Marco Kuhlmann
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Marco Kuhlmann has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
Giorgio Satta
University of Padua
3 shared publicationsJoakim Nivre
3 shared publicationsAlexander Koller
2 shared publicationsDenys Duchier
LORIA
2 shared publications- Emanuel Sanchez Aimar
Linköpings universitet, Datorseende
2 shared publications - Giorgio Satta
Dept. of Information Engineering University of Padua
2 shared publications
Latest Publications
- 31 October 2024
- Published by Springer Nature
- p. 307-327
Flexible Distribution Alignment: Towards Long-Tailed Semi-supervised Learning with Proper Calibration
- 31 October 2024
- Published by Springer Nature
- p. 307-327
Flexible Distribution Alignment: Towards Long-Tailed Semi-supervised Learning with Proper Calibration
- Open Access
Properties and Challenges of LLM-Generated Explanations
The self-rationalising capabilities of large language models (LLMs) have been explored in restricted settings, using task/specific data sets. However, current LLMs do not (only) rely on specifically annotated data; nonetheless, they frequently explain their outputs. The properties of the generated explanations are influenced by the pre-training corpus and by the target data used for instruction fine-tuning. As the pre-training corpus includes a large amount of human-written explanations "in the wild", we hypothesise that LLMs adopt common properties of human explanations. By analysing the outputs for a multi-domain instruction fine-tuning data set, we find that generated explanations show selectivity and contain illustrative elements, but less frequently are subjective or misleading. We discuss reasons and consequences of the properties' presence or absence. In particular, we outline positive and negative implications depending on the goals and user groups of the self-rationalising system.
Open AccessProperties and Challenges of LLM-Generated Explanations
The self-rationalising capabilities of large language models (LLMs) have been explored in restricted settings, using task/specific data sets. However, current LLMs do not (only) rely on specifically annotated data; nonetheless, they frequently explain their outputs. The properties of the generated explanations are influenced by the pre-training corpus and by the target data used for instruction fine-tuning. As the pre-training corpus includes a large amount of human-written explanations "in the wild", we hypothesise that LLMs adopt common properties of human explanations. By analysing the outputs for a multi-domain instruction fine-tuning data set, we find that generated explanations show selectivity and contain illustrative elements, but less frequently are subjective or misleading. We discuss reasons and consequences of the properties' presence or absence. In particular, we outline positive and negative implications depending on the goals and user groups of the self-rationalising system.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 13-27
Properties and Challenges of LLM-Generated Explanations
Jenny Kunz, Marco Kuhlmann. Proceedings of the Third Workshop on Bridging Human--Computer Interaction and Natural Language Processing. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 13-27
Open AccessProperties and Challenges of LLM-Generated Explanations
Jenny Kunz, Marco Kuhlmann. Proceedings of the Third Workshop on Bridging Human--Computer Interaction and Natural Language Processing. 2024.
- Open Access
Balanced Product of Calibrated Experts for Long-Tailed Recognition
Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the test distribution differs from the training distribution, e.g. uniform versus long-tailed, the problem of the distribution shift needs to be addressed. A recent line of work proposes learning multiple diverse experts to tackle this issue. Ensemble diversity is encouraged by various techniques, e.g. by specializing different experts in the head and the tail classes. In this work, we take an analytical approach and extend the notion of logit adjustment to ensembles to form a Balanced Product of Experts (BalPoE). BalPoE combines a family of experts with different test-time target distributions, generalizing several previous approaches. We show how to properly define these distributions and combine the experts in order to achieve unbiased predictions, by proving that the ensemble is Fisher-consistent for minimizing the balanced error. Our theoretical analysis shows that our balanced ensemble requires calibrated experts, which we achieve in practice using mixup. We conduct extensive experiments and our method obtains new state-of-the-art results on three long-tailed datasets: CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018. Our code is available at https://github.com/emasa/BalPoE-CalibratedLT.
Open AccessBalanced Product of Calibrated Experts for Long-Tailed Recognition
Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the test distribution differs from the training distribution, e.g. uniform versus long-tailed, the problem of the distribution shift needs to be addressed. A recent line of work proposes learning multiple diverse experts to tackle this issue. Ensemble diversity is encouraged by various techniques, e.g. by specializing different experts in the head and the tail classes. In this work, we take an analytical approach and extend the notion of logit adjustment to ensembles to form a Balanced Product of Experts (BalPoE). BalPoE combines a family of experts with different test-time target distributions, generalizing several previous approaches. We show how to properly define these distributions and combine the experts in order to achieve unbiased predictions, by proving that the ensemble is Fisher-consistent for minimizing the balanced error. Our theoretical analysis shows that our balanced ensemble requires calibrated experts, which we achieve in practice using mixup. We conduct extensive experiments and our method obtains new state-of-the-art results on three long-tailed datasets: CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018. Our code is available at https://github.com/emasa/BalPoE-CalibratedLT.
- Open Access
- 19 May 2022
- Published by MIT Press in Computational Linguistics
- Vol. 48 (3) , 1-41
Tractable Parsing for CCGs of Bounded Degree
Unlike other mildly context-sensitive formalisms, Combinatory Categorial Grammar (CCG) cannot be parsed in polynomial time when the size of the grammar is taken into account. Refining this result, we show that the parsing complexity of CCG is exponential only in the maximum degree of composition. When that degree is fixed, parsing can be carried out in polynomial time. Our finding is interesting from a linguistic perspective because a bounded degree of composition has been suggested as a universal constraint on natural language grammar. Moreover, ours is the first complexity result for a version of CCG that includes substitution rules, which are used in practical grammars but have been ignored in theoretical work.
- 19 May 2022
- Published by MIT Press in Computational Linguistics
- Vol. 48 (3) , 1-41
Open AccessTractable Parsing for CCGs of Bounded Degree
Unlike other mildly context-sensitive formalisms, Combinatory Categorial Grammar (CCG) cannot be parsed in polynomial time when the size of the grammar is taken into account. Refining this result, we show that the parsing complexity of CCG is exponential only in the maximum degree of composition. When that degree is fixed, parsing can be carried out in polynomial time. Our finding is interesting from a linguistic perspective because a bounded degree of composition has been suggested as a universal constraint on natural language grammar. Moreover, ours is the first complexity result for a version of CCG that includes substitution rules, which are used in practical grammars but have been ignored in theoretical work.
- Open Access
- 1 March 2022
- Published by Elsevier in Journal of Computer and System Sciences
- Vol. 124, 214-233
The tree-generative capacity of combinatory categorial grammars
- 1 March 2022
- Published by Elsevier in Journal of Computer and System Sciences
- Vol. 124, 214-233
Open AccessThe tree-generative capacity of combinatory categorial grammars
- Open Access
On the Complexity of CCG Parsing
We study the parsing complexity of Combinatory Categorial Grammar (CCG) in the formalism of Vijay-Shanker and Weir (1994). As our main result, we prove that any parsing algorithm for this formalism will take in the worst case exponential time when the size of the grammar, and not only the length of the input sentence, is included in the analysis. This sets the formalism of Vijay-Shanker and Weir (1994) apart from weakly equivalent formalisms such as Tree-Adjoining Grammar (TAG), for which parsing can be performed in time polynomial in the combined size of grammar and input sentence. Our results contribute to a refined understanding of the class of mildly context-sensitive grammars, and inform the search for new, mildly context-sensitive versions of CCG.
Open AccessOn the Complexity of CCG Parsing
We study the parsing complexity of Combinatory Categorial Grammar (CCG) in the formalism of Vijay-Shanker and Weir (1994). As our main result, we prove that any parsing algorithm for this formalism will take in the worst case exponential time when the size of the grammar, and not only the length of the input sentence, is included in the analysis. This sets the formalism of Vijay-Shanker and Weir (1994) apart from weakly equivalent formalisms such as Tree-Adjoining Grammar (TAG), for which parsing can be performed in time polynomial in the combined size of grammar and input sentence. Our results contribute to a refined understanding of the class of mildly context-sensitive grammars, and inform the search for new, mildly context-sensitive versions of CCG.
- 14 December 2016
- Published by MIT Press in Computational Linguistics
- p. 1-13
Towards A Catalogue of Linguistic Graph Banks
Graphs exceeding the formal complexity of rooted trees are of growing relevance to much NLP research. Although formally well-understood in graph theory, there is substantial variation in the types of linguistic graphs, as well as in the interpretation of various structural properties. To provide a common terminology and transparent statistics across different collections of graphs in NLP, we propose to establish a shared community resource with an open-source reference implementation for common statistics.
- 14 December 2016
- Published by MIT Press in Computational Linguistics
- p. 1-13
Towards A Catalogue of Linguistic Graph Banks
Graphs exceeding the formal complexity of rooted trees are of growing relevance to much NLP research. Although formally well-understood in graph theory, there is substantial variation in the types of linguistic graphs, as well as in the interpretation of various structural properties. To provide a common terminology and transparent statistics across different collections of graphs in NLP, we propose to establish a shared community resource with an open-source reference implementation for common statistics.
- Open Access
- 1 December 2016
- Published by MIT Press in Computational Linguistics
- Vol. 42 (4) , 819-827
Towards a Catalogue of Linguistic Graph Banks
Graphs exceeding the formal complexity of rooted trees are of growing relevance to much NLP research. Although formally well understood in graph theory, there is substantial variation in the types of linguistic graphs, as well as in the interpretation of various structural properties. To provide a common terminology and transparent statistics across different collections of graphs in NLP, we propose to establish a shared community resource with an open-source reference implementation for common statistics.
- 1 December 2016
- Published by MIT Press in Computational Linguistics
- Vol. 42 (4) , 819-827
Open AccessTowards a Catalogue of Linguistic Graph Banks
Graphs exceeding the formal complexity of rooted trees are of growing relevance to much NLP research. Although formally well understood in graph theory, there is substantial variation in the types of linguistic graphs, as well as in the interpretation of various structural properties. To provide a common terminology and transparent statistics across different collections of graphs in NLP, we propose to establish a shared community resource with an open-source reference implementation for common statistics.