Evaluating Cumulative Spectral Gradient as a Complexity Measure

Abdul Ghani Naim; Ajaz Ahmad Bhat; Haji Gul

REVIEW 4 major objections 5 minor 14 references

Evaluating Cumulative Spectral Gradient as a Complexity Measure

T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper argues that the Cumulative Spectral Gradient, a dataset-complexity metric validated on image classification, does not scale with class count and does not predict mean reciprocal rank when applied to knowledge-graph tail prediction

desk verdict K-sensitivity result is worth a look, but the paper's own R = -0.644 contradicts its 'weak or no correlation' claim and the BERT-embedding setup makes the correlation test off-target. read the letter →

arxiv 2509.02399 v1 pith:PP3VZVCC submitted 2025-09-02 cs.LG cs.CL

Haji Gul , Abdul Ghani Naim , Ajaz Ahmad Bhat This is my paper · ORCID

classification cs.LGcs.CL

keywords datasetcomplexitycumulativespectralgradientknowledgegraphlinkpredictionmeanreciprocalrankclusteringBERTembeddingsmulti-classtailbenchmarkevaluation

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

Dataset-complexity measures would let researchers predict how well a link-prediction model will do before running it. This paper tests whether the Cumulative Spectral Gradient (CSG), a spectral-clustering metric that correlates with image-classification error, carries over to knowledge-graph tail prediction. It argues that CSG's two advertised properties fail there: values rise sharply with the nearest-neighbor/truncation parameter K, so CSG does not inherently scale with the number of classes, and CSG scores computed from BERT embeddings of (head, relation) pairs show little relation to mean reciprocal rank across six KG models. Because the paper itself reports a mean Pearson R of -0.644, the 'weak or no correlation' summary is stronger than its own number supports. If the findings hold, CSG cannot serve as a model-agnostic complexity measure for KG evaluation, and comparisons of dataset difficulty should not rely on it.

What carries the argument

The central object is the Cumulative Spectral Gradient (CSG), defined from the normalized graph Laplacian $L = I - D^{-1/2} S D^{-1/2}$ of a K-by-K class-similarity matrix S. S counts, for each sampled (head, relation) embedding in a tail class, how many of its k nearest neighbors fall in each other tail class. $CSG_{k_c} = \lambda_{k_c} - \lambda_0$ sums the first k_c eigenvalue gaps of L and is meant to quantify class overlap: larger CSG means harder separation. The paper uses this quantity as the target of a sensitivity analysis, varying K and M, and as the independent variable correlated against MRR.

What would settle it

Compute CSG on the learned entity/relation embeddings of TransE, RotatE, ConvE, TuckER, RESCAL, and ComplEx on FB15k-237 and WN18RR (or on those models' last-layer scores), and correlate with MRR across a full sweep of K and M. Near-zero correlation at every K would confirm the paper's conclusion; a recovered systematic negative correlation would show the static-BERT embedding choice, not CSG itself, produced the failure.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central finding is a controlled negative: when CSG is computed for multi-class tail prediction on FB15k-237, WN18RR, CoDEx, and smaller KG datasets—building a class similarity matrix from k-NN overlap over BERT embeddings and taking $CSG_{k_c} = \lambda_{k_c} - \lambda_0$ from the normalized graph Laplacian—the metric's value is dominated by the choice of the parameter K (and, secondarily, M) rather than by an intrinsic property of the dataset. It also reports that CSG values do not track the MRR achieved by TransE, RESCAL, RotatE, ConvE, TuckER, and ComplEx, contradicting the original claim that CSG forecasts downstream performance. The paper concludes that CSG's

Load-bearing premise

The negative result rests on treating static BERT embeddings of (head, relation) pairs as the space in which CSG should be evaluated against the MRR of trained link-prediction models; if those embeddings are not the right space, the test does not refute CSG's original claim about classifier-representation complexity.

Editorial extensions

If this is right

CSG should not be reported as a standalone complexity measure for KG link prediction without specifying and justifying K and M.
Any cross-dataset comparison that takes CSG values at face value is not interpretable unless parameter settings are matched.
The advertised property that CSG naturally scales with the number of classes does not generalize to large multi-class tail-prediction tasks.
Dataset-difficulty estimation for knowledge graphs needs an alternative classifier-agnostic measure that is stable under reasonable parameter choices.

Reading between the lines

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

The paper's own Pearson R = -0.644, reported in Section 2.2, is a moderate negative correlation, not 'no correlation'; read literally, it points the same direction as the original CSG claim (higher complexity with lower accuracy), so the qualitative conclusion depends on how that five-point correlation is interpreted.
The evaluation computes CSG on static BERT embeddings of (h,r) pairs, not on the learned embedding spaces of the six models whose MRR is correlated; a fair transfer test would compute CSG on those models' own representations, and the negative result may be an artifact of that mismatch.
The notation K is used both for the nearest-neighbor count and for the number of classes/eigenvalues entering CSG, so the reported K-sensitivity conflates two distinct design choices; disentangling them would sharpen the conclusion.
A synthetic control set with tunable class overlap and fixed class count could separate K-sensitivity from genuine class-count scaling, giving a cleaner falsifier than benchmark correlation.

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 on arXiv:2509.02399. The strongest piece is the first systematic evaluation of CSG on KG link-prediction benchmarks, and the K-sensitivity observation—CSG rises consistently as K increases across several datasets—is a legitimate caution for anyone using CSG as a default complexity measure. That part has value.

The soft spots are serious. The abstract claims CSG shows 'weak or no correlation' with MRR, but Section 2.2 reports a mean Pearson R = -0.644. That's a substantial magnitude, and the sign is the one the original CSG paper predicts (higher complexity, lower performance). So the paper's own number, if correct, partially supports the original claim rather than refuting it. No p-value or confidence interval is given, and with only five datasets in Figure 4, the effective sample is tiny—so the correlation is not well determined either way. But the wording in the abstract and the reported coefficient are an internal contradiction that undercuts the paper's main thesis.

The second problem is the methodology: CSG is computed on BERT embeddings of concatenated head-relation text, while the MRR values are from trained TransE/RotatE/ConvE/TuckER/ComplEx models with their own learned representations. Unless CSG is explicitly assumed to be representation-agnostic, this test measures the geometry of BERT, not the complexity of the models being evaluated. That's an off-target test of the original claim. Figure 4 is the only presentation; there's no table of exact CSG/MRR values, no code, no error bars, so the result isn't verifiable from the text.

There are smaller issues: K is used for both the number of classes and the nearest-neighbor count; the FB15k-237 and WN18RR references point to Freebase and WordNet rather than the actual benchmark papers. These are easy fixes but signal a need for care.

Net: the K-sensitivity finding could be a useful contribution if presented cleanly with proper statistics and a study that isolates the parameter effect. The correlation claim, as stated, is not supported by the paper's own evidence. I'd still send it to peer review because the question is legitimate and a referee could force the necessary corrections—but as it stands, the central conclusion should not be accepted.

Referee Report

4 major / 5 minor

Summary. The paper evaluates the Cumulative Spectral Gradient (CSG) complexity measure on knowledge-graph link prediction. It computes CSG from BERT embeddings of head-relation pairs, grouping tail entities as classes, building a k-NN similarity matrix, and extracting spectral gaps from the normalized graph Laplacian. The authors claim that CSG is highly sensitive to the neighbor parameter K and that CSG values show weak or no correlation with MRR across five KG benchmarks and several KG embedding models, contradicting the original CSG claims of Branchaud-Charron et al. (2019). The central evidence is Figure 4, with a reported mean Pearson coefficient R = -0.644.

Significance. If the claims were correct, the paper would provide a useful negative result: CSG would be an unstable and uninformative complexity measure for large multi-class KG link prediction. The parameter-sensitivity observation in Figures 2 and 3 could be of interest to practitioners. However, the main quantitative result is internally inconsistent with the paper's headline conclusion, and the experimental setup compares CSG on BERT text embeddings with MRR from separately trained KG embedding models. The paper also does not release code or exact numerical values. As presented, the central claims are not supported, so the significance for the field is not established.

major comments (4)

[Section 2.2, Abstract, Introduction] The paper repeatedly claims that CSG exhibits 'weak or no correlation' and 'near-zero Pearson correlation' with MRR, yet Section 2.2 reports a mean Pearson coefficient R = -0.644. For a five-dataset benchmark, |R| ≈ 0.64 is a substantial effect, and its negative sign is the direction predicted by Branchaud-Charron et al.: higher complexity corresponds to worse downstream performance. This internal contradiction directly undermines the paper's central negative claim. The authors must either correct the reported number or revise the conclusion; as written, the quantitative evidence supports the original CSG hypothesis rather than refuting it.
[Section 2, 'Generating Embeddings'] CSG is computed on BERT embeddings of concatenated head-relation text, while MRR values come from trained KG embedding models (TransE, RotatE, ConvE, etc.) with their own learned representations. Unless CSG is explicitly assumed to be representation-agnostic, this correlation test measures whether BERT embedding geometry predicts the performance of unrelated models, not whether CSG computed on a given model's own embeddings predicts that model's performance. The authors should either compute CSG on the same embeddings that produce the MRR values or provide an argument that CSG is invariant to the choice of embedding. Without this, the 'weak/no correlation' claim cannot be attributed to CSG.
[Section 2.2, Figure 4] The statistical analysis is underspecified. CSG is a single dataset-level value, while MRR is observed for each (dataset, model) pair; plotting five model MRRs against the same dataset CSG creates 25 points that are not independent. This nested structure inflates the effective sample size and makes a single pooled Pearson R difficult to interpret. The authors should report per-dataset correlations, account for the model-level grouping (e.g., a mixed-effects model), and provide exact CSG/MRR values with confidence intervals. Without this, the reported R = -0.644 cannot be verified or interpreted.
[Section 2, Eqs. (4), (10)-(11); Section 2.2, Figures 2-3] The notation for K is inconsistent. Eq. (4) defines K as the total number of unique tail entities (classes), while Eqs. (10)-(11) and Figure 3 use K (or k) for the number of nearest neighbors. This conflation makes the headline claim that 'CSG is sensitive to K' ambiguous: does K refer to the number of classes or the neighbor count? Additionally, Section 2.1 states M = 120, but Figure 2 varies M across 30-280 and Figure 3 uses M = 100. The authors must disambiguate the notation and report the exact experimental setting for each figure.

minor comments (5)

[Section 2, Eq. (9)] The definition M = min(N, |Φ(Ci)|) is confusing because N is not defined and M is later fixed to 120. Please clarify the sampling procedure and define N.
[Section 2, Eq. (19)] The symbol kc is not defined. It appears later as a summation bound, but its relationship to K (the number of classes) is unclear.
[Figures 3 and 4] Figure 3 uses 'k' on the x-axis but does not state whether this is the number of neighbors or the number of classes. Figure 4 has no axis label for CSG and contains a typo 'T ail Pre iction Mo els'. Please improve figure readability.
[References] Several datasets (Countries, Toy, UML, Nations) are cited to Liang et al. (2024), a survey, which is not the original source. The authors should cite the original dataset papers.
[General presentation] The paper contains numerous typos and awkward phrasings (e.g., 'the d embedding dimension is', 'Letter1' in the author list). A thorough proofreading pass is recommended.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the CSG evaluation is self-contained, though internally inconsistent.

full rationale

The paper does not fit any parameter to the MRR values it compares against. It computes CSG from BERT embeddings of (h,r) pairs with stated choices M=120 and k=50 (Equations 9-11), using the original CSG definition (Equations 17-20), and then plots CSG against MRR from independently trained KG models. No equation reduces the predicted correlation to the input: the CSG values are not derived from MRR, nor is any MRR-derived parameter inserted into CSG. The claims about sensitivity to K/M are empirical surface/line plots (Figures 2-3), not consequences of a definitional identity. The paper is marred by an internal inconsistency: Section 2.2 reports a mean Pearson R = -0.644 and calls this 'no meaningful correlation', which contradicts the abstract's 'weak or no correlation' claim and actually aligns in sign with the original CSG hypothesis; there is also a notational conflation of K as number of classes (Eq. 4) and K/k as nearest-neighbor count (Abstract, Eq. 10). These are correctness/reporting flaws, not circularity. Self-citations (Gul et al. 2024, 2025a, 2025b) appear only in the introduction's application examples and are not load-bearing. No uniqueness theorem is imported, and no external result is renamed. Therefore the derivation chain is self-contained and the appropriate circularity score is 0.

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

The paper introduces no new entities. Its main assumptions are the validity of the BERT-based feature space and the fidelity of the CSG implementation.

free parameters (2)

M (number of sampled vectors per class) = 120 (default)
Algorithmic hyperparameter, chosen ad hoc and varied in experiments; not fitted to the target result.
K (number of nearest neighbors) = 50 (default), varied over range
Algorithmic hyperparameter, chosen ad hoc and varied to test sensitivity; not fitted to the target result.

assumptions (3)

domain assumption CSG as defined in Eq (19)-(20) (sum of eigenvalue gaps equals lambda_{K-1}) is the metric proposed by Branchaud-Charron et al. (2019).
The paper's CSG reduces to the largest eigenvalue of the normalized Laplacian; if the original CSG includes additional weighting, the evaluation targets a different quantity.
domain assumption BERT embeddings of (h,r) pairs provide a feature space suitable for measuring dataset complexity that predicts KG link prediction performance.
The original CSG was computed on model activations; this paper uses static BERT embeddings, which may not reflect the difficulty faced by the evaluated KG embedding models.
standard math The normalized Laplacian eigenvalue spectrum (Eq 14-16) captures class separability.
Standard spectral graph theory, assumed from Branchaud-Charron et al.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Evaluating Cumulative Spectral Gradient as a Complexity Measure." pith.science (2026). https://pith.science/paper/PP3VZVCC

@misc{pith2026250902399,
  author       = {Pith},
  title        = {Pith review of: Evaluating Cumulative Spectral Gradient as a Complexity Measure},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PP3VZVCC}},
  note         = {Machine review of arXiv:2509.02399}
}

read the original abstract

Accurate estimation of dataset complexity is crucial for evaluating and comparing link prediction models for knowledge graphs (KGs). The Cumulative Spectral Gradient (CSG) metric derived from probabilistic divergence between classes within a spectral clustering framework was proposed as a dataset complexity measure that (1) naturally scales with the number of classes and (2) correlates strongly with downstream classification performance. In this work, we rigorously assess CSG behavior on standard knowledge graph link prediction benchmarks a multi class tail prediction task, using two key parameters governing its computation, M, the number of Monte Carlo sampled points per class, and K, the number of nearest neighbors in the embedding space. Contrary to the original claims, we find that (1) CSG is highly sensitive to the choice of K and therefore does not inherently scale with the number of target classes, and (2) CSG values exhibit weak or no correlation with established performance metrics such as mean reciprocal rank (MRR). Through experiments on FB15k 237, WN18RR, and other standard datasets, we demonstrate that CSG purported stability and generalization predictive power break down in link prediction settings. Our results highlight the need for more robust, classifier agnostic complexity measures in KG link prediction evaluation.

Figures

Figures reproduced from arXiv: 2509.02399 by the authors.

**Figure 1.** Left box showing triplets where the heads are (h1, h2, . . . , hk) green, relations (r1, r2, . . . , rk), tails (t1, t2, t3) are in blue, yellow and purple. The next box denotes the grouping of their tail entities into classes: c1 for t1, with (h1, r1, t1) and, (h2, r2, t1) belonging to the same class, for example. BERT is used to embed head-relation pairs, producing 768-dimensional vectors, and then concatenates th… view at source ↗

**Figure 3.** A plot of CSG as a function of K values at M = 100. 3. Conclusion CSG is significantly influenced by parameters, K and M, challenging prior assumptions of their minimal impact on complexity assessments as well as the application of CSG as a reliable complexity metric for large multi-class datasets. References Bodenreider, O. The unified medical language system (umls): integrating biomedical terminology. Nucleic Aci… view at source ↗

**Figure 2.** illustrates the Cumulative Spectral Gradient (CSG) of Codex-S dataset in the tail-prediction task setting, represented as a surface function of parameters K and M. Both parameters influence the CSG values, the impact of K on CSG however is quite significant. This observation counters the previously held belief, as discussed in (Branchaud-Charron et al., 2019), where it was argued that these parameters had minimal … view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Relationship Between MRR from different tail-prediction models on five standard KG datasets and the corresponding CSG values. Bollacker, K., Evans, C., Paritosh, P., Sturge, T., and Taylor, J. Freebase: A collaboratively created graph database for structuring human kno…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 12 canonical work pages

[1]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...
[2]

The unified medical language system (umls): integrating biomedical terminology

Bodenreider, O. The unified medical language system (umls): integrating biomedical terminology. Nucleic Acids Research, 32 0 (Database issue): 0 D267--270, 2004. doi:10.1093/nar/gkh061. URL https://www.ncbi.nlm.nih.gov/pmc/articles/PMC308795/

doi:10.1093/nar/gkh061 2004
[3]

Freebase: A collaboratively created graph database for structuring human knowledge

Bollacker, K., Evans, C., Paritosh, P., Sturge, T., and Taylor, J. Freebase: A collaboratively created graph database for structuring human knowledge. In Proc. of ACM SIGMOD, 2008

work page 2008
[4]

and Pitoura, E

Bourli, S. and Pitoura, E. Bias in knowledge graph embeddings. In 2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pp.\ 6--10. IEEE, 2020

work page 2020
[5]

Spectral metric for dataset complexity assessment

Branchaud-Charron, F., Achkar, A., and Jodoin, P.-M. Spectral metric for dataset complexity assessment. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 3215--3224, 2019

work page 2019
[6]

A Contextualized BERT model for Knowledge Graph Completion

Gul, H., Naim, A. G., and Bhat, A. A. A contextualized bert model for knowledge graph completion. In Muslims in Machine Learning (MusIML) Workshop, Co-located with Advances in neural information processing systems-24 (NeurIPS'24), https://arxiv.org/html/2412.11016v1, volume 3, 2024

work page Pith review arXiv 2024
[7]

MuCoS: Efficient Drug Target Discovery via Multi Context Aware Sampling in Knowledge Graphs

Gul, H., Naim, A. G., and Bhat, A. A. Mucos: Efficient drug target discovery via multi context aware sampling in knowledge graphs. In 24th BioNLP workshop, Co-located with Association for Computational Linguistics (ACL-2025), https://arxiv.org/pdf/2503.08075, 2025 a

work page Pith review arXiv 2025
[8]

Gul, H., Naim, A. G. H., and Bhat, A. A. Muco-kgc: Multi-context-aware knowledge graph completion. In 29th edition of Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), https://arxiv.org/html/2503.03091v2, 2025 b

work page Pith review arXiv 2025

Show all 14 references

[9]

A survey of knowledge graph reasoning on graph types: Static, dynamic, and multi-modal

Liang, K., Meng, L., Liu, M., Liu, Y., Tu, W., Wang, S., Zhou, S., Liu, X., Sun, F., and He, K. A survey of knowledge graph reasoning on graph types: Static, dynamic, and multi-modal. IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.\ 1--20, 2024. doi:10.1109...

2024
[10]

Miller, G. A. Wordnet: A lexical database for english. Communications of the ACM, 38 0 (11), 1995

1995
[11]

and Koutra, D

Safavi, T. and Koutra, D. Codex: A comprehensive knowledge graph completion benchmark. In Proc. of EMNLP, 2020

2020
[12]

Evaluating content-based pre-training strategies for a knowledge-aware recommender system based on graph neural networks

Spillo, G., Bottalico, F., Musto, C., De Gemmis, M., Lops, P., and Semeraro, G. Evaluating content-based pre-training strategies for a knowledge-aware recommender system based on graph neural networks. In Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and ...

2024
[13]

Kosel: Knowledge subgraph enhanced large language model for medical question answering

Zeng, Z., Cheng, Q., Hu, X., Zhuang, Y., Liu, X., He, K., and Liu, Z. Kosel: Knowledge subgraph enhanced large language model for medical question answering. Knowledge-Based Systems, 309: 0 112837, 2025

2025
[14]

A comprehensive large-scale biomedical knowledge graph for ai-powered data-driven biomedical research

Zhang, Y., Sui, X., Pan, F., Yu, K., Li, K., Tian, S., Erdengasileng, A., Han, Q., Wang, W., Wang, J., et al. A comprehensive large-scale biomedical knowledge graph for ai-powered data-driven biomedical research. Nature Machine Intelligence, pp.\ 1--13, 2025

2025

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [1]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

[2] [2]

The unified medical language system (umls): integrating biomedical terminology

Bodenreider, O. The unified medical language system (umls): integrating biomedical terminology. Nucleic Acids Research, 32 0 (Database issue): 0 D267--270, 2004. doi:10.1093/nar/gkh061. URL https://www.ncbi.nlm.nih.gov/pmc/articles/PMC308795/

doi:10.1093/nar/gkh061 2004

[3] [3]

Freebase: A collaboratively created graph database for structuring human knowledge

Bollacker, K., Evans, C., Paritosh, P., Sturge, T., and Taylor, J. Freebase: A collaboratively created graph database for structuring human knowledge. In Proc. of ACM SIGMOD, 2008

work page 2008

[4] [4]

and Pitoura, E

Bourli, S. and Pitoura, E. Bias in knowledge graph embeddings. In 2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pp.\ 6--10. IEEE, 2020

work page 2020

[5] [5]

Spectral metric for dataset complexity assessment

Branchaud-Charron, F., Achkar, A., and Jodoin, P.-M. Spectral metric for dataset complexity assessment. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 3215--3224, 2019

work page 2019

[6] [6]

A Contextualized BERT model for Knowledge Graph Completion

Gul, H., Naim, A. G., and Bhat, A. A. A contextualized bert model for knowledge graph completion. In Muslims in Machine Learning (MusIML) Workshop, Co-located with Advances in neural information processing systems-24 (NeurIPS'24), https://arxiv.org/html/2412.11016v1, volume 3, 2024

work page Pith review arXiv 2024

[7] [7]

MuCoS: Efficient Drug Target Discovery via Multi Context Aware Sampling in Knowledge Graphs

Gul, H., Naim, A. G., and Bhat, A. A. Mucos: Efficient drug target discovery via multi context aware sampling in knowledge graphs. In 24th BioNLP workshop, Co-located with Association for Computational Linguistics (ACL-2025), https://arxiv.org/pdf/2503.08075, 2025 a

work page Pith review arXiv 2025

[8] [8]

Gul, H., Naim, A. G. H., and Bhat, A. A. Muco-kgc: Multi-context-aware knowledge graph completion. In 29th edition of Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), https://arxiv.org/html/2503.03091v2, 2025 b

work page Pith review arXiv 2025