Hasil untuk "Animal culture"

Mohsinul Kabir, Tasnim Ahmed, Md Mezbaur Rahman et al.

Cross-cultural competence in large language models (LLMs) requires the ability to identify Culture-Specific Items (CSIs) and to adapt them appropriately across cultural contexts. Progress in evaluating this capability has been constrained by the scarcity of high-quality CSI-annotated corpora with parallel cross-cultural sentence pairs. To address this limitation, we introduce XCR-Bench, a Cross(X)-Cultural Reasoning Benchmark consisting of 4.9k parallel sentences and 1,098 unique CSIs, spanning three distinct reasoning tasks with corresponding evaluation metrics. Our corpus integrates Newmark's CSI framework with Hall's Triad of Culture, enabling systematic analysis of cultural reasoning beyond surface-level artifacts and into semi-visible and invisible cultural elements such as social norms, beliefs, and values. Our findings show that state-of-the-art LLMs exhibit consistent weaknesses in identifying and adapting CSIs related to social etiquette and cultural reference. Additionally, we find evidence that LLMs encode regional and ethno-religious biases even within a single linguistic setting during cultural adaptation. We release our corpus and code to facilitate future research on cross-cultural NLP.

F. Carichon, R. Rampa, G. Farnadi

Large Language Models (LLMs) are increasingly used to generate and shape cultural content, ranging from narrative writing to artistic production. While these models demonstrate impressive fluency and generative capacity, prior work has shown that they also exhibit systematic cultural biases, raising concerns about stereotyping, homogenization, and the erasure of culturally specific forms of expression. Understanding whether LLMs can meaningfully align with diverse cultures beyond the dominant ones remains a critical challenge. In this paper, we study cultural adaptation in LLMs through the lens of cooking recipes, a domain in which culture, tradition, and creativity are tightly intertwined. We build on the \textit{GlobalFusion} dataset, which pairs human recipes from different countries according to established measures of cultural distance. Using the same country pairs, we generate culturally adapted recipes with multiple LLMs, enabling a direct comparison between human and LLM behavior in cross-cultural content creation. Our analysis shows that LLMs fail to produce culturally representative adaptations. Unlike humans, the divergence of their generated recipes does not correlate with cultural distance. We further provide explanations for this gap. We show that cultural information is weakly preserved in internal model representations, that models inflate novelty in their production by misunderstanding notions such as creativity and tradition, and that they fail to identify adaptation with its associated countries and to ground it in culturally salient elements such as ingredients. These findings highlight fundamental limitations of current LLMs for culturally oriented generation and have important implications for their use in culturally sensitive applications.

Arijit Maji, Raghvendra Kumar, Akash Ghosh et al.

Language Models (LMs) are indispensable tools shaping modern workflows, but their global effectiveness depends on understanding local socio-cultural contexts. To address this, we introduce SANSKRITI, a benchmark designed to evaluate language models' comprehension of India's rich cultural diversity. Comprising 21,853 meticulously curated question-answer pairs spanning 28 states and 8 union territories, SANSKRITI is the largest dataset for testing Indian cultural knowledge. It covers sixteen key attributes of Indian culture: rituals and ceremonies, history, tourism, cuisine, dance and music, costume, language, art, festivals, religion, medicine, transport, sports, nightlife, and personalities, providing a comprehensive representation of India's cultural tapestry. We evaluate SANSKRITI on leading Large Language Models (LLMs), Indic Language Models (ILMs), and Small Language Models (SLMs), revealing significant disparities in their ability to handle culturally nuanced queries, with many models struggling in region-specific contexts. By offering an extensive, culturally rich, and diverse dataset, SANSKRITI sets a new standard for assessing and improving the cultural understanding of LMs.

Sougata Saha, Saurabh Kumar Pandey, Harshit Gupta et al.

In a rapidly globalizing and digital world, content such as book and product reviews created by people from diverse cultures are read and consumed by others from different corners of the world. In this paper, we investigate the extent and patterns of gaps in understandability of book reviews due to the presence of culturally-specific items and elements that might be alien to users from another culture. Our user-study on 57 book reviews from Goodreads reveal that 83\% of the reviews had at least one culture-specific difficult-to-understand element. We also evaluate the efficacy of GPT-4o in identifying such items, given the cultural background of the reader; the results are mixed, implying a significant scope for improvement. Our datasets are available here: https://github.com/sougata-ub/reading_between_lines

Yiman Chen, Linan Zhang, Xianshan Lin et al.

Hybrid breeding is a crucial approach in aquaculture for improving stock performance. This study systematically evaluated heterosis in hybrid offspring from a cross between the zig-zag eel (<i>Mastacembelus armatus</i>) and the tire track eel (<i>Mastacembelus favus</i>). The hybrids exhibited significantly superior growth performance and a higher survival rate than both parental populations. Physiological analyses revealed distinct advantages in digestive enzyme activity and intestinal tissue structure. Liver transcriptome sequencing revealed that most of the differentially expressed genes (DEGs) in the hybrids displayed non-additive expression patterns and were significantly enriched in pathways associated with growth hormone synthesis, insulin secretion and lipid metabolism. The upregulation of <i>igf2</i> and the downregulation of <i>socs1</i> were identified as key molecular drivers of enhanced growth. This multi-level analysis, spanning from phenotype to molecular mechanisms, sheds light on the heterogeneous basis of these hybrids and provides theoretical insights and practical guidance for breeding work with zig-zag eels and related species.

Federica Meluzzi

Gender norms are widely recognized as key determinants of persistent gender gaps in the labor market. However, our understanding of the drivers of gender norms, and their implications for preferences, remain lacking. This paper addresses this gap by examining how cultural assimilation from college peers influences women's early-career labor market decisions. For identification of causal effects, I exploit cross-cohort idiosyncratic variation in peers' geographical origins within Master's programs, combined with unique administrative and survey data covering the universe of college students in Italy. The main finding is that exposure to female classmates originating from areas with more egalitarian gender culture significantly increases women's labor supply, primarily through increased uptake of full-time jobs. A one standard deviation increase in peers' culture increases female earnings by 3.7%. The estimated peer effects are economically significant, representing more than a third of the gender earnings gap. Drawing on comprehensive data on students' job search preferences and newly collected data on their beliefs, I shed novel light on two distinct mechanisms driving peer influence: (1) shifts in preferences for non-pecuniary job attributes, and (2) social learning, particularly on the characteristics of the job offer distribution.

Jenna Kline, Austin O'Quinn, Tanya Berger-Wolf et al.

Platforms that run artificial intelligence (AI) pipelines on edge computing resources are transforming the fields of animal ecology and biodiversity, enabling novel wildlife studies in animals' natural habitats. With emerging remote sensing hardware, e.g., camera traps and drones, and sophisticated AI models in situ, edge computing will be more significant in future AI-driven animal ecology (ADAE) studies. However, the study's objectives, the species of interest, its behaviors, range, habitat, and camera placement affect the demand for edge resources at runtime. If edge resources are under-provisioned, studies can miss opportunities to adapt the settings of camera traps and drones to improve the quality and relevance of captured data. This paper presents salient features of ADAE studies that can be used to model latency, throughput objectives, and provision edge resources. Drawing from studies that span over fifty animal species, four geographic locations, and multiple remote sensing methods, we characterized common patterns in ADAE studies, revealing increasingly complex workflows involving various computer vision tasks with strict service level objectives (SLO). ADAE workflow demands will soon exceed individual edge devices' compute and memory resources, requiring multiple networked edge devices to meet performance demands. We developed a framework to scale traces from prior studies and replay them offline on representative edge platforms, allowing us to capture throughput and latency data across edge configurations. We used the data to calibrate queuing and machine learning models that predict performance on unseen edge configurations, achieving errors as low as 19%.

Keqi Wang, Sarah W. Harcum, Wei Xie

To advance understanding of cellular metabolism and reduce batch-to-batch variability in cell culture processes, this study introduces a multi-scale hybrid modeling framework designed to simulate and predict the dynamic behavior of CHO cell cultures undergoing metabolic phase transitions. The model captures dependencies across molecular, cellular, and macro-kinetic levels, accounting for variability in single-cell metabolic phases. It integrates three components: (i) a stochastic mechanistic model of single-cell metabolic networks, (ii) a probabilistic model of phase transitions, and (iii) a macro-kinetic model of heterogeneous population dynamics. This modular architecture enables flexible representation of process trajectories under diverse conditions and incorporates heterogeneous online (e.g., oxygen uptake, pH) and offline measurements (e.g., viable cell density, metabolite concentrations). Leveraging these data and single-cell insights, the framework predicts culture dynamics using only readily available online measurements and initial conditions, delivering accurate long-term forecasts of multivariate culture behavior and uncertainty-aware estimates of batch-to-batch variation. Overall, this work establishes a robust foundation for digital twin platforms and predictive bioprocess analytics, supporting systematic experimental design and process control to improve yield and production stability in biomanufacturing.

Julia Kharchenko, Tanya Roosta, Aman Chadha et al.

Large Language Models (LLMs) attempt to imitate human behavior by responding to humans in a way that pleases them, including by adhering to their values. However, humans come from diverse cultures with different values. It is critical to understand whether LLMs showcase different values to the user based on the stereotypical values of a user's known country. We prompt different LLMs with a series of advice requests based on 5 Hofstede Cultural Dimensions -- a quantifiable way of representing the values of a country. Throughout each prompt, we incorporate personas representing 36 different countries and, separately, languages predominantly tied to each country to analyze the consistency in the LLMs' cultural understanding. Through our analysis of the responses, we found that LLMs can differentiate between one side of a value and another, as well as understand that countries have differing values, but will not always uphold the values when giving advice, and fail to understand the need to answer differently based on different cultural values. Rooted in these findings, we present recommendations for training value-aligned and culturally sensitive LLMs. More importantly, the methodology and the framework developed here can help further understand and mitigate culture and language alignment issues with LLMs.

Tatik Hernawati, Suherni Susilowati, Tri Wahyu Suprayogi et al.

One of the many efforts to increase the quality of livestock genetics is through artificial insemination (AI). Other than increasing it, AI can be conducted to preserve semen. A successful AI is determined by seminal quality, therefore, a method to preserve semen for a longer storage time is needed. The method used is adding an extender that fulfills prerequisites for a semen extender such as coconut water combined with egg yolk citrate extender. Coconut water is rich in carbohydrates, protein, vitamins, and antioxidants while egg yolk contains lecithin. This study aims to find out the Pote buck spermatozoa quality stored in coconut water and egg yolk extender. This study uses three groups of treatments (T0: 0.1 ml semen + 0.9 ml egg yolk citrate, T1: 0.1 ml semen + 0.9 ml coconut water, and T2: 0.1 ml semen + egg yolk citrate (20%) + coconut water). All three of these are stored at 5oC and evaluated every day until day 5 of their motility, viability, intact plasma membrane, abnormality, and MDA level. Data analysis used is ANOVA and a further test called BNT is conducted if a significant difference is determined. No significant difference was found between T0 and T1 (p>0.05). The highest progressive motility, viability, and intact plasma membrane (%) among the three groups of treatments happened to be from T2. Meanwhile, a low percentage of spermatozoa abnormality and MDA level were also found in T2 with its extender being coconut water combined with egg yolk citrate. To conclude, the best extender for storing Pote buck semen is stored at 5oC is coconut water combined with egg yolk citrate extender.

Theethad Tangpakornsak, Phasamon Saisawart, Somchin Sutthigran et al.

The vertebral heart score (VHS) is the sum of the ratio of the cardiac dimensions to the number of thoracic vertebrae, starting from the fourth thoracic vertebra (T4) to the intervertebral disk space (IVS). Breed-specific VHSs, in most cases, were different from the original reference value. Characteristics of the thoracic vertebrae and IVS may influence this variation. This study was conducted to investigate the characteristics of the T4 and IVS on the thoracic radiographs of Corgis in comparison with other small-to-medium breed dogs to evaluate the Corgi-specific VHSs in healthy dogs. The ratio of the T4’s length/height (T4L/H) was significantly different among dog breeds but not the IVS between the T4 and T5. The T4L/H was highest in the Shih Tzu and lowest in Beagle dogs. The Corgi-specific VHS obtained from the ventrodorsal radiograph was significantly higher than that from the dorsoventral radiograph, but a significant difference was not observed between the right and left lateral radiographs. In contrast, the Corgi-specific VHS derived from the right lateral thoracic radiograph was significantly lower than the reference value. This may be correlated with the characteristics of the thoracic vertebrae of Corgis, which were slightly higher than those of the other breeds.

Enrico Volpe, Francesca Errani, Samuele Zamparo et al.

Aquaculture is a constantly growing sector. The intensification of fish production and the movement of aquatic animals could cause the spread of infectious diseases. Remarkably, the diffusion of viral agents represents the major bottleneck for finfish production, and viral encephalopathy and retinopathy (VER) is considered the most impacting disease for Mediterranean aquaculture. No effective therapies are available to contrast VER, and vaccination can be applied only in grow-out facilities. Hence, programs to minimize the sanitary risks in farms are paramount to implementing hygienic standards and biosecurity. This study aimed to evaluate the in vitro virucidal activity of a peroxy-acid disinfectant (Virkon^® S, DuPont, Sudbury, UK) towards the two NNV strains most widespread in the Mediterranean Sea. Remarkably, two protocols were applied to assess the virucidal activity under different conditions of use: the suspension test and the net test. The latter has been applied to evaluate the efficacy of the biocide on instruments, simulating the in-field application. The obtained results demonstrated the suitability of the tested biocide for NNV inactivation, being effective under some of the tested conditions. However, the presence of organic matter, the concentration of the product, and the application conditions can significantly affect the result of the disinfection procedure.

J.M. dos Santos Neto, C.M. Prom, A.L. Lock

ABSTRACT: We evaluated the effects of infusing an exogenous emulsifier (polysorbates-C18:1) either into the rumen or abomasum on fatty acid (FA) digestibility and production responses of lactating dairy cows. Nine ruminally cannulated multiparous Holstein cows (170 ± 13.6 d in milk) were assigned to a treatment sequence in replicated 3 × 3 Latin squares with 18-d periods consisting of 7 d of washout and 11 d of infusion. Treatments were abomasal infusions of water carrier only into the rumen and abomasum (control, CON), 30 g/d polysorbate-C18:1 (T80) infused into the rumen (RUM), or 30 g/d T80 infused into the abomasum (ABO). Emulsifiers were dissolved in water and delivered at 6-h intervals (total daily infusion was divided into 4 equal infusions per day). Cows were fed the same diet that contained [% diet dry matter (DM)] 32.2% neutral detergent fiber (NDF), 16.1% crude protein, 26.5% starch, and 3.41% FA (including 1.96% FA from a saturated FA supplement containing 28.0% C16:0 and 54.6% C18:0). Two orthogonal contrasts were evaluated: (1) the overall effect of T80 {CON vs. average of the T80 infusions [1/2 (ABO + RUM)]}, and (2) the effect of ABO versus RUM infusion. Compared with CON, infusing T80 increased the digestibilities of NDF (2.85 percentage units), total (4.35 percentage units), 16-carbon (3.25 percentage units), and 18-carbon FA (4.60 percentage units), and tended to increase DM digestibility and total and 18-carbon FA absorption. Compared with RUM, ABO decreased the intakes of total (28 g/d), 16-carbon (7 g/d), and 18-carbon FA (19 g/d); tended to increase the digestibility of total and 18-carbon FA; and had no effect on the absorption of total, 16-carbon, or 18-carbon FA. Production responses did not change among our treatments. In conclusion, infusing 30 g/d polysorbates-C18:1 increased NDF and total, 16-carbon, and 18-carbon FA digestibility. Compared with RUM, ABO tended to increase the digestibilities of total and 18-carbon FA; however, this may be related to the fact that ABO reduced the intakes of total, 16-carbon, and 18-carbon FA, not necessarily due to better emulsifying action per se. In summary, ABO and RUM both improved FA absorption.

Ziping Jiang, Paul L. Chazot, Richard Jiang

As the bridge between genetic and physiological aspects, animal behaviour analysis is one of the most significant topics in biology and ecological research. However, identifying, tracking and recording animal behaviour are labour intensive works that require professional knowledge. To mitigate the spend for annotating data, researchers turn to computer vision techniques for automatic label algorithms, since most of the data are recorded visually. In this work, we explore a variety of behaviour detection algorithms, covering traditional vision methods, statistical methods and deep learning methods. The objective of this work is to provide a thorough investigation of related work, furnishing biologists with a scratch of efficient animal behaviour detection methods. Apart from that, we also discuss the strengths and weaknesses of those algorithms to provide some insights for those who already delve into this field.

Marta Moya-Jódar, Giulia Coppiello, Juan Roberto Rodríguez-Madoz et al.

Each year, tens of thousands of people worldwide die of end-stage organ failure due to the limited availability of organs for use in transplantation. To meet this clinical demand, one of the last frontiers of regenerative medicine is the generation of humanized organs in pigs from pluripotent stem cells (PSCs) via blastocyst complementation. For this, organ-disabled pig models are needed. As endothelial cells (ECs) play a critical role in xenotransplantation rejection in every organ, we aimed to produce hematoendothelial-disabled pig embryos targeting the master transcription factor <i>ETV2</i> via CRISPR-Cas9-mediated genome modification. In this study, we designed five different guide RNAs (gRNAs) against the DNA-binding domain of the porcine <i>ETV2</i> gene, which were tested on porcine fibroblasts in vitro. Four out of five guides showed cleavage capacity and, subsequently, these four guides were microinjected individually as ribonucleoprotein complexes (RNPs) into one-cell-stage porcine embryos. Next, we combined the two gRNAs that showed the highest targeting efficiency and microinjected them at higher concentrations. Under these conditions, we significantly improved the rate of biallelic mutation. Hence, here, we describe an efficient one-step method for the generation of hematoendothelial-disabled pig embryos via CRISPR-Cas9 microinjection in zygotes. This model could be used in experimentation related to the in vivo generation of humanized organs.

Song-Hsien Lin, Guan-Ying Tsai, Meng-Jiy Wang et al.

Photodynamic therapy (PDT)-based cancer vaccines are shown to be more effective modalities for treating cancer in animal models compared to other methods used to generate cancer cell-derived vaccines. The higher efficacy seems to stem from the generation of cell membrane nanovesicles or fragments that carry both cancer cell-specific antigens and high surface content of damage-associated molecular pattern (DAMP) molecules induced by oxidative stress. To develop more effective cancer vaccines in this direction, we explored the generation of cancer vaccines by applying different sources of oxidative stress on cancer cell cultures followed by spontaneous release or filter extrusions to produce cancer cell-derived DAMP-expressing nanovesicles. Through an in-vitro test based on the co-culture of cancer cells and macrophages, it was found that the nanovesicle vaccines generated by H<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mrow></mrow><mn>2</mn></msub></semantics></math></inline-formula>O<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mrow></mrow><mn>2</mn></msub></semantics></math></inline-formula> are as effective as those generated by PDT in diminishing cancer cell culture masses, providing a simpler way to manufacture vaccines. In addition, the nanovesicle vaccines produced by filter extrusion are as potent as those produced by spontaneous release, rendering a more stable way for vaccine production.

Katerina Papantoniou, Panagiotis Papadakos, Theodore Patkos et al.

Deception detection is a task with many applications both in direct physical and in computer-mediated communication. Our focus is on automatic deception detection in text across cultures. We view culture through the prism of the individualism/collectivism dimension and we approximate culture by using country as a proxy. Having as a starting point recent conclusions drawn from the social psychology discipline, we explore if differences in the usage of specific linguistic features of deception across cultures can be confirmed and attributed to norms in respect to the individualism/collectivism divide. We also investigate if a universal feature set for cross-cultural text deception detection tasks exists. We evaluate the predictive power of different feature sets and approaches. We create culture/language-aware classifiers by experimenting with a wide range of n-gram features based on phonology, morphology and syntax, other linguistic cues like word and phoneme counts, pronouns use, etc., and token embeddings. We conducted our experiments over 11 datasets from 5 languages i.e., English, Dutch, Russian, Spanish and Romanian, from six countries (US, Belgium, India, Russia, Mexico and Romania), and we applied two classification methods i.e, logistic regression and fine-tuned BERT models. The results showed that our task is fairly complex and demanding. There are indications that some linguistic cues of deception have cultural origins, and are consistent in the context of diverse domains and dataset settings for the same language. This is more evident for the usage of pronouns and the expression of sentiment in deceptive language. The results of this work show that the automatic deception detection across cultures and languages cannot be handled in a unified manner, and that such approaches should be augmented with knowledge about cultural differences and the domains of interest.

Wei-Lin Hsiao, Kristen Grauman

Fashion is intertwined with external cultural factors, but identifying these links remains a manual process limited to only the most salient phenomena. We propose a data-driven approach to identify specific cultural factors affecting the clothes people wear. Using large-scale datasets of news articles and vintage photos spanning a century, we present a multi-modal statistical model to detect influence relationships between happenings in the world and people's choice of clothing. Furthermore, on two image datasets we apply our model to improve the concrete vision tasks of visual style forecasting and photo timestamping. Our work is a first step towards a computational, scalable, and easily refreshable approach to link culture to clothing.

Jone Ibarruri, Idoia Goiri, Marta Cebrián et al.

This research aimed to evaluate in vitro organic matter digestibility, fermentation characteristics and methane production of fruit and vegetable discards processed by solid state fermentation (SSF) by <i>Rhizopus</i> sp. Mixtures were composed of approximately 28% citric fruits, 35% other fruits and 37% vegetables. Fruit and vegetables were processed and fermented to obtain a stabilized product. Nutritional characterization and in vitro ruminal fermentation tests were performed to determine the effect of fungal bioconversion on digestibility, end products and gas production kinetics. Results indicate that SSF reduced organic matter and reducing sugars, while it increased crude protein and neutral detergent fiber, acid detergent fiber and neutral detergent insoluble protein. The in vitro gas production showed that SSF led to a reduction of the organic matter digestibility (<i>p</i> < 0.001), short chain fatty acids (SCFA; <i>p</i> = 0.003) and CH₄ (<i>p</i> = 0.002). SSF reduced the gas production from the insoluble fraction (<i>p</i> = 0.001), without modifying the production rate (<i>p</i> = 0.676) or the lag time (<i>p</i> = 0.574). Regarding SCFA profile, SSF increased acetic (<i>p</i> = 0.020) and decreased propionic (<i>p</i> = 0.004) and butyric (<i>p</i> = 0.006) acids proportions, increasing acetic to propionic (<i>p</i> = 0.008) and acetic plus butyric to propionic (<i>p</i> = 0.011) ratios. SSF succeeded in obtaining a stabilized material enriched in protein, but at the expense of a reduction of protein availability and organic matter digestibility. These changes should be considered before including them in a ruminant’s rations.

Julia Ramos Sampaio, Rodrigo Pedro Soares, Thiago Doria Barral et al.

Lipophosphoglycan (LPG), when used as an ELISA target, confers high specificity and sensitivity to the detection of Leishmania infantum antibodies in dogs. Glycoconjugates are economically viable because the yield is very high after extraction/purification. In addition, they are very stable, which allows their use in point-of-care testing without special storage conditions. During the glycoconjugate extraction, a glycoinositolphospholipid (GIPL)-enriched fraction is obtained in similar quantities as LPG. Since GIPLs can be extracted from the same parasite pellet as LPGs, this work aimed to evaluate the immune recognition of GIPLs by Leishmania infantum-infected dogs and its use for canine leishmaniasis (CanL) immunodiagnosis. Like LPG, GIPLs were recognized by sera from L. infantum-infected dogs, but with less sensitivity (83.8%). However, 80% (16/20) of subclinically infected dogs were detected as positive in the assay. Different from LPG, the GIPL-based assay achieved a lower specificity (73.7%) and cross-reactions occurred with T. cruzi and L. braziliensis-infected dogs. Although GIPLs exhibited a similar performance to LPG for subclinically L. infantum-infected dogs, the occurrence of cross-reactivities with other protozoa and a lower sensitivity hinders its use for an immunodiagnostic test. In places where those diseases do not co-exist such as in the Mediterranean region, its use for subclinically dogs could be an alternative.