61 Search Results for "Liu, Allen"

Document
DOI: 10.4230/LIPIcs.FORC.2026.7
When to Ask a Question: Understanding Communication Strategies in Generative AI Tools

Authors: Charlotte Park, Kate Donahue, and Manish Raghavan

Published in: LIPIcs, Volume 368, 7th Symposium on Foundations of Responsible Computing (FORC 2026)

Abstract
Generative AI models differ from traditional machine learning tools in that they allow users to provide as much or as little information as they choose in their inputs. This flexibility often leads users to omit certain details, relying on the models to infer and fill in under-specified information based on distributional knowledge of user preferences. Such inferences may privilege majority viewpoints and disadvantage users with atypical preferences, raising concerns about fairness. Unlike more traditional recommender systems, LLMs can explicitly solicit more information from users through natural language. However, while directly eliciting user preferences could increase personalization and mitigate inequality, excessive querying places a burden on users who value efficiency. We develop a stylized model of user-LLM interaction and develop an objective that captures tradeoff between user burden and preference representation. Building on the observation that individual preferences are often correlated, we analyze how AI systems should balance inference and elicitation, characterizing the optimal amount of information to solicit before content generation. Ultimately, we show that information elicitation can mitigate the systematic biases of preference inference, enabling the design of generative tools that better incorporate diverse user perspectives while maintaining efficiency. We complement this theoretical analysis with an empirical evaluation illustrating the model’s predictions and exploring their practical implications.
Cite as

Charlotte Park, Kate Donahue, and Manish Raghavan. When to Ask a Question: Understanding Communication Strategies in Generative AI Tools. In 7th Symposium on Foundations of Responsible Computing (FORC 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 368, pp. 7:1-7:25, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{park_et_al:LIPIcs.FORC.2026.7,
  author =	{Park, Charlotte and Donahue, Kate and Raghavan, Manish},
  title =	{{When to Ask a Question: Understanding Communication Strategies in Generative AI Tools}},
  booktitle =	{7th Symposium on Foundations of Responsible Computing (FORC 2026)},
  pages =	{7:1--7:25},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-419-2},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{368},
  editor =	{Lin, Huijia (Rachel)},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.FORC.2026.7},
  URN =		{urn:nbn:de:0030-drops-259782},
  doi =		{10.4230/LIPIcs.FORC.2026.7},
  annote =	{Keywords: human-AI interaction, user modeling, personalization}
}
Document
Research
DOI: 10.4230/TGDK.4.1.1
On the Computational Cost of Knowledge Graph Embeddings

Authors: Victor Charpenay, Mansour Zoubeirou A Mayaki, and Antoine Zimmermann

Published in: TGDK, Volume 4, Issue 1 (2026). Transactions on Graph Data and Knowledge, Volume 4, Issue 1

Abstract
Over a decade, numerous Knowledge Graph Embedding (KGE) models have been designed and evaluated on reference datasets, always with increasing performance. In this paper, we re-evaluate these models with respect to their computational efficiency during training, by estimating the computational cost of the procedure expressed in floating-point operations. We design a cost model based on analytical expressions and apply it on a collection of 20 KGE models, representative of the state-of-the-art. We show that dimensionality or parameter efficiency, used in the literature to compare models with each other, are not suitable to evaluate the true cost of models. Through fixed-budget experiments, a novel approach to evaluate KGE models based on cost estimates, we re-assess the relative performance of model families compared to the state-of-the-art. Bilinear models such as ComplEx underperform with a low computational budget while hyperbolic linear models appear to offer no particular benefit compared to simpler Euclidian models, especially the MuRE model. Neural models, such as ConvE or CompGCN, achieve reasonable performance in the literature but their high computational cost appears unnecessary when compared with other models. The trade-off between efficiency and expressivity of both linear and neural models is to be further explored.
Cite as

Victor Charpenay, Mansour Zoubeirou A Mayaki, and Antoine Zimmermann. On the Computational Cost of Knowledge Graph Embeddings. In Transactions on Graph Data and Knowledge (TGDK), Volume 4, Issue 1, pp. 1:1-1:30, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@Article{charpenay_et_al:TGDK.4.1.1,
  author =	{Charpenay, Victor and Zoubeirou A Mayaki, Mansour and Zimmermann, Antoine},
  title =	{{On the Computational Cost of Knowledge Graph Embeddings}},
  journal =	{Transactions on Graph Data and Knowledge},
  pages =	{1:1--1:30},
  ISSN =	{2942-7517},
  year =	{2026},
  volume =	{4},
  number =	{1},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/TGDK.4.1.1},
  URN =		{urn:nbn:de:0030-drops-256863},
  doi =		{10.4230/TGDK.4.1.1},
  annote =	{Keywords: Knowledge Graph Embedding, Parameter Efficiency, Computational Budget, Green AI}
}
Document
Survey
DOI: 10.4230/TGDK.4.1.2
Temporal Modelling in Cultural Heritage Knowledge Graphs: Use Cases, Requirements, Evaluation, and Decision Support

Authors: Oleksandra Bruns, Jörg Waitelonis, Jeff Z. Pan, and Harald Sack

Published in: TGDK, Volume 4, Issue 1 (2026). Transactions on Graph Data and Knowledge, Volume 4, Issue 1

Abstract
Our culture, history and world are in constant motion, continuously shaped by the flow of time, evolving narratives, and shifting relationships. Capturing this temporal complexity within cultural heritage (CH) knowledge graphs is essential for preserving the dynamic nature of human heritage. However, standard RDF predicates fail to effectively model the temporal aspects of cultural data, such as changing facts, evolving relationships, and temporal concepts. Over the past two decades, a variety of RDF-based approaches have been proposed to address this limitation, yet guidance is missing on which method best suits specific CH contexts. This paper presents a systematic evaluation of temporal RDF modelling approaches from a CH perspective. Based on an analysis of real-world CH use cases, core temporal requirements are identified that reflect both modelling expressivity and practical concerns. Six prominent approaches - RDF*, tRDF, Named Graphs, Singleton Property, N-ary Relations, and 4D Fluents - are assessed across these requirements. Our findings reveal that no single solution fits all scenarios, but suitable approaches can be selected based on project-specific priorities. To support practitioners, a decision-support tool is introduced to guide them in selecting the most suitable extension for their specific needs. This work provides practical guidance for CH modelling and contributes to the broader development of temporally aware Linked Data.
Cite as

Oleksandra Bruns, Jörg Waitelonis, Jeff Z. Pan, and Harald Sack. Temporal Modelling in Cultural Heritage Knowledge Graphs: Use Cases, Requirements, Evaluation, and Decision Support. In Transactions on Graph Data and Knowledge (TGDK), Volume 4, Issue 1, pp. 2:1-2:46, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@Article{bruns_et_al:TGDK.4.1.2,
  author =	{Bruns, Oleksandra and Waitelonis, J\"{o}rg and Pan, Jeff Z. and Sack, Harald},
  title =	{{Temporal Modelling in Cultural Heritage Knowledge Graphs: Use Cases, Requirements, Evaluation, and Decision Support}},
  journal =	{Transactions on Graph Data and Knowledge},
  pages =	{2:1--2:46},
  ISSN =	{2942-7517},
  year =	{2026},
  volume =	{4},
  number =	{1},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/TGDK.4.1.2},
  URN =		{urn:nbn:de:0030-drops-256871},
  doi =		{10.4230/TGDK.4.1.2},
  annote =	{Keywords: Temporal Data Representation, RDF Extensions, Cultural Heritage, Knowledge Graphs}
}
Document
DOI: 10.4230/LIPIcs.ICDT.2026.5
Rule Rewriting Revisited: A Fresh Look at Static Filtering for Datalog and ASP

Authors: Philipp Hanisch and Markus Krötzsch

Published in: LIPIcs, Volume 365, 29th International Conference on Database Theory (ICDT 2026)

Abstract
Static filtering is a data-independent optimisation method for Datalog, which generalises algebraic query rewriting techniques from relational databases. In spite of its early discovery by Kifer and Lozinskii in 1986, the method has been overlooked in recent research and system development, and special cases are being rediscovered independently. We therefore recall the original approach, using updated terminology and more general filter predicates that capture features of modern systems, and we show how to extend its applicability to answer set programming (ASP). The outcome is strictly more general but also more complex than the classical approach: double exponential in general and single exponential even for predicates of bounded arity. As a solution, we propose tractable approximations of the algorithm that can still yield much improved logic programs in typical cases, e.g., it can improve the performance of rule systems over real-world data in the order of magnitude.
Cite as

Philipp Hanisch and Markus Krötzsch. Rule Rewriting Revisited: A Fresh Look at Static Filtering for Datalog and ASP. In 29th International Conference on Database Theory (ICDT 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 365, pp. 5:1-5:19, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{hanisch_et_al:LIPIcs.ICDT.2026.5,
  author =	{Hanisch, Philipp and Kr\"{o}tzsch, Markus},
  title =	{{Rule Rewriting Revisited: A Fresh Look at Static Filtering for Datalog and ASP}},
  booktitle =	{29th International Conference on Database Theory (ICDT 2026)},
  pages =	{5:1--5:19},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-413-0},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{365},
  editor =	{ten Cate, Balder and Funk, Maurice},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ICDT.2026.5},
  URN =		{urn:nbn:de:0030-drops-256197},
  doi =		{10.4230/LIPIcs.ICDT.2026.5},
  annote =	{Keywords: Rule rewriting, static optimisation, static filtering, Datalog, Answer Set Programming}
}
Document
DOI: 10.4230/LIPIcs.STACS.2026.28
Conditional Complexity Hardness: Monotone Circuit Size, Matrix Rigidity, and Tensor Rank

Authors: Nikolai Chukhin, Alexander S. Kulikov, Ivan Mihajlin, and Arina Smirnova

Published in: LIPIcs, Volume 364, 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)

Abstract
Proving complexity lower bounds remains a challenging task: currently, we only know how to prove conditional uniform (algorithm) lower bounds and nonuniform (circuit) lower bounds in restricted circuit models. About a decade ago, Williams (STOC 2010) showed how to derive nonuniform lower bounds from uniform upper bounds: roughly, by designing a fast algorithm for checking satisfiability of circuits, one gets a lower bound for this circuit class. Since then, a number of results of this kind have been proved. For example, Jahanjou et al. (ICALP 2015) and Carmosino et al. (ITCS 2016) proved that if NSETH fails, then E^{NP} has series-parallel circuit size ω(n). One can also derive nonuniform lower bounds from nondeterministic uniform lower bounds. Perhaps the most well-known example is the Karp-Lipton theorem (STOC 1980): if Σ₂ ≠ Π₂, then NP ⊄ P/poly. Some recent examples include the following. Nederlof (STOC 2020) proved a lower bound on the matrix multiplication tensor rank under an assumption that TSP cannot be solved faster than in 2ⁿ time. Belova et al. (SODA 2024) proved that there exists an explicit polynomial family of arithmetic circuit size Ω(n^{δ}), for any δ > 0, assuming that MAX-3-SAT cannot be solved faster than in 2ⁿ nondeterministic time. Williams (FOCS 2024) proved an exponential lower bound for ETHR ∘ ETHR circuits under the Orthogonal Vectors conjecture. Whereas all the lower bounds above are proved under strong assumptions that might eventually be refuted, the revealed connections are of great interest and may still give further insights: one may be able to weaken the used assumptions or to construct generators from other fine-grained reductions. In this paper, we continue developing this line of research and show how uniform nondeterministic lower bounds can be used to construct generators of various types of combinatorial objects that are notoriously hard to analyze: Boolean functions of high circuit size, matrices of high rigidity, and tensors of high rank. Specifically, we prove the following. - If, for some ε and k, k-SAT cannot be solved in input-oblivious co-nondeterministic time O(2^{(1/2+ε)n}), then there exists a monotone Boolean function family in coNP of monotone circuit size 2^{Ω(n / log n)}. Combining this with the result above, we get win-win circuit lower bounds: either E^{NP{}} requires series-parallel circuits of size ω(n) or coNP requires monotone circuits of size 2^{Ω(n / log n)}. - If, for all ε > 0, MAX-3-SAT cannot be solved in co-nondeterministic time O(2^{(1 - ε)n}), then there exist small families of matrices with rigidity exceeding the best known constructions as well as small families of three-dimensional tensors of rank n^{1+Δ}, for some Δ > 0.
Cite as

Nikolai Chukhin, Alexander S. Kulikov, Ivan Mihajlin, and Arina Smirnova. Conditional Complexity Hardness: Monotone Circuit Size, Matrix Rigidity, and Tensor Rank. In 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 364, pp. 28:1-28:21, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{chukhin_et_al:LIPIcs.STACS.2026.28,
  author =	{Chukhin, Nikolai and Kulikov, Alexander S. and Mihajlin, Ivan and Smirnova, Arina},
  title =	{{Conditional Complexity Hardness: Monotone Circuit Size, Matrix Rigidity, and Tensor Rank}},
  booktitle =	{43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)},
  pages =	{28:1--28:21},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-412-3},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{364},
  editor =	{Mahajan, Meena and Manea, Florin and McIver, Annabelle and Thắng, Nguy\~{ê}n Kim},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.STACS.2026.28},
  URN =		{urn:nbn:de:0030-drops-255177},
  doi =		{10.4230/LIPIcs.STACS.2026.28},
  annote =	{Keywords: computational complexity, circuit complexity, lower bounds, conditional lower bounds, monotone circuits, matrix rigidity, tensor rank, arithmetic circuits, fine-grained complexity}
}
Document
DOI: 10.4230/LIPIcs.STACS.2026.38
Fully Dynamic Spectral Sparsification for Directed Hypergraphs

Authors: Sebastian Forster, Gramoz Goranci, and Ali Momeni

Published in: LIPIcs, Volume 364, 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)

Abstract
There has been a surge of interest in spectral hypergraph sparsification, a natural generalization of spectral sparsification for graphs. In this paper, we present a simple fully dynamic algorithm for maintaining spectral hypergraph sparsifiers of directed hypergraphs. Our algorithm achieves a near-optimal size of O(n² / ε ² log ⁷ m) and amortized update time of O(r² log ³ m), where n is the number of vertices, and m and r respectively upper bound the number of hyperedges and the rank of the hypergraph at any time. We also extend our approach to the parallel batch-dynamic setting, where a batch of any k hyperedge insertions or deletions can be processed with O(kr² log ³ m) amortized work and O(log ² m) depth. This constitutes the first spectral-based sparsification algorithm in this setting.
Cite as

Sebastian Forster, Gramoz Goranci, and Ali Momeni. Fully Dynamic Spectral Sparsification for Directed Hypergraphs. In 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 364, pp. 38:1-38:20, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{forster_et_al:LIPIcs.STACS.2026.38,
  author =	{Forster, Sebastian and Goranci, Gramoz and Momeni, Ali},
  title =	{{Fully Dynamic Spectral Sparsification for Directed Hypergraphs}},
  booktitle =	{43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)},
  pages =	{38:1--38:20},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-412-3},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{364},
  editor =	{Mahajan, Meena and Manea, Florin and McIver, Annabelle and Thắng, Nguy\~{ê}n Kim},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.STACS.2026.38},
  URN =		{urn:nbn:de:0030-drops-255272},
  doi =		{10.4230/LIPIcs.STACS.2026.38},
  annote =	{Keywords: Spectral sparsification, Dynamic algorithms, (Directed) hypergraphs, Data structures}
}
Document
DOI: 10.4230/LIPIcs.STACS.2026.69
Approximating q → p Norms of Non-Negative Matrices in Nearly-Linear Time

Authors: Etienne Objois and Adrian Vladu

Published in: LIPIcs, Volume 364, 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)

Abstract
We provide the first nearly-linear time algorithm for approximating 𝓁_{q → p}-norms of non-negative matrices, for q ≥ p ≥ 1. Our algorithm returns a (1-ε)-approximation to the matrix norm in time Õ(1/(q ε) ⋅ nnz(A)), where A is the input matrix, and improves upon the previous state of the art, which either proved convergence only in the limit [Boyd '74], or had very high polynomial running times [Bhaskara-Vijayraghavan, SODA '11]. Our algorithm is extremely simple, and is largely inspired from the coordinate-scaling approach used for positive linear program solvers. Our algorithm can readily be used in the [Englert-Räcke, FOCS '09] to improve the running time of constructing O(log n)-competitive 𝓁_p-oblivious routings.
Cite as

Etienne Objois and Adrian Vladu. Approximating q → p Norms of Non-Negative Matrices in Nearly-Linear Time. In 43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 364, pp. 69:1-69:19, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{objois_et_al:LIPIcs.STACS.2026.69,
  author =	{Objois, Etienne and Vladu, Adrian},
  title =	{{Approximating q → p Norms of Non-Negative Matrices in Nearly-Linear Time}},
  booktitle =	{43rd International Symposium on Theoretical Aspects of Computer Science (STACS 2026)},
  pages =	{69:1--69:19},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-412-3},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{364},
  editor =	{Mahajan, Meena and Manea, Florin and McIver, Annabelle and Thắng, Nguy\~{ê}n Kim},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.STACS.2026.69},
  URN =		{urn:nbn:de:0030-drops-255585},
  doi =		{10.4230/LIPIcs.STACS.2026.69},
  annote =	{Keywords: matrix norm, Perron-Frobenius theory, oblivious routings, input-sparsity time, lp norm}
}
Document
DOI: 10.4230/LIPIcs.ITCS.2026.30
New Bounds for Circular Trace Reconstruction

Authors: Arnav Burudgunte, Paul Valiant, and Hongao Wang

Published in: LIPIcs, Volume 362, 17th Innovations in Theoretical Computer Science Conference (ITCS 2026)

Abstract
The "trace reconstruction" problem asks, given an unknown binary string x and a channel that repeatedly returns "traces" of x with each bit randomly deleted with some probability p, how many traces are needed to recover x? There is an exponential gap between the best known upper and lower bounds for this problem. Many variants of the model have been introduced in hopes of motivating or revealing new approaches to narrow this gap. We study the variant of circular trace reconstruction introduced by Narayanan and Ren (ITCS 2021), in which traces undergo a random cyclic shift in addition to random deletions. We show an improved lower bound of Ω̃(n⁵) for circular trace reconstruction. This contrasts with the (previously) best known lower bounds of Ω̃(n³) in the circular case and Ω̃(n^{3/2}) in the linear case. Our bound shows the indistinguishability of traces from two sparse strings x,y that each have a constant number of nonzeros. Can this technique be extended significantly? How hard is it to reconstruct a sparse string x under a cyclic deletion channel? We resolve these questions by showing, using Fourier techniques, that Õ(n⁶) traces suffice for reconstructing any constant-sparse string in a circular deletion channel, in contrast to the best known upper bound of exp(Õ(n^{1/3})) for general strings in the circular deletion channel. This shows that new algorithms or new lower bounds must focus on non-constant-sparse strings.
Cite as

Arnav Burudgunte, Paul Valiant, and Hongao Wang. New Bounds for Circular Trace Reconstruction. In 17th Innovations in Theoretical Computer Science Conference (ITCS 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 362, pp. 30:1-30:23, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{burudgunte_et_al:LIPIcs.ITCS.2026.30,
  author =	{Burudgunte, Arnav and Valiant, Paul and Wang, Hongao},
  title =	{{New Bounds for Circular Trace Reconstruction}},
  booktitle =	{17th Innovations in Theoretical Computer Science Conference (ITCS 2026)},
  pages =	{30:1--30:23},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-410-9},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{362},
  editor =	{Saraf, Shubhangi},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ITCS.2026.30},
  URN =		{urn:nbn:de:0030-drops-253176},
  doi =		{10.4230/LIPIcs.ITCS.2026.30},
  annote =	{Keywords: Trace reconstruction, algorithmic statistics, Fourier analysis}
}
Document
DOI: 10.4230/DagMan.11.1.19
Conversational Agents: A Framework for Evaluation (CAFE) (Dagstuhl Perspectives Workshop 24352)

Authors: Christine Bauer, Li Chen, Nicola Ferro, Norbert Fuhr, Avishek Anand, Timo Breuer, Guglielmo Faggioli, Ophir Frieder, Hideo Joho, Jussi Karlgren, Johannes Kiesel, Bart P. Knijnenburg, Aldo Lipani, Lien Michiels, Andrea Papenmeier, Maria Soledad Pera, Mark Sanderson, Scott Sanner, Benno Stein, Johanne R. Trippas, Karin Verspoor, and Martijn C. Willemsen

Published in: Dagstuhl Manifestos, Volume 11, Issue 1 (2025)

Abstract
During the workshop, we deeply discussed what CONversational Information ACcess (CONIAC) is and its unique features, proposing a world model abstracting it, and defined the Conversational Agents Framework for Evaluation (CAFE) for the evaluation of CONIAC systems, consisting of six major components: 1) goals of the system’s stakeholders, 2) user tasks to be studied in the evaluation, 3) aspects of the users carrying out the tasks, 4) evaluation criteria to be considered, 5) evaluation methodology to be applied, and 6) measures for the quantitative criteria chosen.
Cite as

Christine Bauer, Li Chen, Nicola Ferro, Norbert Fuhr, Avishek Anand, Timo Breuer, Guglielmo Faggioli, Ophir Frieder, Hideo Joho, Jussi Karlgren, Johannes Kiesel, Bart P. Knijnenburg, Aldo Lipani, Lien Michiels, Andrea Papenmeier, Maria Soledad Pera, Mark Sanderson, Scott Sanner, Benno Stein, Johanne R. Trippas, Karin Verspoor, and Martijn C. Willemsen. Conversational Agents: A Framework for Evaluation (CAFE) (Dagstuhl Perspectives Workshop 24352). In Dagstuhl Manifestos, Volume 11, Issue 1, pp. 19-67, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{bauer_et_al:DagMan.11.1.19,
  author =	{Bauer, Christine and Chen, Li and Ferro, Nicola and Fuhr, Norbert and Anand, Avishek and Breuer, Timo and Faggioli, Guglielmo and Frieder, Ophir and Joho, Hideo and Karlgren, Jussi and Kiesel, Johannes and Knijnenburg, Bart P. and Lipani, Aldo and Michiels, Lien and Papenmeier, Andrea and Pera, Maria Soledad and Sanderson, Mark and Sanner, Scott and Stein, Benno and Trippas, Johanne R. and Verspoor, Karin and Willemsen, Martijn C.},
  title =	{{Conversational Agents: A Framework for Evaluation (CAFE) (Dagstuhl Perspectives Workshop 24352)}},
  pages =	{19--67},
  journal =	{Dagstuhl Manifestos},
  ISSN =	{2193-2433},
  year =	{2025},
  volume =	{11},
  number =	{1},
  editor =	{Bauer, Christine and Chen, Li and Ferro, Nicola and Fuhr, Norbert and Anand, Avishek and Breuer, Timo and Faggioli, Guglielmo and Frieder, Ophir and Joho, Hideo and Karlgren, Jussi and Kiesel, Johannes and Knijnenburg, Bart P. and Lipani, Aldo and Michiels, Lien and Papenmeier, Andrea and Pera, Maria Soledad and Sanderson, Mark and Sanner, Scott and Stein, Benno and Trippas, Johanne R. and Verspoor, Karin and Willemsen, Martijn C.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagMan.11.1.19},
  URN =		{urn:nbn:de:0030-drops-252722},
  doi =		{10.4230/DagMan.11.1.19},
  annote =	{Keywords: Conversational Agents, Evaluation, Information Access}
}
Document
Research
DOI: 10.4230/TGDK.3.3.4
Mining Inter-Document Argument Structures in Scientific Papers for an Argument Web

Authors: Florian Ruosch, Cristina Sarasua, and Abraham Bernstein

Published in: TGDK, Volume 3, Issue 3 (2025). Transactions on Graph Data and Knowledge, Volume 3, Issue 3

Abstract
In Argument Mining, predicting argumentative relations between texts (or spans) remains one of the most challenging aspects, even more so in the cross-document setting. This paper makes three key contributions to advance research in this domain. We first extend an existing dataset, the Sci-Arg corpus, by annotating it with explicit inter-document argumentative relations, thereby allowing arguments to be distributed over several documents forming an Argument Web; these new annotations are published using Semantic Web technologies (RDF, OWL). Second, we explore and evaluate three automated approaches for predicting these inter-document argumentative relations, establishing critical baselines on the new dataset. We find that a simple classifier based on discourse indicators with access to context outperforms neural methods. Third, we conduct a comparative analysis of these approaches for both intra- and inter-document settings, identifying statistically significant differences in results that indicate the necessity of distinguishing between these two scenarios. Our findings highlight significant challenges in this complex domain and open crucial avenues for future research on the Argument Web of Science, particularly for those interested in leveraging Semantic Web technologies and knowledge graphs to understand scholarly discourse. With this, we provide the first stepping stones in the form of a benchmark dataset, three baseline methods, and an initial analysis for a systematic exploration of this field relevant to the Web of Data and Science.
Cite as

Florian Ruosch, Cristina Sarasua, and Abraham Bernstein. Mining Inter-Document Argument Structures in Scientific Papers for an Argument Web. In Transactions on Graph Data and Knowledge (TGDK), Volume 3, Issue 3, pp. 4:1-4:33, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{ruosch_et_al:TGDK.3.3.4,
  author =	{Ruosch, Florian and Sarasua, Cristina and Bernstein, Abraham},
  title =	{{Mining Inter-Document Argument Structures in Scientific Papers for an Argument Web}},
  journal =	{Transactions on Graph Data and Knowledge},
  pages =	{4:1--4:33},
  ISSN =	{2942-7517},
  year =	{2025},
  volume =	{3},
  number =	{3},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/TGDK.3.3.4},
  URN =		{urn:nbn:de:0030-drops-252159},
  doi =		{10.4230/TGDK.3.3.4},
  annote =	{Keywords: Argument Mining, Large Language Models, Knowledge Graphs, Link Prediction}
}
Document
Invited Paper
DOI: 10.4230/OASIcs.RW.2024/2025.7
Modern Datalog: Concepts, Methods, Applications (Invited Paper)

Authors: Markus Krötzsch

Published in: OASIcs, Volume 138, Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 & RW 2025)

Abstract
Pure Datalog is arguably the most fundamental rule language, elegant and simple, but also often too limited to be useful in practice. This has motivated the introduction of many new expressive features, ranging from datatypes and related functions, over aggregates and semi-ring generalisations, to existential quantifiers and complex terms. In spite of their variety, all these approaches remain true to the nature of Datalog as a direct, pattern-based way of computing on structured data. We therefore find that a modern notion of Datalog is emerging, distinctly different from other approaches of logic programming and with its own set of related methods and applications. In this course, we introduce Datalog and its most common extensions, and explain when and how these features can be used together (which is often, but not always, safe to do). We further look at modern Datalog systems and some of their primary use cases. Hands-on work with Datalog and its extensions is done with the free Datalog engine https://knowsys.github.io/nemo-doc/. The course is accessible to all audiences and does not assume specific prior knowledge.
Cite as

Markus Krötzsch. Modern Datalog: Concepts, Methods, Applications (Invited Paper). In Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 & RW 2025). Open Access Series in Informatics (OASIcs), Volume 138, pp. 7:1-7:41, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@InProceedings{krotzsch:OASIcs.RW.2024/2025.7,
  author =	{Kr\"{o}tzsch, Markus},
  title =	{{Modern Datalog: Concepts, Methods, Applications}},
  booktitle =	{Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 \& RW 2025)},
  pages =	{7:1--7:41},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-405-5},
  ISSN =	{2190-6807},
  year =	{2025},
  volume =	{138},
  editor =	{Artale, Alessandro and Bienvenu, Meghyn and Garc{\'\i}a, Yazm{\'\i}n Ib\'{a}\~{n}ez and Murlak, Filip},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.RW.2024/2025.7},
  URN =		{urn:nbn:de:0030-drops-250524},
  doi =		{10.4230/OASIcs.RW.2024/2025.7},
  annote =	{Keywords: Datalog, query language, knowlegde representation and reasoning, logic programming, Horn logic, SPARQL, datatypes and aggregation, lecture notes, tutorial}
}
Document
Invited Paper
DOI: 10.4230/OASIcs.RW.2024/2025.1
Rule-Based Knowledge Graph Completion (Invited Paper)

Authors: Patrick Betz, Christian Meilicke, and Heiner Stuckenschmidt

Published in: OASIcs, Volume 138, Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 & RW 2025)

Abstract
The field of knowledge graph completion is concerned with augmenting knowledge graphs with missing information. Symbolic rule-based approaches are not only efficient and interpretable but also competitive with embedding-based methods in regard to predictive quality. Rule-based knowledge graph completion can be separated into two stages, the learning stage and the application stage, which are both individually challenging. In the learning stage, horn rules are mined from a given knowledge graph. Given the vast size of the space of all possible rules, the mining approach must select relevant rules effectively. In the application stage, the mined rules are used to make new predictions which are assigned with plausibility scores. These scores need to be set by aggregating individual confidence values of rules that have the same consequence. This tutorial covers the fundamental aspects required to build a symbolic rule-based approach for knowledge graph completion. It will discuss the different rule types, mining strategies, and how to effectively apply the rules in different scenarios. Finally, we discuss practical examples for rule application by using the Python-based PyClause library.
Cite as

Patrick Betz, Christian Meilicke, and Heiner Stuckenschmidt. Rule-Based Knowledge Graph Completion (Invited Paper). In Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 & RW 2025). Open Access Series in Informatics (OASIcs), Volume 138, pp. 1:1-1:45, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@InProceedings{betz_et_al:OASIcs.RW.2024/2025.1,
  author =	{Betz, Patrick and Meilicke, Christian and Stuckenschmidt, Heiner},
  title =	{{Rule-Based Knowledge Graph Completion}},
  booktitle =	{Joint Proceedings of the 20th and 21st Reasoning Web Summer Schools (RW 2024 \& RW 2025)},
  pages =	{1:1--1:45},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-405-5},
  ISSN =	{2190-6807},
  year =	{2025},
  volume =	{138},
  editor =	{Artale, Alessandro and Bienvenu, Meghyn and Garc{\'\i}a, Yazm{\'\i}n Ib\'{a}\~{n}ez and Murlak, Filip},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.RW.2024/2025.1},
  URN =		{urn:nbn:de:0030-drops-250461},
  doi =		{10.4230/OASIcs.RW.2024/2025.1},
  annote =	{Keywords: Knowledge Graph Completion, Rule Learning, Symbolic AI}
}
Document
DOI: 10.4230/LIPIcs.GD.2025.11
Towards a Better Understanding of Graph Perception in Immersive Environments

Authors: Lin Zhang, Yao Wang, Ying Zhang, Wilhelm Kerle-Malcharek, Karsten Klein, Falk Schreiber, and Andreas Bulling

Published in: LIPIcs, Volume 357, 33rd International Symposium on Graph Drawing and Network Visualization (GD 2025)

Abstract
As Immersive Analytics (IA) increasingly uses Virtual Reality (VR) for stereoscopic 3D (S3D) graph visualisation, it is crucial to understand how users perceive network structures in these immersive environments. However, little is known about how humans read S3D graphs during task solving, and how gaze behaviour indicates task performance. To address this gap, we report a user study with 18 participants asked to perform three analytical tasks on S3D graph visualisations in a VR environment. Our findings reveal systematic relationships between network structural properties and gaze behaviour. Based on these insights, we contribute a comprehensive eye tracking methodology for analysing human perception in immersive environments and establish eye tracking as a valuable tool for objectively evaluating cognitive load in S3D graph visualisation.
Cite as

Lin Zhang, Yao Wang, Ying Zhang, Wilhelm Kerle-Malcharek, Karsten Klein, Falk Schreiber, and Andreas Bulling. Towards a Better Understanding of Graph Perception in Immersive Environments. In 33rd International Symposium on Graph Drawing and Network Visualization (GD 2025). Leibniz International Proceedings in Informatics (LIPIcs), Volume 357, pp. 11:1-11:19, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@InProceedings{zhang_et_al:LIPIcs.GD.2025.11,
  author =	{Zhang, Lin and Wang, Yao and Zhang, Ying and Kerle-Malcharek, Wilhelm and Klein, Karsten and Schreiber, Falk and Bulling, Andreas},
  title =	{{Towards a Better Understanding of Graph Perception in Immersive Environments}},
  booktitle =	{33rd International Symposium on Graph Drawing and Network Visualization (GD 2025)},
  pages =	{11:1--11:19},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-403-1},
  ISSN =	{1868-8969},
  year =	{2025},
  volume =	{357},
  editor =	{Dujmovi\'{c}, Vida and Montecchiani, Fabrizio},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GD.2025.11},
  URN =		{urn:nbn:de:0030-drops-249976},
  doi =		{10.4230/LIPIcs.GD.2025.11},
  annote =	{Keywords: Stereoscopic 3D, Graph Visualisation, Eye Tracking, Graph Perception}
}
Document
Survey
DOI: 10.4230/TGDK.3.2.1
Resilience in Knowledge Graph Embeddings

Authors: Arnab Sharma, N'Dah Jean Kouagou, and Axel-Cyrille Ngonga Ngomo

Published in: TGDK, Volume 3, Issue 2 (2025). Transactions on Graph Data and Knowledge, Volume 3, Issue 2

Abstract
In recent years, knowledge graphs have gained interest and witnessed widespread applications in various domains, such as information retrieval, question-answering, recommendation systems, amongst others. Large-scale knowledge graphs to this end have demonstrated their utility in effectively representing structured knowledge. To further facilitate the application of machine learning techniques, knowledge graph embedding models have been developed. Such models can transform entities and relationships within knowledge graphs into vectors. However, these embedding models often face challenges related to noise, missing information, distribution shift, adversarial attacks, etc. This can lead to sub-optimal embeddings and incorrect inferences, thereby negatively impacting downstream applications. While the existing literature has focused so far on adversarial attacks on KGE models, the challenges related to the other critical aspects remain unexplored. In this paper, we, first of all, give a unified definition of resilience, encompassing several factors such as generalisation, in-distribution generalization, distribution adaption, and robustness. After formalizing these concepts for machine learning in general, we define them in the context of knowledge graphs. To find the gap in the existing works on resilience in the context of knowledge graphs, we perform a systematic survey, taking into account all these aspects mentioned previously. Our survey results show that most of the existing works focus on a specific aspect of resilience, namely robustness. After categorizing such works based on their respective aspects of resilience, we discuss the challenges and future research directions.
Cite as

Arnab Sharma, N'Dah Jean Kouagou, and Axel-Cyrille Ngonga Ngomo. Resilience in Knowledge Graph Embeddings. In Transactions on Graph Data and Knowledge (TGDK), Volume 3, Issue 2, pp. 1:1-1:38, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{sharma_et_al:TGDK.3.2.1,
  author =	{Sharma, Arnab and Kouagou, N'Dah Jean and Ngomo, Axel-Cyrille Ngonga},
  title =	{{Resilience in Knowledge Graph Embeddings}},
  journal =	{Transactions on Graph Data and Knowledge},
  pages =	{1:1--1:38},
  ISSN =	{2942-7517},
  year =	{2025},
  volume =	{3},
  number =	{2},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/TGDK.3.2.1},
  URN =		{urn:nbn:de:0030-drops-248117},
  doi =		{10.4230/TGDK.3.2.1},
  annote =	{Keywords: Knowledge graphs, Resilience, Robustness}
}
Document
Research
DOI: 10.4230/TGDK.3.2.3
GraphRAG on Technical Documents - Impact of Knowledge Graph Schema

Authors: Henri Scaffidi, Melinda Hodkiewicz, Caitlin Woods, and Nicole Roocke

Published in: TGDK, Volume 3, Issue 2 (2025). Transactions on Graph Data and Knowledge, Volume 3, Issue 2

Abstract
Retrieval Augmented Generation (RAG) is seeing rapid adoption in industry to enable employees to query information captured in proprietary data for their organisation. In this work, we test the impact of domain-relevant knowledge graph schemas on the results of Microsoft’s GraphRAG pipeline. Our approach aims to address the poor quality of GraphRAG responses on technical reports rich in domain-specific terms. The use case involves technical reports about geology, chemistry and mineral processing published by the Minerals Research Institute of Western Australia (MRIWA). Four schemas are considered: a simple five-class minerals domain expert-developed schema, an expanded minerals domain schema, the Microsoft GraphRAG auto-generated schema, and a schema-less GraphRAG. These are compared to a conventional baseline RAG. Performance is evaluated using a scoring approach that accounts for the mix of correct, incorrect, additional, and missing content in RAG responses. The results show that the simple five-class minerals domain schema extracts approximately 10% more entities from the MRIWA reports than the other schema options. Additionally, both the five-class and the expanded eight-class minerals domain schemas produce the most factually correct answers and the fewest hallucinations. We attribute this to the minerals-specific schemas extracting more relevant, domain-specific information during the Indexing stage. As a result, the Query stage’s context window includes more high-value content. This contributes to the observed improvement in answer quality compared to the other pipelines. In contrast, pipelines with fewer domain-related entities in the KG retrieve less valuable information, leaving more room for irrelevant content in the context window. Baseline RAG responses were typically shorter, less complete, and contained more hallucinations compared to our GraphRAG pipelines. We provide a complete set of resources at https://github.com/nlp-tlp/GraphRAG-on-Minerals-Domain/tree/main. These resources include links to the MRIWA reports, a set of questions (from simple to challenging) along with domain-expert curated answers, schemas, and evaluations of the pipelines.
Cite as

Henri Scaffidi, Melinda Hodkiewicz, Caitlin Woods, and Nicole Roocke. GraphRAG on Technical Documents - Impact of Knowledge Graph Schema. In Transactions on Graph Data and Knowledge (TGDK), Volume 3, Issue 2, pp. 3:1-3:24, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{scaffidi_et_al:TGDK.3.2.3,
  author =	{Scaffidi, Henri and Hodkiewicz, Melinda and Woods, Caitlin and Roocke, Nicole},
  title =	{{GraphRAG on Technical Documents - Impact of Knowledge Graph Schema}},
  journal =	{Transactions on Graph Data and Knowledge},
  pages =	{3:1--3:24},
  ISSN =	{2942-7517},
  year =	{2025},
  volume =	{3},
  number =	{2},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/TGDK.3.2.3},
  URN =		{urn:nbn:de:0030-drops-248131},
  doi =		{10.4230/TGDK.3.2.3},
  annote =	{Keywords: RAG, minerals, local search, global search, entity extraction, competency questions}
}
  • Refine by Type
  • 61 Document/PDF
  • 50 Document/HTML

  • Refine by Publication Year
  • 8 2026
  • 34 2025
  • 2 2024
  • 9 2023
  • 2 2022
  • Show More...

  • Refine by Author
  • 3 Biswas, Russa
  • 3 Lissandrini, Matteo
  • 3 Liu, Allen
  • 3 Scherp, Ansgar
  • 3 de Melo, Gerard
  • Show More...

  • Refine by Series/Journal
  • 39 LIPIcs
  • 3 OASIcs
  • 1 LITES
  • 17 TGDK
  • 1 DagMan

  • Refine by Classification
  • 7 Computing methodologies → Knowledge representation and reasoning
  • 4 Computing methodologies → Natural language processing
  • 4 Computing methodologies → Semantic networks
  • 4 Information systems → Graph-based database models
  • 3 Theory of computation
  • Show More...

  • Refine by Keyword
  • 7 Knowledge Graphs
  • 3 Large Language Models
  • 2 Datalog
  • 2 Explainable AI
  • 2 Knowledge graphs
  • Show More...

Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact