arXiv subjects

Takuma Udagawa

Publications and source records attributed to Takuma Udagawa.

23 records · Page 2Linked to original sources

arXiv2020

A Linguistic Analysis of Visually Grounded Dialogues Based on Spatial Expressions

Recent models achieve promising results in visually grounded dialogues. However, existing datasets often contain undesirable biases and lack sophisticated linguistic analyses, which make it difficult to understand how well current models recognize their precise linguistic structures. To address this problem, we make two design choices: first, we focus on OneCommon Corpus \citep{udagawa2019natural,udagawa2020annotated}, a simple yet challenging common grounding dataset which contains minimal bias by design. Second, we analyze their linguistic structures based on \textit{spatial expressions} and provide comprehensive and reliable annotation for 600 dialogues. We show that our annotation captures important linguistic structures including predicate-argument structure, modification and ellipsis. In our experiments, we assess the model's understanding of these structures through reference resolution. We demonstrate that our annotation can reveal both the strengths and weaknesses of baseline models in essential levels of detail. Overall, we propose a novel framework and resource for investigating fine-grained language understanding in visually grounded dialogues.

cs.CL↗

arXiv2019

An Annotated Corpus of Reference Resolution for Interpreting Common Grounding

Common grounding is the process of creating, repairing and updating mutual understandings, which is a fundamental aspect of natural language conversation. However, interpreting the process of common grounding is a challenging task, especially under continuous and partially-observable context where complex ambiguity, uncertainty, partial understandings and misunderstandings are introduced. Interpretation becomes even more challenging when we deal with dialogue systems which still have limited capability of natural language understanding and generation. To address this problem, we consider reference resolution as the central subtask of common grounding and propose a new resource to study its intermediate process. Based on a simple and general annotation schema, we collected a total of 40,172 referring expressions in 5,191 dialogues curated from an existing corpus, along with multiple judgements of referent interpretations. We show that our annotation is highly reliable, captures the complexity of common grounding through a natural degree of reasonable disagreements, and allows for more detailed and quantitative analyses of common grounding strategies. Finally, we demonstrate the advantages of our annotation for interpreting, analyzing and improving common grounding in baseline dialogue systems.

cs.CL↗

arXiv2019

A Natural Language Corpus of Common Grounding under Continuous and Partially-Observable Context

Common grounding is the process of creating, repairing and updating mutual understandings, which is a critical aspect of sophisticated human communication. However, traditional dialogue systems have limited capability of establishing common ground, and we also lack task formulations which introduce natural difficulty in terms of common grounding while enabling easy evaluation and analysis of complex models. In this paper, we propose a minimal dialogue task which requires advanced skills of common grounding under continuous and partially-observable context. Based on this task formulation, we collected a largescale dataset of 6,760 dialogues which fulfills essential requirements of natural language corpora. Our analysis of the dataset revealed important phenomena related to common grounding that need to be considered. Finally, we evaluate and analyze baseline neural models on a simple subtask that requires recognition of the created common ground. We show that simple baseline models perform decently but leave room for further improvement. Overall, we show that our proposed task will be a fundamental testbed where we can train, evaluate, and analyze dialogue system's ability for sophisticated common grounding.

cs.CL↗

arXiv2017

Finite-size Gap, Magnetization, and Entanglement of Deformed Fredkin Spin Chain

We investigate ground- and excited-state properties of the deformed Fredkin spin chain proposed by Salberger, Zhang, Klich, Korepin, and the authors. This model is a one-parameter deformation of the Fredkin spin chain, whose Hamiltonian is $3$-local and translationally invariant in the bulk. The model is frustration-free and its unique ground state can be expressed as a weighted superposition of colored Dyck paths. We focus on the case where the deformation parameter $t>1$. By using a variational method, we prove that the finite-size gap decays at least exponentially with increasing the system size. We prove that the magnetization in the ground state is along the $z$-direction, namely $\langle s^x \rangle =\langle s^y \rangle=0$, and show that the $z$-component $\langle s^z \rangle$ exhibits a domain-wall structure. We then study the entanglement properties of the chain. In particular, we derive upper and lower bounds for the von Neumann and Rényi entropies, and entanglement spectrum for any bipartition of the chain.

cond-mat.stat-mech↗

arXiv2016

Deformed Fredkin Spin Chain with Extensive Entanglement

We introduce a new spin chain which is a deformation of the Fredkin spin chain and has a phase transition between bounded and extensive entanglement entropy scaling. In this chain, spins have a local interaction of three nearest neighbors. The Hamiltonian is frustration-free and its ground state can be described analytically as a weighted superposition of Dyck paths. In the purely spin $1/2$ case, the entanglement entropy obeys an area law: it is bounded from above by a constant, when the size of the block $n$ increases (and $t>1$). When a local color degree of freedom is introduced the entanglement entropy increases linearly with the size of the block (and $t>1$). The entanglement entropy of half of the chain is tightly bounded by ${ n}\log s$ where $n$ is the size of the block, and $s$ is the number of colors. Our chain fosters a new example for a significant boost to entropy and for the existence of the associated critical rainbow phase where the entanglement entropy scales with volume that has recently been discovered in Zhang et al. (arXiv:1606.07795)

cond-mat.stat-mech↗