When LLM-based multi-agent systems disagree, current practice treats this as noise to be resolved through consensus. We propose it can be signal. We focus on hate speech moderation, a domain where judgments depend on cultural context and individual value weightings, producing high legitimate disagreement among human annotators. We hypothesize that convergent disagreement, where agents reason similarly but conclude differently, indicates genuine value pluralism that humans also struggle to resolve. Using the Measuring Hate Speech corpus, we embed reasoning traces from five perspective-differentiated agents and classify disagreement patterns using a four-category taxonomy based on reasoning similarity and conclusion agreement. We find that raw reasoning divergence weakly predicts human annotator conflict, but the structure of agent discord carries additional signal: cases where agents agree on a verdict show markedly lower human disagreement than cases where they do not, with large effect sizes (d>0.8) surviving correction for multiple comparisons. Our taxonomy-based ordering correlates with human disagreement patterns. These preliminary findings motivate a shift from consensus-seeking to uncertainty-surfacing multi-agent design, where disagreement structure - not magnitude - guides when human judgment is needed.
Summary: Antiretroviral therapy (ART) inhibits human immunodeficiency virus (HIV) replication to maintain undetectable viral loads but does not result in a cure. Due to the significant challenges of lifelong ART for many, there is a strong interest in therapeutic strategies that result in a cure. Recent clinical trials have shown that administration of broadly neutralizing antibodies (bnAbs) in the presence of some viremia can lead to ART-free viral control in some people; however, the underlying mechanisms are unclear. Our computational modeling shows that bnAbs administered in the presence of some viremia promote the evolution of autologous antibodies (aAbs) that target diverse epitopes of HIV envelope proteins. This “net” of polyclonal aAbs could lead to sustained control because viral escape from such a response would require mutations in multiple epitopes. Our results provide a common mechanistic framework underlying recent clinical observations upon bnAb/ART therapy and could inform future trials.
Elena A. Leonova, Albina A. Bakhova, Irina L. Cherednik
et al.
The article presents current views of scientists on the physiological effects of oxytocin, its influence on social behaviour and on the psychological state of humans and animals. It is known that at the early stages of invertebrate evolution, signals were transmitted through oxytocin-like substances. In the reviewed papers, the development of oxytocin receptors in vertebrates is considered. Facts are presented about the control the oxytocinergic neurotransmitter system and its homologues exert over social behaviour, both in mammals and other vertebrates. The evolution of communicative behaviour associated with the influence of oxytocin is traced. A number of studies have demonstrated a link of the oxytocinergic neurotransmitter system with aggressive behaviour. A correlation has been established between the level of expression of oxytocin receptors in the brain and partner preference. The effect of oxytocin on the formation of social bonds between humans and domestic animals has been described. In domestic animals, oxytocin concentrations have been shown to be positively correlated with the number of contacts with the owner. The anti-stress and anxiolytic effects of this neuropeptide blocking the effects of the main stress hormone, cortisol, have been detected. Thus, oxytocin has the potential for treatment of stress and its consequences. The oxytocinergic system can modulate the mechanisms of emotions and can be used to alleviate social dysfunction in mental illnesses, including schizophrenia and autism spectrum disorders. The studied effects of oxytocin confirm the special role of this ancient hormone in anthropogenesis and its importance for human health and socialization, as well as demonstrate its potential use in pharmacotherapy for a number of pathologies.
ABSTRACT: Sarcotheca macrophylla Blume (Oxalidaceae), an endemic plant species to Borneo, consists of short-styled (S-morph) and long-styled (L-morph) flowers, referred to as heterostyly. However, the character states related to heterostyly have not been fully disclosed in S. macrophylla. Additionally, S. macrophylla has a reproductive issue due to a lack of seeds in fruits. This research aims to examine the floral characteristics and detail of sexual organs (androecium and gynoecium) in this species and to analyze the consequences of heterostyly on its natural pollination. We collect flowers from 34 individuals and investigate the inaccuracy index and all of floral character states. We used light and stereo microscopes to evaluate the heterostyly syndrome through morpho-anatomical observations of the flowers. Our results reveal that the two floral types (S-morph and L-morph) of S. macrophylla possess reciprocal herkogamy with ancillary pollen polymorphism in palynological characters, differences in pollen number, and pollen viability. These characters suggest that S. macrophylla has distyly type. Furthermore, the reproductive organs of the floral morphs appear to contribute differently to reproductive fitness. The S-morph flowers demonstrate enhanced male function through higher pollen production and viability, while the L-morph flowers exhibit enhanced female function through floral architecture that promotes effective pollen reception.
Ambiguity in natural language instructions poses significant risks in safety-critical human-robot interaction, particularly in domains such as surgery. To address this, we propose a framework that uses Large Language Models (LLMs) for ambiguity detection specifically designed for collaborative surgical scenarios. Our method employs an ensemble of LLM evaluators, each configured with distinct prompting techniques to identify linguistic, contextual, procedural, and critical ambiguities. A chain-of-thought evaluator is included to systematically analyze instruction structure for potential issues. Individual evaluator assessments are synthesized through conformal prediction, which yields non-conformity scores based on comparison to a labeled calibration dataset. Evaluating Llama 3.2 11B and Gemma 3 12B, we observed classification accuracy exceeding 60% in differentiating ambiguous from unambiguous surgical instructions. Our approach improves the safety and reliability of human-robot collaboration in surgery by offering a mechanism to identify potentially ambiguous instructions before robot action.
Florian Mai, David Kaczér, Nicholas Kluge Corrêa
et al.
Two core challenges of alignment are 1) scalable oversight and 2) accounting for the dynamic nature of human values. While solutions like recursive reward modeling address 1), they do not simultaneously account for 2). We sketch a roadmap for a novel algorithmic framework that trains a superhuman reasoning model to decompose complex tasks into subtasks that are still amenable to human-level guidance. Our approach relies on what we call the part-to-complete generalization hypothesis, which states that the alignment of subtask solutions generalizes to the alignment of complete solutions. We advocate for the need to measure this generalization and propose ways to improve it in the future.
Social identity theory (SIT) and social categorization theory (SCT) are two facets of the social identity approach (SIA) to understanding social phenomena. SIT and SCT are models that describe and explain how people interact with one another socially, connecting the individual to the group through an understanding of underlying psychological mechanisms and intergroup behaviour. SIT, originally developed in the 1970s, and SCT, a later, more general offshoot, have been broadly applied to a range of social phenomena among people. The rise of increasingly social machines embedded in daily life has spurned efforts on understanding whether and how artificial agents can and do participate in SIA activities. As agents like social robots and chatbots powered by sophisticated large language models (LLMs) advance, understanding the real and potential roles of these technologies as social entities is crucial. Here, I provide a primer on SIA and extrapolate, through case studies and imagined examples, how SIT and SCT can apply to artificial social agents. I emphasize that not all human models and sub-theories will apply. I further argue that, given the emerging competence of these machines and our tendency to be taken in by them, we experts may need to don the hat of the uncanny killjoy, for our own good.
In this study, we trace the evolution of a data work team in an artificial intelligence (AI) startup in India. By bringing attention to data work, which is the indispensable work of preparing annotated datasets for training AI systems, conducted within a formal organisational set up, we underline: 1. how organisational approaches adopted to balance investor and client preferences shape work arrangements and spatial division of the data workers; 2. how relations between the data team and the ‘core’ technical team serve to invisibilise human labour in the production of AI; and 3. how increasing codification of data work leads to devaluation of data work within the organisation and deskilling of young data workers at large, making them vulnerable in choosing a meaningful career path of their choice. In tracing this trajectory of displacement of data workers employed in a formal sector, we show that the prevalent characterisation of data work as being invisible or precarious is not inherent to AI nor inevitable in its labour processes. Rather, it is produced through the specific embedding of AI production within the political economy of startup capitalism. Through this, we seek to recentre the discourse on AI and future-of-work away from deterministic projections of AI’s impact on work and towards the specific labour processes of AI and its implications for the skills and career trajectories of a young and growing workforce in the Global South.
Influenza A/H9 viruses circulate worldwide in wild and domestic avian species, continuing to evolve and posing a zoonotic risk. A substantial increase in human infections with A/H9N2 subtype avian influenza viruses (AIVs) and the emergence of novel reassortants carrying A/H9N2-origin internal genes has occurred in recent years. Different names have been used to describe the circulating and emerging A/H9 lineages. To address this issue, an international group of experts from animal and public health laboratories, endorsed by the WOAH/FAO Network of Expertise on Animal Influenza, has created a practical lineage classification and nomenclature system based on the analysis of 10,638 hemagglutinin sequences from A/H9 AIVs sampled worldwide. This system incorporates phylogenetic relationships and epidemiologic characteristics designed to trace emerging and circulating lineages and clades. To aid in lineage and clade assignment, an online tool has been created. This proposed classification enables rapid comprehension of the global spread and evolution of A/H9 AIVs.
<p>Due to a large number of automatic auroral camera systems on the ground, image data analysis requires more efficiency than what human expert visual inspection can provide. Furthermore, there is no solid consensus on how many different types or shapes exist in auroral displays. We report the first attempt to classify auroral morphological forms by an unsupervised learning method on an image set that contains both nightside and dayside aurora. We used 6 months of full-colour auroral all-sky images captured at a high-Arctic observatory on Svalbard, Norway, in 2019–2020. The selection of images containing aurora was performed manually. These images were then input into a convolutional neural network called SimCLR for feature extraction. The clustered and fused features resulted in 37 auroral morphological clusters. In the clustering of auroral image data with two different time resolutions, we found that the occurrence of 8 clusters strongly increased when the image cadence was high (24 s), while the occurrence of 14 clusters experienced little or no change with changes in input image cadence. We therefore investigated the temporal evolution of a group of eight “active aurora” clusters. Time periods for which this active aurora persisted for longer than two consecutive images with a maximum cadence of 6 min coincided with ground-magnetic deflections, and their occurrence was found to maximize around magnetic midnight. The active aurora onsets typically included vortical auroral structures and equivalent current patterns typical for substorms. Our findings therefore suggest that our unsupervised image clustering method can be used to detect auroral breakups in ground-based image datasets with a temporal accuracy determined by the image cadence.</p>
Davood Anvari, Shirzad Gholami, Adel Spotin
et al.
Background: Current study was designed to provide a better insight into the circulating genotypes, genetic diversity, and population structure of Echinococcus spp. between southeast of Iran and Pakistan.
Methods: From Jun 2020 to Dec 2020, 46 hydatid cysts were taken from human (n: 6), camel (n: 10), goat (n: 10), cattle (n: 10) and sheep (n: 10) in various cities of Sistan and Baluchestan Province of Iran, located at the neighborhood of Pakistan. DNA samples were extracted, amplified, and subjected to sequence analysis of cox1 and nad1 genes. Results: The phylogeny inferred by the Maximum Likelihood algorithm indicated that G1 genotype (n: 19), G3 genotype (n: 14) and G6 genotype (n: 13) assigned into their specific clades. The diversity indices showed a moderate (nad1: Hd: 0.485) to high hap- lotype diversity (cox1: Hd: 0.867) of E. granulosus s.s. (G1/G3) and low nucleotide diver- sity. The negative value of Tajima’s D and Fu’s Fs test displayed deviation from neutrali- ty indicating a recent population expansion. A parsimonious network of the haplotypes of cox1 displayed star-like features in the overall population containing IR9/PAK1/G1, IR2/PAK2/G3 and IR18/G6 as the most common haplotypes. A pairwise fixation index (Fst) indicated that E. granulosus s.s. populations are genetically moderate differen- tiated between southeast of Iran and Pakistan. The extension of haplotypes PAK18/G1 (sheep) and PAK26/G1 (cattle) toward Iranian haplogroup revealed that there is dawn of Echinococcus flow due to a transfer of alleles between mentioned populations through transport of livestock or their domestication.
Conclusion: The current findings strengthen our knowledge concerning the evolution- ary paradigms of E. granulosus in southeastern borders of Iran and is effective in control- ling of hydatidosis.
Geoparks, as sustainable tourism products, embody a range of values and functions aimed at mitigating the negative impacts of tourism on the landscape. The current increasing trend in establishing and supporting geoparks has extended to the territory of the Slovak Republic, which possesses valuable natural potential. However, effectively harnessing this potential requires a clearly defined management structure aligned with the goals and mission of each geopark. The paper presents the operational aspects of geoparks in Slovakia, providing a comprehensive overview of geopark management followed by a subsequent evaluation. Assessing the management approach yields valuable insights into the ongoing development of Slovak geoparks, serving as a stepping stone for their further advancement. These geoparks are in a state of continual evolution, demanding significant support to ensure their efficient functioning. The study establishes clear quality management criteria for optimal staffing. The main objective of the paper is to demonstrate the need for the professionalization of human resources in geopark management. Additionally, the article concludes with a model for a strategic approach to human resources management, covering all its areas. Geoparks, as a modern product of geotourism and territorial management, represent innovative strategies for protecting and conserving geodiversity. They are closely intertwined with knowledge dissemination, education, and fostering a broader appreciation for the Earth’s value. Moreover, they play a vital role in the sustainable development and preservation of territories.
Summary: The basal breast cancer subtype is enriched for triple-negative breast cancer (TNBC) and displays consistent large chromosomal deletions. Here, we characterize evolution and maintenance of chromosome 4p (chr4p) loss in basal breast cancer. Analysis of The Cancer Genome Atlas data shows recurrent deletion of chr4p in basal breast cancer. Phylogenetic analysis of a panel of 23 primary tumor/patient-derived xenograft basal breast cancers reveals early evolution of chr4p deletion. Mechanistically we show that chr4p loss is associated with enhanced proliferation. Gene function studies identify an unknown gene, C4orf19, within chr4p, which suppresses proliferation when overexpressed—a member of the PDCD10-GCKIII kinase module we name PGCKA1. Genome-wide pooled overexpression screens using a barcoded library of human open reading frames identify chromosomal regions, including chr4p, that suppress proliferation when overexpressed in a context-dependent manner, implicating network interactions. Together, these results shed light on the early emergence of complex aneuploid karyotypes involving chr4p and adaptive landscapes shaping breast cancer genomes.
Olga Vladimirovna Semenova, Alexey Alexeevich Brazhnikov, Marina Lvovna Butovskaya
In this paper, we address the evolutionary dynamic of parental roles using game theory. The main purpose of the article was to expand a classical list of evolutionary dynamic parental conflicts by adding some important cases which hitherto have not been intensively studied. Our models are apt to deliver some novel insights into the evolution of parental care. We also introduced several hypothetical events that served as illustrations of an arising alteration in cost-benefits for both parents and simulated a subsequent evolutionary endpoint. Our models revealed that evolutionary outcomes for reproductive decisions of both parents could be completely predicted by certain payoff matrices, which serve as proxies for a Darwinian fitness gain. In this sense, the result of a frequency-dependent selection on reproductive traits would inevitably depend on fitness costs and benefits arising for both parents in various circumstances. We demonstrated that population division could be a plausible evolutionary consequence for any human mating game where ‘reproductive defection’ represents the best response to any action by the reproductive opponent. We conclude that future evolutionary studies of human reproductive behavior should be more oriented on estimating a sex-biased asymmetry in potential fitness gains obtained by cooperative and deceptive parents in diverse environments and cultures.
Humans have internal models of robots (like their physical capabilities), the world (like what will happen next), and their tasks (like a preferred goal). However, human internal models are not always perfect: for example, it is easy to underestimate a robot's inertia. Nevertheless, these models change and improve over time as humans gather more experience. Interestingly, robot actions influence what this experience is, and therefore influence how people's internal models change. In this work we take a step towards enabling robots to understand the influence they have, leverage it to better assist people, and help human models more quickly align with reality. Our key idea is to model the human's learning as a nonlinear dynamical system which evolves the human's internal model given new observations. We formulate a novel optimization problem to infer the human's learning dynamics from demonstrations that naturally exhibit human learning. We then formalize how robots can influence human learning by embedding the human's learning dynamics model into the robot planning problem. Although our formulations provide concrete problem statements, they are intractable to solve in full generality. We contribute an approximation that sacrifices the complexity of the human internal models we can represent, but enables robots to learn the nonlinear dynamics of these internal models. We evaluate our inference and planning methods in a suite of simulated environments and an in-person user study, where a 7DOF robotic arm teaches participants to be better teleoperators. While influencing human learning remains an open problem, our results demonstrate that this influence is possible and can be helpful in real human-robot interaction.