relation types)
Module A in Section 4.3 restricts the LLM to a hand-selected set: USES_METHOD, ADDRESSES_PROBLEM, CHALLENGES_FINDING, ENABLES_APPLICATION, IMPROVES_UPON, CONTRASTS_WITH. This schema choice is ad hoc and shapes all downstream extractions and analytics.
LLM model and prompt configuration
The modules use 'gpt-4' as an example model, but no specific models, temperature settings, prompt versions, or reproducibility constraints are fixed. LLM choice is a practical free parameter that affects candidate quality and stability.

assumptions (7)

domain assumption LLM outputs are probabilistic and unreliable and therefore must be treated as provisional candidates rather than accepted facts.
Relied on throughout Sections 3.3-3.4 and 4.3; supported by cited hallucination literature, but treated as a premise for why validation is needed rather than as a measured property of the specific pipeline.
domain assumption Open scholarly infrastructures such as OpenAlex provide sufficient coverage, quality, and reproducibility to serve as the backbone for STI analytics.
Section 4.1 selects OpenAlex as the starting point; Section 6.5 later acknowledges coverage and bias asymmetries, but the framework requires the backbone to be an adequate 'initial condition for analytical accountability' without quantifying that adequacy.
domain assumption Multi-layer validation (structural, evidentiary, comparative, expert) can filter unreliable LLM candidates at scale while preserving enough recall to be analytically useful.
This is the load-bearing operational premise of Sections 4.4 and 5. The paper provides no acceptance rates, precision/recall measurements, or expert-reliability data, so the effectiveness of the funnel is assumed.
standard math A versioned property graph with provenance metadata supports as-of queries and historical reconstruction as described.
Section 4.2 defines G = (V, E, τ, ρ, λ, θ) and Section 4.6 assumes versioning and provenance replay. This is standard graph-engineering background, not introduced by the paper.
domain assumption Domain experts are available, timely, and reliable for selective oversight of high-impact or uncertain candidates.
Section 4.4 (selective human oversight) routes candidates to experts, and Section 5.5 calls for expert Likert-scale evaluation. Expert availability, inter-rater reliability, and expert bias are not analyzed.
ad hoc to paper Validation is the necessary mediator between semantic flexibility and epistemic discipline.
Central theoretical thesis of Section 3.4. It is stipulated as the framework's organizing principle rather than derived or empirically tested; it defines the paper's notion of 'analytically admissible'.
ad hoc to paper Decay functions with domain-specific half-lives approximate the changing relevance of relations over time.
Introduced in Section 4.6 as an optional mechanism with no calibration data or comparison to alternative temporal-weighting schemes.

how reviews work

0 comments

Cite this review

Pith. "Pith review of From Static Bibliometrics to Dynamic Knowledge Graphs: An LLM-Powered Framework for Modernizing Science, Technology, and Innovation (STI) Analytics." pith.science (2026). https://pith.science/paper/BOWNBWJX

@misc{pith2026260721327,
  author       = {Pith},
  title        = {Pith review of: From Static Bibliometrics to Dynamic Knowledge Graphs: An LLM-Powered Framework for Modernizing Science, Technology, and Innovation (STI) Analytics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BOWNBWJX}},
  note         = {Machine review of arXiv:2607.21327}
}

read the original abstract

Bibliometric indicators - citation counts, h-indexes, co-authorship networks - have long anchored science, technology, and innovation (STI) analytics, yet suffer from temporal lag, semantic shallowness, and an inability to capture the non-linear dynamics of contemporary knowledge ecosystems. Dynamic knowledge graphs and large language models (LLMs) have each been proposed as remedies, but neither is sufficient alone: existing scholarly knowledge graphs remain largely static, while LLM-driven pipelines are prone to hallucination, opacity, and corpus bias without structured grounding. This paper proposes a hybrid, symbolic-first framework integrating all three traditions under explicit methodological constraint. Organized across five layers - an open scholarly data backbone, a dynamic versioned knowledge graph, a constrained LLM-assisted semantic augmentation layer, a multi-layer validation pipeline, and an analytics layer - the framework positions LLMs strictly as generators of provisional candidate enrichments. Candidates become analytically admissible only after passing structural, evidentiary, comparative, and selective expert validation, with full provenance recorded at every stage. The analytics layer supports both established bibliometric indicators and extended graph-based analyses, including trend emergence detection, science-to-technology pathway mapping, and policy-oriented gap analysis. The framework's central theoretical contribution is treating validation as the mediating principle between semantic flexibility and epistemic discipline, enabling STI analytics that is semantically richer and temporally more responsive than static bibliometrics while remaining aligned with the evidentiary standards of science-of-science research. Governance considerations addressing reproducibility, bias, and auditability are also discussed.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 5 linked inside Pith

[1]

inside out

017; Devlin et al., 2019), culminating in frontier LLMs as of 2025–2026 (e.g., successors to GPT-4, Llama, and Qwen series), has fundamentally altered the landscape of natural language processing. LLMs exhibit remarkable capabilities in zero-shot and few-shot semantic understanding, text generation, and reasoning, leading to their rapid adoption across sc...

2019
[2]

Coccia, M

[15]. Coccia, M. (2018). A theory of the general causes of long waves: War, general purpose technologies, and economic change. Technological Forecasting and Social Change, 128, 287–295. [16]. Culbert, J., et al. (2024). Reference coverage analysis of OpenAlex compared to Web of Science and Scopus. arXiv:2401.16359 / Scientometrics. [17]. De Bellis, N. (20...

arXiv 2018
[3]

Traditional bibliometric approaches remain indispensable because they provide standardized, reproducible, and widely interpretable indicators of scientific activity

Theoretical Foundations The central premise of this paper is that contemporary STI analytics requires a representational and methodological shift. Traditional bibliometric approaches remain indispensable because they provide standardized, reproducible, and widely interpretable indicators of scientific activity. At the same time, their capacity to represen...

2010
[4]

NONE” is a valid output. Output structure: json {

A Framework for LLM-Assisted Dynamic Knowledge Graph Analytics in STI Building on the theoretical position developed above, this section presents the proposed framework for modernizing STI analytics. The framework is designed around a simple methodological principle: semantic expansion should take place under structural and evidentiary constraint. Its pur...

2022
[7]

Relations with high analytical impact, unusual novelty, low confidence from earlier layers, or potential policy relevance are routed to domain experts

Selective human oversight. Relations with high analytical impact, unusual novelty, low confidence from earlier layers, or potential policy relevance are routed to domain experts. A purpose-built interface presents the candidate, its evidence, and the outputs of automated checks; experts accept, reject, or modify the candidate, and their decisions are reco...

2024
[8]

Conclusion This paper has argued that the modernization of science, technology, and innovation (STI) analytics requires more than incremental refinement of existing bibliometric indicators. While traditional bibliometric and scientometric approaches remain indispensable for standardized, reproducible, and institutionally legible evaluation, they are limit...

arXiv 2024
[30]

Wadden, D., Wennberg, U., Luan, Y., & Hajishirzi, H

[70]. Wadden, D., Wennberg, U., Luan, Y., & Hajishirzi, H. (2019, November). Entity, relation, and event extraction with contextualized span representations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (pp. 5784–5789). [71]. Wang, J., & ...

2019
[260]

[39]. Koh, H. Y., Ju, J., Liu, M., & Pan, S. (2022). An empirical survey on long document summarization: Datasets, models, and metrics. ACM Computing Surveys, 55(8), 1–35. [40]. Lairgi, Y., Moncla, L., Benabdeslem, K., Cazabet, R., & Cléau, P. (2026, March). ATOM: AdapTive and OptiMized dynamic temporal knowledge graph construction using LLMs. In Findings...

arXiv 2022

Show all 15 references

[462]

Bender, E

[10]. Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (pp. 610–623). [11]. Bian, H. (2025). LLM-...

2021
[964]

Devlin, J., Chang, M

[20]. Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019, June). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human ...

2019 arXiv
[1260]

[50]. Moed, H. F. (2005). Citation analysis in research evaluation. Dordrecht: Springer Netherlands. [51]. Newman, M. E. (2001). The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences, 98(2), 404–409. [52]. OECD. (2024). OECD agenda...

2005 arXiv
[1849]

Bai, J., Mosbach, S., Taylor, C

[9]. Bai, J., Mosbach, S., Taylor, C. J., Karan, D., Lee, K. F., Rihm, S. D., ... & Kraft, M. (2024). A dynamic knowledge graph approach to distributed self-driving laboratories. Nature Communications, 15(1),

2024
[2020]

Closer to our focus, recent work has explored using LLMs to populate knowledge graphs, either by generating triples directly (Ye et al.,

is now a standard technique, and its application to scientific question answering is active (Ahrabian et al., 2023). Closer to our focus, recent work has explored using LLMs to populate knowledge graphs, either by generating triples directly (Ye et al.,

2023
[2022]

Multi-agent frameworks like KARMA (Lu et al.,

or by aligning textual mentions with KG entities (Wang et al., 2021). Multi-agent frameworks like KARMA (Lu et al.,

2021
[2025]

employ collaborative agents for entity discovery, relation extraction, schema alignment, and conflict resolution, achieving ~83% verified correctness on PubMed data while reducing conflicts via debate mechanisms. Neurosymbolic hybrids integrate LLMs with human-in-the-loop or s...

2025

Pith tools

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

[1] [1]

inside out

017; Devlin et al., 2019), culminating in frontier LLMs as of 2025–2026 (e.g., successors to GPT-4, Llama, and Qwen series), has fundamentally altered the landscape of natural language processing. LLMs exhibit remarkable capabilities in zero-shot and few-shot semantic understanding, text generation, and reasoning, leading to their rapid adoption across sc...

2019

[2] [2]

Coccia, M

[15]. Coccia, M. (2018). A theory of the general causes of long waves: War, general purpose technologies, and economic change. Technological Forecasting and Social Change, 128, 287–295. [16]. Culbert, J., et al. (2024). Reference coverage analysis of OpenAlex compared to Web of Science and Scopus. arXiv:2401.16359 / Scientometrics. [17]. De Bellis, N. (20...

arXiv 2018

[3] [3]

Traditional bibliometric approaches remain indispensable because they provide standardized, reproducible, and widely interpretable indicators of scientific activity

Theoretical Foundations The central premise of this paper is that contemporary STI analytics requires a representational and methodological shift. Traditional bibliometric approaches remain indispensable because they provide standardized, reproducible, and widely interpretable indicators of scientific activity. At the same time, their capacity to represen...

2010

[4] [4]

NONE” is a valid output. Output structure: json {

A Framework for LLM-Assisted Dynamic Knowledge Graph Analytics in STI Building on the theoretical position developed above, this section presents the proposed framework for modernizing STI analytics. The framework is designed around a simple methodological principle: semantic expansion should take place under structural and evidentiary constraint. Its pur...

2022

[5] [7]

Relations with high analytical impact, unusual novelty, low confidence from earlier layers, or potential policy relevance are routed to domain experts

Selective human oversight. Relations with high analytical impact, unusual novelty, low confidence from earlier layers, or potential policy relevance are routed to domain experts. A purpose-built interface presents the candidate, its evidence, and the outputs of automated checks; experts accept, reject, or modify the candidate, and their decisions are reco...

2024

[6] [8]

Conclusion This paper has argued that the modernization of science, technology, and innovation (STI) analytics requires more than incremental refinement of existing bibliometric indicators. While traditional bibliometric and scientometric approaches remain indispensable for standardized, reproducible, and institutionally legible evaluation, they are limit...

arXiv 2024

[7] [30]

Wadden, D., Wennberg, U., Luan, Y., & Hajishirzi, H

[70]. Wadden, D., Wennberg, U., Luan, Y., & Hajishirzi, H. (2019, November). Entity, relation, and event extraction with contextualized span representations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (pp. 5784–5789). [71]. Wang, J., & ...

2019

[8] [260]

[39]. Koh, H. Y., Ju, J., Liu, M., & Pan, S. (2022). An empirical survey on long document summarization: Datasets, models, and metrics. ACM Computing Surveys, 55(8), 1–35. [40]. Lairgi, Y., Moncla, L., Benabdeslem, K., Cazabet, R., & Cléau, P. (2026, March). ATOM: AdapTive and OptiMized dynamic temporal knowledge graph construction using LLMs. In Findings...

arXiv 2022