Tootfinder

This https://arxiv.org/abs/2403.17645 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

DANCER: Entity Description Augmented Named Entity Corrector for Automatic Speech Recognition
End-to-end automatic speech recognition (E2E ASR) systems often suffer from mistranscription of domain-specific phrases, such as named entities, sometimes leading to catastrophic failures in downstream tasks. A family of fast and lightweight named entity correction (NEC) models for ASR have recently been proposed, which normally build on phonetic-level edit distance algorithms and have shown impressive NEC performance. However, as the named entity (NE) list grows, the problems of phonetic confu…

This https://arxiv.org/abs/2403.17645 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

DANCER: Entity Description Augmented Named Entity Corrector for Automatic Speech Recognition
End-to-end automatic speech recognition (E2E ASR) systems often suffer from mistranscription of domain-specific phrases, such as named entities, sometimes leading to catastrophic failures in downstream tasks. A family of fast and lightweight named entity correction (NEC) models for ASR have recently been proposed, which normally build on phonetic-level edit distance algorithms and have shown impressive NEC performance. However, as the named entity (NE) list grows, the problems of phonetic confu…

Quantification of chirality based on electric toroidal monopole
Akane Inda, Rikuto Oiwa, Satoru Hayami, Hiroshi M Yamamoto, Hiroaki Kusunose
https://arxiv.org/abs/2402.13611

Quantification of chirality based on electric toroidal monopole
Chirality ubiquitously appears in nature, however, its quantification remains obscure owing to the lack of microscopic description at the quantum-mechanical level. We propose a way of evaluating chirality in terms of electric toroidal monopole, a practical entity of time-reversal even pseudoscalar (parity-odd) object reflecting relevant electronic wave functions. For this purpose, we analyze a twisted methane at the quantum-mechanical level, showing that the electric toroidal monopoles become a…

New point of view about optical activity in helically-coiled fiber
Chun-Fang Li, Zhi-Juan Hu
https://arxiv.org/abs/2404.14334 https://

New point of view about optical activity in helically-coiled fiber
The optical activity in a helically-coiled optical fiber is reexamined. It is proven that not only is there no circular birefringence in the fiber but the polarization relative to the laboratory reference frame is not rotated along the fiber. The reason for this is that in contrast with the polarization vector, the Jones vector does not give a complete description of the polarization. As a mathematical entity in some local reference frame that depends on the instantaneous propagation direction,…

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Models
Jio Oh, Soyeon Kim, Junseok Seo, Jindong Wang, Ruochen Xu, Xing Xie, Steven Euijong Whang
https://arxiv.org/abs/2403.05266

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Models
Large language models (LLMs) have achieved unprecedented performance in various applications, yet their evaluation remains a critical issue. Existing hallucination benchmarks are either static or lack adjustable complexity for thorough analysis. We contend that utilizing existing relational databases is a promising approach for constructing benchmarks due to their accurate knowledge description via functional dependencies. We propose ERBench to automatically convert any relational database into…

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Models
Jio Oh, Soyeon Kim, Junseok Seo, Jindong Wang, Ruochen Xu, Xing Xie, Steven Euijong Whang
https://arxiv.org/abs/2403.05266

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Models
Large language models (LLMs) have achieved unprecedented performance in various applications, yet their evaluation remains a critical issue. Existing hallucination benchmarks are either static or lack adjustable complexity for thorough analysis. We contend that utilizing existing relational databases is a promising approach for constructing benchmarks due to their accurate knowledge description via functional dependencies. We propose ERBench to automatically convert any relational database into…