ALQAC 2022: A Summary of the Competition

Nguyen, Chau; Bui, Minh-Quan; Do, Dinh-Truong; Le, Nguyen-Khang; Nguyen, Dieu-Hien; Nguyen, Thu-Trang; Nguyen, Ha-Thanh; Tran, Vu; Nguyen, Le-Minh; Ngoc-Cam, Le,; Le, Thi-Thuy; Nguyen, Minh-Phuong; Dang, Tran-Binh; Nguyen, T. D.; Phan, Viet-Anh; Vuong, Thi-Hai-Yen; Nguyen, Minh-Tien; Le, Tung

doi:10.1109/kse56063.2022.9953764

Cited by 8 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The use of artificial intelligence (AI) and machine learning in the legal field has attracted significant attention in recent years, resulting in the development of numerous state-ofthe-art models capable of performing various legal text processing tasks, such as contract analysis and case law retrieval [3,4,5,6]. Deep learning models, which offer automated feature extraction, have been utilized not only in similarity matching tasks but also in other semantic matching tasks such as question answering [7,8], machine reading comprehension [9,10], image retrieval [11], and entity matching [12,13].…”

Section: Background and Related Workmentioning

confidence: 99%

“…Legal retrieval tasks involve identifying relevant legal regulations based on given queries and are fundamental components of intelligent legal counsel systems [3,4,5,6]. Moreover, deep learning and transfer learning methods have successfully addressed the limited data challenges commonly encountered in legal retrieval tasks, leading to the development of various effective approaches [7,14,15,16,17].…”

Section: Background and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

LogiLaw Dataset Towards Reinforcement Learning from Logical Feedback (RLLF)

Nguyen,

Fungwacharakorn,

Satoh

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Large Language Models (LLMs) face limitations in logical reasoning, which restrict their applicability in critical domains such as law. Current evaluation methods often lead to inaccurate assessments of LLMs’ capabilities due to their simplicity. This paper presents a refined evaluation method for assessing LLMs’ capability to answer legal questions by eliminating the possibility of obtaining correct responses by chance. Furthermore, we introduce the LogiLaw dataset, which aims to enhance the models’ logical reasoning capacities in general and legal reasoning specifically. By leveraging the refined evaluation technique, the LogiLaw dataset, and the proposed Reinforcement Learning from Logical Feedback (RLLF) approach, our work aims to open new avenues for research to bolster LLMs’ performance in law and other logic-intensive disciplines while addressing the shortcomings of conventional evaluation approaches.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Section: Background and Related Workmentioning

confidence: 99%

LogiLaw Dataset Towards Reinforcement Learning from Logical Feedback (RLLF)

Nguyen,

Fungwacharakorn,

Satoh

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

show abstract

“…Антибиотикорезистентность является глобальной проблемой здравоохранения разных стран. Значительный вклад в распространение вносят антропогенные химические стрессы, которые связаны с экспозицией к металлам, углеводородам, органическим соединениям [6]. Многие микроорганизмы способны модифицировать антибиотики и мишени для них, снижать внутриклеточное накопление, изменяя проницаемость мембран и увеличивая активность оттока [6][7][8][9].…”

Section: вредные производственные факторы влияют на иммунологическую ...unclassified

Immune response status and development of streptococcus pneumoniae carriage as health risk factors for workers engaged in coke production and basic oxygen steelmaking

Bushueva,

Karpova,

Roslaya

et al. 2023

Health Risk Analysis

View full text Add to dashboard Cite

Occupational hazards affect immunological reactivity and increase the risk of infection with respiratory pathogens. The purpose of our work was to study the mechanisms and causes of developing carriage of streptococci possessing genetic determinants of resistance to antibiotics as health risk factors for workers engaged in coke production and basic oxygen steelmaking. We examined 136 workers of a ferrous metals industry and compared the prevalence of the carriage of Streptococcus pneumoniae, Streptococcus spp. and macrolide resistance genes (Mef, ErmB) in those engaged in coke production and basic oxygen steelmaking. We also collated the results of immunological examination of the carriers and non-carriers of S. pneumoniae. The control cohort was formed of engineering and technical personnel. We established that, compared with the controls, Streptococcus pneumoniae was a more frequent finding in the workers engaged in coke production and basic oxygen steelmaking (p ˂ 0.05). The cycle threshold value of Streptococcus spp. in them was also statistically higher. The Mef gene was more abundant in the steelmakers, including those carrying Streptococcus pneumoniae (p ˂ 0.05), and a change in immunological reactivity was detected in the cases carrying Streptococcus pneumoniae. In those engaged in basic oxygen steelmaking, significant changes were detected at the cellular level expressed by a reduced absolute number of mature T-lymphocytes (CD3+), T-helpers (CD4+), and T-suppressor/cytotoxic cells (CD8+) (p ≤ 0.05). At the humoral level, both the relative and absolute number of B-lymphocytes (CD19+) and the level of IgM were reduced, while that of IgG was, on the opposite, increased (p ≤ 0.05). In the coke plant workers carrying Streptococcus pneumoniae, the level of IgG was significantly higher than the control and reference values. We conclude that Streptococcus pneumoniae carriage in the steelmakers as a health risk factor was accompanied by a change in immunological reactivity and higher abundance of the Mef gene compared to the controls while in the coke production workers the related differences were detected only in the IgG level.

show abstract

“…This results in an incorrect interpretation of standard triaxial tests from the macroscale for constitutive behaviour. 26 To evaluate the behaviour of a localized sample at scales beyond the sample size, semi-analytical methods based on the 2-scale framework 11,12,14,27 have been recently developed to interpret the standard triaxial results. 28 They show promising potentials and enable the quantification of behaviour inside the localization band from the experimental measurement of specimen behaviour.…”

Section: Introductionmentioning

confidence: 99%

Strain localization in the standard triaxial tests of granular materials: Insights into meso‐ and macro‐scale behaviours

Phan,

Bui,

Nguyen

et al. 2024

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

The standard triaxial tests cease to be valid as material tests since the homogeneity of the granular mass is lost when localized failures such as shear bands occur requiring a different approach to interpreting and analyzing material responses at the meso scale from the macro behaviour. This study sheds light on the above issue by analyzing the standard triaxial tests of the granular specimens undergoing localized failure at different scales using DEM. For the first time, the behaviour of material inside and outside of the shear band as well as the entire sample are quantitatively quantified through DEM simulations. The results enable confirmation of various theoretical hypotheses and experimental observations on the localized failure in granular materials. For example, by quantifying the inter‐particle contact forces of materials inside and outside the shear band zone, it is confirmed that the material outside the localization band undergoes inelastic unloading beyond the bifurcation point, while those inside the localization band experience inelastic shearing to reach the critical state. Moreover, the analysis in this study suggests that if the volume of the localization zone can be precisely measured from the experiment, the mesoscale constitutive responses that truly represent the inelastic behaviour can be quantified. This gives rise to a more appropriate method to obtain inelastic constitutive responses of granular materials from specimens undergoing localized failure in triaxial tests.

show abstract

ALQAC 2022: A Summary of the Competition

Cited by 8 publications

References 6 publications

LogiLaw Dataset Towards Reinforcement Learning from Logical Feedback (RLLF)

LogiLaw Dataset Towards Reinforcement Learning from Logical Feedback (RLLF)

Immune response status and development of streptococcus pneumoniae carriage as health risk factors for workers engaged in coke production and basic oxygen steelmaking

Strain localization in the standard triaxial tests of granular materials: Insights into meso‐ and macro‐scale behaviours

Contact Info

Product

Resources

About