Selected work

Representative Papers

Delegated Fair Division

Jul 30, 2026

This study addresses fair allocation of indivisible resources in hierarchical settings, such as charitable organizations distributing goods first to central agencies and then to individual recipients. Focusing on this two-tier delegated allocation scenario, the work presents the first systematic modeling framework that distinguishes between two distinct structures of individual preference information. The authors propose a polynomial-time algorithm that simultaneously guarantees envy-freeness at both the central-agency and individual levels. By integrating techniques from combinatorial optimization and fairness theory, the designed mechanism efficiently achieves two-level fairness under both information structures, offering practical applicability in real-world contexts such as food donation distribution.

0 citationsRead paper

Time Series Decomposition using the Fréchet Distance

Jul 05, 2026

This work proposes the Fréchet decomposition problem, which seeks to approximate a collection of univariate time series as Fréchet combinations of a small set of basis curves while minimizing the total Fréchet distance. Analogous to principal component analysis, this is the first effort to incorporate the Fréchet distance into time series decomposition, and it introduces two distinct variants of the problem. For the single-basis setting, the authors design a (1+ε)-approximation algorithm; for an arbitrary number k of basis curves, they present an exact solution method based on projection distance, assuming a given candidate set of basis curves. By delivering both an efficient approximation scheme and a general exact algorithm, this study establishes a novel paradigm for modeling nonlinear structures in time series data.

0 citationsRead paper

A Note on EFX Inapproximability for Chores

May 20, 2026

This study investigates the existence of constant-factor approximations to envy-freeness up to any good (EFX) in the allocation of indivisible chores under complement-free cost functions. Focusing on monotone subadditive and submodular cost settings, the authors introduce a novel approach combining refined counterexample constructions with a weighted covering model and ordinal preference profiles. They establish the first explicit inapproximability lower bounds for EFX allocations: in a three-agent, six- chore instance, any α-EFX allocation must satisfy α ≥ 2^{1/3} ≈ 1.26 under subadditive costs and α ≥ 20/19 under submodular costs. These results significantly narrow the gap between previously known upper and lower bounds, advancing the understanding of fairness guarantees in chore division under realistic cost structures.

0 citationsRead paper

Match-Any-Events: Zero-Shot Motion-Robust Feature Matching Across Wide Baselines for Event Cameras

Apr 20, 2026

This work addresses the challenge of feature matching with event cameras under large baselines, arbitrary viewpoints, and extreme motion—particularly the lack of effective correspondences across datasets. We propose the first event-based feature matching model capable of zero-shot cross-dataset deployment. Our approach introduces a motion-robust and computationally efficient attention-based backbone that integrates multi-temporal-scale features, coupled with a sparsity-aware event token selection strategy to enhance both training and inference efficiency. Leveraging a large-scale synthetic event-motion framework and wide-baseline supervision, the model achieves strong generalization without target-domain fine-tuning, outperforming state-of-the-art methods by 37.7% across multiple benchmarks and establishing the first zero-shot solution for large-baseline cross-dataset event matching.

0 citationsRead paper

Detecting Media Clones in Cultural Repositories Using a Positive Unlabeled Learning Approach

Apr 05, 2026

This study addresses the challenge of detecting unlabeled cross-record media duplicates in cultural heritage databases by formulating it as a positive-unlabeled (PU) learning task. The authors propose a lightweight method that requires no explicit negative samples: a query encoder is trained using only a single positive example, leveraging data augmentation and latent representation learning, while L2-norm thresholds in the embedding space transparently filter candidate clones. This approach represents the first effective application of PU learning to media deduplication in cultural heritage contexts. Evaluated on the AtticPOT dataset, it achieves an F1 score of 90.79 (AUROC = 98.99), outperforming the SVDD baseline by 7.70 F1 points; on CIFAR-10, it attains an F1 of 96.37, demonstrating robustness to viewpoint and condition variations. The method is well-suited for curator workflows involving duplicate detection and record linkage.

0 citationsRead paper

Recent publications

Latest Papers

Delegated Fair Division

Jul 30, 2026

This study addresses fair allocation of indivisible resources in hierarchical settings, such as charitable organizations distributing goods first to central agencies and then to individual recipients. Focusing on this two-tier delegated allocation scenario, the work presents the first systematic modeling framework that distinguishes between two distinct structures of individual preference information. The authors propose a polynomial-time algorithm that simultaneously guarantees envy-freeness at both the central-agency and individual levels. By integrating techniques from combinatorial optimization and fairness theory, the designed mechanism efficiently achieves two-level fairness under both information structures, offering practical applicability in real-world contexts such as food donation distribution.

0 citationsRead paper

Time Series Decomposition using the Fréchet Distance

Jul 05, 2026

This work proposes the Fréchet decomposition problem, which seeks to approximate a collection of univariate time series as Fréchet combinations of a small set of basis curves while minimizing the total Fréchet distance. Analogous to principal component analysis, this is the first effort to incorporate the Fréchet distance into time series decomposition, and it introduces two distinct variants of the problem. For the single-basis setting, the authors design a (1+ε)-approximation algorithm; for an arbitrary number k of basis curves, they present an exact solution method based on projection distance, assuming a given candidate set of basis curves. By delivering both an efficient approximation scheme and a general exact algorithm, this study establishes a novel paradigm for modeling nonlinear structures in time series data.

0 citationsRead paper

A Note on EFX Inapproximability for Chores

May 20, 2026

This study investigates the existence of constant-factor approximations to envy-freeness up to any good (EFX) in the allocation of indivisible chores under complement-free cost functions. Focusing on monotone subadditive and submodular cost settings, the authors introduce a novel approach combining refined counterexample constructions with a weighted covering model and ordinal preference profiles. They establish the first explicit inapproximability lower bounds for EFX allocations: in a three-agent, six- chore instance, any α-EFX allocation must satisfy α ≥ 2^{1/3} ≈ 1.26 under subadditive costs and α ≥ 20/19 under submodular costs. These results significantly narrow the gap between previously known upper and lower bounds, advancing the understanding of fairness guarantees in chore division under realistic cost structures.

0 citationsRead paper

Match-Any-Events: Zero-Shot Motion-Robust Feature Matching Across Wide Baselines for Event Cameras

Apr 20, 2026

This work addresses the challenge of feature matching with event cameras under large baselines, arbitrary viewpoints, and extreme motion—particularly the lack of effective correspondences across datasets. We propose the first event-based feature matching model capable of zero-shot cross-dataset deployment. Our approach introduces a motion-robust and computationally efficient attention-based backbone that integrates multi-temporal-scale features, coupled with a sparsity-aware event token selection strategy to enhance both training and inference efficiency. Leveraging a large-scale synthetic event-motion framework and wide-baseline supervision, the model achieves strong generalization without target-domain fine-tuning, outperforming state-of-the-art methods by 37.7% across multiple benchmarks and establishing the first zero-shot solution for large-baseline cross-dataset event matching.

0 citationsRead paper

Detecting Media Clones in Cultural Repositories Using a Positive Unlabeled Learning Approach

Apr 05, 2026

This study addresses the challenge of detecting unlabeled cross-record media duplicates in cultural heritage databases by formulating it as a positive-unlabeled (PU) learning task. The authors propose a lightweight method that requires no explicit negative samples: a query encoder is trained using only a single positive example, leveraging data augmentation and latent representation learning, while L2-norm thresholds in the embedding space transparently filter candidate clones. This approach represents the first effective application of PU learning to media deduplication in cultural heritage contexts. Evaluated on the AtticPOT dataset, it achieves an F1 score of 90.79 (AUROC = 98.99), outperforming the SVDD baseline by 7.70 F1 points; on CIFAR-10, it attains an F1 of 96.37, demonstrating robustness to viewpoint and condition variations. The method is well-suited for curator workflows involving duplicate detection and record linkage.

0 citationsRead paper