ABSTRACT Architectural lighting has potent biological effects but applied lighting practices that capitalize on this potential have been limited. In this review, we endeavor to consolidate and synthesize key references that will be useful for lighting professionals, with the goal of supporting knowledge translation into pragmatic lighting strategies. Specifically, we explain relevant terminology, outline basic concepts, identify key references, provide a balanced overview of the current state of knowledge, and highlight important remaining questions. We summarize the physiological effects of light on human health and well-being, including a description of the processes underlying the photic regulation of circadian, neuroendocrine, and neurobehavioral functions. We review seminal work elucidating the elements mediating the potency of light for these physiological responses, with specific attention to factors critical for interpreting those findings. In parallel, we explain and endorse melanopic Equivalent Daylight Illuminance ( ) as the preferred measure to quantify the biological potency of light. Ultimately, while future studies are necessary to further facilitate the translation of laboratory knowledge to domestic and workplace settings, the immediate potential for applied lighting to better support human health is clear. Aiming for integrative lighting solutions that have biologically high potency light during the day and low potency during the night is perhaps the most immediate improvement to be made in order to better support applications for humans.
Most of modern neural machine translation (NMT) models are based on an encoder-decoder framework with an attention mechanism. While they perform well on standard datasets, they can have trouble in translation of long inputs that are rare or unseen during training. Incorporating target syntax is one approach to dealing with such length-related problems. We propose a novel syntactic decoder that generates a target-language dependency tree in a top-down, left-to-right order. Experiments show that the proposed top-down string-to-tree decoding generalizes better than conventional sequence-to-sequence decoding in translating long inputs that are not observed in the training data.
Owing to the rapid evolution of technologies and project requirements, organizations need to upgrade the code base in their software projects to a new version of the programming language or even translating to an entirely new one. However, code translation is resource-intensive and requires expertise in both the source and target languages. While researchers have made progress in automating translations between legacy and modern languages, recent work has increasingly turned to pre-trained Large Language Models (LLMs) to translate efficiently. Given the proprietary nature of code, organizations prefer fine-tuning LLMs locally rather than relying on external APIs. This is one of the first empirical studies that proposes a Federated LLM-based approach for code translation. The proposed approach enables clients to jointly train a code translator without sharing sensitive data. This study demonstrates that participants can collaboratively develop a FedLLM for efficient code translation (particularly C\# to Java and vice-versa) with superior results (more than 40\% improvement in CodeLLaMA's CodeBLEU score) compared to individual client models. Our findings indicate that FedLLM offers a collaborative approach to code translation and could serve as a promising direction for future research in this field.
Pronoun translation is a longstanding challenge in neural machine translation (NMT), often requiring inter-sentential context to ensure linguistic accuracy. To address this, we introduce ProNMT, a novel framework designed to enhance pronoun and overall translation quality in context-aware machine translation systems. ProNMT leverages Quality Estimation (QE) models and a unique Pronoun Generation Likelihood-Based Feedback mechanism to iteratively fine-tune pre-trained NMT models without relying on extensive human annotations. The framework combines QE scores with pronoun-specific rewards to guide training, ensuring improved handling of linguistic nuances. Extensive experiments demonstrate significant gains in pronoun translation accuracy and general translation quality across multiple metrics. ProNMT offers an efficient, scalable, and context-aware approach to improving NMT systems, particularly in translating context-dependent elements like pronouns.
Image quality scoring and interpreting are two fundamental components of Image Quality Assessment (IQA). The former quantifies image quality, while the latter enables descriptive question answering about image quality. Traditionally, these two tasks have been addressed independently. However, from the perspective of the Human Visual System (HVS) and the Perception-Decision Integration Model, they are inherently interconnected: interpreting serves as the foundation for scoring, while scoring provides an abstract summary of interpreting. Thus, unifying these capabilities within a single model is both intuitive and logically coherent. In this paper, we propose Q-SiT (Quality Scoring and Interpreting joint Teaching), a unified framework that enables large multimodal models (LMMs) to learn both image quality scoring and interpreting simultaneously. We achieve this by transforming conventional IQA datasets into learnable question-answering datasets and incorporating human-annotated quality interpreting data for training. Furthermore, we introduce an efficient scoring & interpreting balance strategy, which first determines the optimal data mix ratio on lightweight LMMs and then maps this ratio to primary LMMs for fine-tuning adjustment. This strategy not only mitigates task interference and enhances cross-task knowledge transfer but also significantly reduces computational costs compared to direct optimization on full-scale LMMs. With this joint learning framework and corresponding training strategy, we develop Q-SiT, the first model capable of simultaneously performing image quality scoring and interpreting tasks, along with its lightweight variant, Q-SiT-mini. Experimental results demonstrate that Q-SiT achieves strong performance in both tasks with superior generalization IQA abilities.Project page at https://github.com/Q-Future/Q-SiT.
This study investigates the types of difficulties faced by translators in achieving optimal interpretation. The relevance degree is adapted to achieve an optimal meaning when conveying the translated text. The text should be fully comprehensible without unneeded effort. To achieve this goal, the translator should understand the source text. Then, she or he recreates a translated text that is close to the original one, taking into consideration the intended audience and cultural norms. To apply it in the realm of optimality, the concepts of decision theory and utility theory are utilized. The study clarifies that the translator's selection of an utterance is based on relevance, as they aim to convey their intention and be understood. This in turn suggests that literal utterances take more effort to process than implied ones. For optimal relevance, the utterance should be the most relevant one, according to the readers' preferences and abilities. Finally, it is notable to mention that the study suggests developing and implementing this process of translation. As a result, it could significantly improve the translator's achievements in decoding any text appropriately.
This study evaluates the machine translation (MT) quality of two state-of-the-art large language models (LLMs) against a traditional neural machine translation (NMT) system across four language pairs in the legal domain. It combines automatic evaluation metrics (AEMs) and human evaluation (HE) by professional translators to assess translation ranking, fluency and adequacy. The results indicate that while Google Translate generally outperforms LLMs in AEMs, human evaluators rate LLMs, especially GPT-4, comparably or slightly better in terms of producing contextually adequate and fluent translations. This discrepancy suggests LLMs' potential in handling specialized legal terminology and context, highlighting the importance of human evaluation methods in assessing MT quality. The study underscores the evolving capabilities of LLMs in specialized domains and calls for reevaluation of traditional AEMs to better capture the nuances of LLM-generated translations.
Large language models (LLMs) show promise in code translation due to their ability to generate idiomatic code. However, a significant limitation when using LLMs for code translation is scalability: existing works have shown a drop in translation success rates for code exceeding around 100 lines. We overcome this limitation by developing a modular approach to translation, where we partition the code into small code fragments which can be translated independently and semantically validated (that is, checking I/O equivalence). When this approach is applied naively, we discover that LLMs are unreliable when translating features of the source language that do not have a direct mapping to the target language, and that the LLM often gets stuck in repair loops when attempting to fix errors. To address these issues, we introduce two key concepts: (1) feature mapping, which integrates predefined translation rules with LLM-based translation to guide the LLM in navigating subtle language differences and producing semantically accurate code; and (2) type-compatibility, which facilitates localized checks at the function signature level to detect errors early, thereby narrowing the scope of potential repairs. We apply our approach to translating real-world Go codebases to Rust, demonstrating that we can consistently generate reliable Rust translations for projects up to 6,600 lines of code and 369 functions, with an average of 73% of functions successfully validated for I/O equivalence, considerably higher than any existing work.
It is known that there is no a Type I singularity for the Lagrangian mean curvature flow with zero Maslov class. In this paper, we study translating solitons which are important models of Type II singularities. A necessary condition for a blow-up limit arising at a Type II singularity of a Lagrangian mean curvature flow with zero Maslov class is provided. As an application, we try to understand the important open question proposed by Joyce-Lee-Tsui and Neves-Tian, whether the Lagrangian translating solitons constructed by Joyce-Lee-Tsui can be a blow-up limit for a Lagrangian mean curvature flow with zero Maslov class.
The rapid growth of location acquisition technologies makes Point-of-Interest(POI) recommendation possible due to redundant user check-in records. In this paper, we focus on next POI recommendation in which next POI is based on previous POI. We observe that time plays an important role in next POI recommendation but is neglected in the recent proposed translating embedding methods. To tackle this shortage, we propose a time-adaptive translating embedding model (TransTARec) for next POI recommendation that naturally incorporates temporal influence, sequential dynamics, and user preference within a single component. Methodologically, we treat a (previous timestamp, user, next timestamp) triplet as a union translation vector and develop a neural-based fusion operation to fuse user preference and temporal influence. The superiority of TransTARec, which is confirmed by extensive experiments on real-world datasets, comes from not only the introduction of temporal influence but also the direct unification with user preference and sequential dynamics.
Ewa Ludera, Agnieszka Szarkowska, David Orrego-Carmona
Subtitling quality is a rather slippery notion and its assessment in interlingual subtitling continues to present a challenge to subtitling trainers, broadcasters, language service providers and other stakeholders. Using unexplored data from a subtitling process research study by Orrego-Carmona et al. (2018), we examined the quality of interlingual English-to-Polish subtitles created by professionals and novices. First, we implemented Pedersen’s (2017) FAR model to assess the quality of subtitles created by the participants, predicting that professionals would achieve higher quality scores than novices. Then, we followed up the FAR model examination with a quantitative analysis of a set of quality parameters related to text condensation, which is considered a key skill in interlingual subtitling. Despite our hypotheses, professional subtitlers in our study did not gain higher scores in the FAR model analysis; they also made similar types of errors as novices. However, their expertise was demonstrated in better condensation skills. We have also discovered an interesting relationship between subtitlers’ age and their condensation skills. Our study may contribute to a fuller understanding of expertise in interlingual subtitling and provide subtitling trainers with clues about areas most problematic for students.
Efficiently and accurately translating a corpus into a low-resource language remains a challenge, regardless of the strategies employed, whether manual, automated, or a combination of the two. Many Christian organizations are dedicated to the task of translating the Holy Bible into languages that lack a modern translation. Bible translation (BT) work is currently underway for over 3000 extremely low resource languages. We introduce the eBible corpus: a dataset containing 1009 translations of portions of the Bible with data in 833 different languages across 75 language families. In addition to a BT benchmarking dataset, we introduce model performance benchmarks built on the No Language Left Behind (NLLB) neural machine translation (NMT) models. Finally, we describe several problems specific to the domain of BT and consider how the established data and model benchmarks might be used for future translation efforts. For a BT task trained with NLLB, Austronesian and Trans-New Guinea language families achieve 35.1 and 31.6 BLEU scores respectively, which spurs future innovations for NMT for low-resource languages in Papua New Guinea.
Raquel Ferreira Ribeiro, Isabela David de Lima Damasceno, Sara Mabel Ancelmo Benvenuto
Este trabalho propõe uma análise comparativa entre as personagens do conto “Eveline”, de James Joyce, que se passa na Irlanda do início do século XX, e Eveline, protagonista do curta-metragem Válvula, dirigido por Sara Benvenuto (2020). O filme se passa em Iguatu, cidade do sertão do Ceará nos tempos atuais. Ambas as personagens contemplam as possibilidades de mudança em suas vidas, mas lidam com as amarras impostas pelo seu sexo e contexto sociocultural. Consideramos o conceito de adaptação (Hutcheon 2013), a noção de reescritura (Lefevere 2007) e o estudo da estática cinematográfica (Gardies et al. 2008). A análise foi conduzida em três perspectivas: o da análise literária, do processo de construção da adaptação fílmica e a partir de um olhar pedagógico. Concluímos que, embora haja semelhanças entre as narrativas, os contextos socioculturais da mulher dublinense dos anos 1900 e da mulher interiorana de 2020 se sobressaem.
This research addresses the under-explored role of translation as a form of activism in China, particularly in the context of online platforms. While online activism in China has garnered significant attention, the specific role of translation as a transformative practice has received limited scholarly focus. This study examines the translation campaign surrounding The Whistle Giver, a censored article pertaining to COVID-19, which involved a collaborative effort by non-professional translators who strategically rendered the original text into various unreadable forms. By employing translation as a tool of metramorphosis, these translators challenged the dominance of official discourse and asserted their individual agency in shaping public narratives. This case study demonstrates the potential of Web 2.0-enabled translation as a means of combating censorship and fostering grassroots ideological struggles. Web 2.0 platforms play a pivotal role in facilitating the collaborative nature of this translation campaign, allowing for the dissemination and sharing of the transformed texts. The interactive and participatory features of Web 2.0 platforms provide fertile ground for the emergence of a playful and ritualistic dimension to the translation process. By harnessing the power of Web 2.0-enabled translation, activists forge a collective identity, challenging any existing power structures and fostering collective memories. This form of activism transcends traditional boundaries, utilizing translation as a transformative practice to shape public discourse and effect social and political change. The study sheds light on the crucial role of non-professional translators, the transformative potential of translation as a tool for resistance, and the impact of Web 2.0 platforms in facilitating collaborative and participatory practices.
Automatic subtitling is the task of automatically translating the speech of audiovisual content into short pieces of timed text, i.e. subtitles and their corresponding timestamps. The generated subtitles need to conform to space and time requirements, while being synchronised with the speech and segmented in a way that facilitates comprehension. Given its considerable complexity, the task has so far been addressed through a pipeline of components that separately deal with transcribing, translating, and segmenting text into subtitles, as well as predicting timestamps. In this paper, we propose the first direct ST model for automatic subtitling that generates subtitles in the target language along with their timestamps with a single model. Our experiments on 7 language pairs show that our approach outperforms a cascade system in the same data condition, also being competitive with production tools on both in-domain and newly-released out-domain benchmarks covering new scenarios.
Digital pen technologies have enabled new forms of interpreting and provided new ways to train interpreting students in consecutive interpreting; however, digital pens also provide an opportunity to collect interpreting process data that can serve as a proxy for cognitive behavior. This article examines the extant literature on notetaking and cognition to advocate for using digital pens to document the notetaking process and describes several ways in which this notetaking data can be analyzed to understand interpreter behavior. In addition, several potential research avenues are outlined to enhance the methodological tools available to undertake interpreting process research in dialogue and consecutive interpreting.
The paper describes the BUT's speech translation systems. The systems are English$\longrightarrow$German offline speech translation systems. The systems are based on our previous works \cite{Jointly_trained_transformers}. Though End-to-End and cascade~(ASR-MT) spoken language translation~(SLT) systems are reaching comparable performances, a large degradation is observed when translating ASR hypothesis compared to the oracle input text. To reduce this performance degradation, we have jointly-trained ASR and MT modules with ASR objective as an auxiliary loss. Both the networks are connected through the neural hidden representations. This model has an End-to-End differentiable path with respect to the final objective function and also utilizes the ASR objective for better optimization. During the inference both the modules(i.e., ASR and MT) are connected through the hidden representations corresponding to the n-best hypotheses. Ensembling with independently trained ASR and MT models have further improved the performance of the system.