Jesper Milàn, Sten Lennart Jakobsen, Bent Erik Kramer Lindow
A fragment of pterosaur finger bone was found in the chalk in the uppermost Maastrichtian, Højerup Member of the Møns Klint Formation strata of Holtug quarry at the UNESCO World Heritage site Stevns Klint. This represents the first record of this group from the chalk of Denmark. The specimen is identified a fragment of a left proximal phalanx 1 of digit IV by comparison with similar elements showing the overall three-pronged expression of the posterior, ventral and olecranon processes. The dimensions of the specimen shows that small-bodied pterosaurs with a wingspan of less than 50 cm persisted through to the last 50 000–60 000 years of the Cretaceous. It overlaps in size with contemporaneous birds, rejecting previous hypotheses that Late Cretaceous pterosaurs and birds avoided competition through size-based niche partitioning.
The detailed dental anatomy of sauropod mamenchisaurids remains largely unexplored. Here, we describe a well-preserved isolated sauropod tooth from the Upper Jurassic Qigu Formation of the Turpan-Hami Basin, Xinjiang, using high-resolution micro-CT and three-dimensional reconstruction to investigate its internal anatomy. This tooth exhibits a unique combination of features of Mamenchisauridae, specifically the presence of marginal denticles restricted to the mesial margin, and a circular distolingual boss. CT scan data provide novel insights into mamenchisaurid dental anatomy and present the first three-dimensional enamel distribution in sauropod teeth. The lingual ridge forms from thickened enamel and dentine, whereas the lingual boss arises solely from dentine expansion. A labiolingual enamel thickness asymmetry appears in the apical region, convergent with certain neosauropods. The pulp cavity shows a distinct volumetric transition, expanding basally into a bulbous root canal and appearing as a labiolingually compressed lamina structure in the crown. Taxonomic comparison indicates that this tooth represents a mamenchisaurid lineage distinct from the previously only known sauropod tooth from the Qigu Formation. Our study supports a diverse sauropod assemblage in the Qigu Formation and provides new anatomical evidence for understanding sauropod dental evolution.
Faezeh Vahedi, Morteza Memari, Ramtin Tabatabaei
et al.
Nonverbal behaviors, particularly gaze direction, play a crucial role in enhancing effective communication in social interactions. As social robots increasingly participate in these interactions, they must adapt their gaze based on human activities and remain receptive to all cues, whether human-generated or not, to ensure seamless and effective communication. This study aims to increase the similarity between robot and human gaze behavior across various social situations, including both human and non-human stimuli (e.g., conversations, pointing, door openings, and object drops). A key innovation in this study, is the investigation of gaze responses to non-human stimuli, a critical yet underexplored area in prior research. These scenarios, were simulated in the Unity software as a 3D animation and a 360-degree real-world video. Data on gaze directions from 41 participants were collected via virtual reality (VR) glasses. Preprocessed data, trained two neural networks-LSTM and Transformer-to build predictive models based on individuals' gaze patterns. In the animated scenario, the LSTM and Transformer models achieved prediction accuracies of 67.6% and 70.4%, respectively; In the real-world scenario, the LSTM and Transformer models achieved accuracies of 72% and 71.6%, respectively. Despite the gaze pattern differences among individuals, our models outperform existing approaches in accuracy while uniquely considering non-human stimuli, offering a significant advantage over previous literature. Furthermore, deployed on the NAO robot, the system was evaluated by 275 participants via a comprehensive questionnaire, with results demonstrating high satisfaction during interactions. This work advances social robotics by enabling robots to dynamically mimic human gaze behavior in complex social contexts.
Automated decision systems increasingly rely on human oversight to ensure accuracy in uncertain cases. This paper presents a practical framework for optimizing such human-in-the-loop classification systems using a double-threshold policy. Conventional classifiers usually produce a confidence score and apply a single cutoff, but our approach uses two thresholds (a lower and an upper) to automatically accept or reject high-confidence cases while routing ambiguous instances to human reviewers. We formulate this problem as an optimization task that balances system accuracy against the cost of human review. Through analytical derivations and Monte Carlo simulations, we show how different confidence score distributions impact the efficiency of human intervention and reveal regions of diminishing returns, where additional review yields minimal benefit. The framework provides a general, reproducible method for improving reliability in any decision pipeline requiring selective human validation, including applications in entity resolution, fraud detection, medical triage, and content moderation.
The fossil record of the Palaeozoic echinoderm class Soluta
suggests they originated in the Miaolingian (middle Cam
brian) of Laurentia as permanently attached suspension
feeders, demonstrating a stepwise shift towards vagility in
successive strata. Here, we report a new specimen of Pah
vanticystis cf. utahensis associated with three putative juve
niles interpreted as belonging to the same species. We inter
pret this as evidence of facultative attachment in juveniles
of Pahvanticystis, which had not previously been reported
in this taxon, but is known in the earlier genus Castericystis.
Our findings indicate that attachment as a juvenile was
more widespread in solutans than previously thought.
Abstract Otolith finds in situ are important for connecting the independent data sets of articulated fish skeletons and isolated otoliths in the fossil record. Here we describe an otolith in situ retrieved from a skeleton of the stromateid Pinichthys shirvanensis Bannikov, 2021, which was collected from the Tarkhanian (Langhian, Middle Miocene) from the prolific Pshekha River locality in the Krasnodar Region of the northern Caucasus, Russia. It represents the first find of an otolith in situ made in a fossil stromateoid skeleton. The extracted otolith is compared to fossil otolith-based species allocated to the genus Pampus. One of these species is being reallocated in the process to Pinichthys: Pinichthys steurbauti (Schwarzhans, 1994) from the late Oligocene to early Middle Miocene in the North Sea Basin.
Halfdan Nordahl Fundal, Johannes Eide Rambøll, Karsten Olsen
Human-AI interactions are increasingly part of everyday life, yet the interpersonal dynamics that unfold during such exchanges remain underexplored. This study investigates how emotional alignment, semantic exploration, and linguistic innovation emerge within a collaborative storytelling paradigm that paired human participants with a large language model (LLM) in a turn-taking setup. Over nine days, more than 3,000 museum visitors contributed to 27 evolving narratives, co-authored with an LLM in a naturalistic, public installation. To isolate the dynamics specific to human involvement, we compared the resulting dataset with a simulated baseline where two LLMs completed the same task. Using sentiment analysis, semantic embeddings, and information-theoretic measures of novelty and resonance, we trace how humans and models co-construct stories over time. Our results reveal that affective alignment is primarily driven by the model, with limited mutual convergence in human-AI interaction. At the same time, human participants explored a broader semantic space and introduced more novel, narratively influential contributions. These patterns were significantly reduced in the simulated AI-AI condition. Together, these findings highlight the unique role of human input in shaping narrative direction and creative divergence in co-authored texts. The methods developed here provide a scalable framework for analysing dyadic interaction and offer a new lens on creativity, emotional dynamics, and semantic coordination in human-AI collaboration.
Amidst the race to create more intelligent machines there is a risk that we will rely on AI in ways that reduce our own agency as humans. To reduce this risk, we could aim to create tools that prioritize and enhance the human role in human-AI interactions. This paper outlines a human-centered augmented reasoning paradigm by 1. Articulating fundamental principles for augmented reasoning tools, emphasizing their ergonomic, pre-conclusive, directable, exploratory, enhancing, and integrated nature; 2. Proposing a 'many tasks, many tools' approach to ensuring human influence and control, and 3. Offering examples of interaction modes that can serve as bridges between human reasoning and AI algorithms.
The idea of augmented or hybrid intelligence offers a compelling vision for combining human and AI capabilities, especially in tasks where human wisdom, expertise, or common sense are essential. Unfortunately, human reasoning can be flawed and shortsighted, resulting in adverse individual impacts or even long-term societal consequences. While strong efforts are being made to develop and optimize the AI aspect of hybrid reasoning, the real urgency lies in fostering wiser and more intelligent human participation. Tools that enhance critical thinking, ingenuity, expertise, and even wisdom could be essential in addressing the challenges of our emerging future. This paper proposes the development of generative AI-based tools that enhance both the human ability to reflect upon a problem as well as the ability to explore the technical aspects of it. A high-level model is also described for integrating AI and human capabilities in a way that centralizes human participation and control.
Abstract Palaeontology has seen widespread and growing use of machine learning to classify and analyse large datasets of fossils. However, palaeontology is a challenging field in which to apply machine learning. Datasets may be small or unlabelled, images may be complex and different from standard datasets and palaeontologists may lack specialist training and access to necessary computational resources. We show how these challenges can be addressed by utilising recent developments in self-supervised learning (SSL). Using a frozen DINOv3 feature extractor and a simple linear classifier, with reduced data, we can achieve comparable results to literature benchmarks using Convolutional Neural Networks (CNNs), the previous standard, when classifying fossil tracks, pollen, radiolaria, foraminifera and a dataset of diverse fossil images. Additionally, the rich feature vectors generated by the model can be used for few-shot learning, unsupervised clustering and quantification of disparity. Using state-of-the-art self-supervised methods increases accessibility by reducing code, compute and data required. It also maintains accuracy, while increasing reproducibility by reducing parameters and allowing simple future-proof model agnostic pipelines which may become the new standard approach in palaeontology.
La especie de conodonte Tripodus laevis se ha utilizado como conodonte guía para marcar el límite inferior del Ordovícico Medio en los cuadros bioestratigráficos de conodontes de Midcontinent y Precordillera, a pesar de que T. laevis o especímenes asignables a esta especie se registran desde el Floiano inferior hasta el Darriwiliano medio. Además, esta especie de conodonte tiene una historia taxonómica compleja. Se ha ubicado en diferentes familias de conodontes y su aparato ha estado compuesto por distintos morfotipos de elementos conodontales. En la presente contribución se realizó una revisión crítica sobre la historia taxonómica de T. laevis, así como de su valor bioestratigráfico como especie guía para señalar el límite inferior del Dapingiano. Con base en este análisis, proponemos que T. laevis sensu lato (s. l.) no debe ser considerada una especie válida y evitar el uso de la misma en la Precordillera argentina. En consecuencia, se sugiere discontinuar la primera aparición (first appearence datum) de T. laevis s. l. como un marcador para el límite inferior del Ordovícico Medio en esta región. A la luz de esta propuesta, sugerimos el uso de un nuevo cuadro bioestratigráfico de conodontes para el Dapingiano de la Precordillera argentina.
Abstract The Triassic is considered a crucial period for the establishment of the modern insect fauna and fossil records from this period are fundamental for understanding the real impact that the end Permian Mass Extinction events had on these animals. Here, we review the insect fossils from one of the main deposits of this period in the world, Monte San Giorgio, which is considered one of the nine main insect Fossillagerstätten. In this Lagerstätte, located on the border between Switzerland and Italy, a total of 273 fossil insects have been collected in five localities. The fossils found in Val Mara site D, one of the two richest insect fossils sites of Monte San Giorgio, present peculiar features, such as extraordinary sizes and phosphatisation of internal tissues revealing fine internal details. In contrast, the Val Mara site VM 12 fossil record (248 specimens) is dominated by small to medium size insects, usually almost intact, preserving details such as setae on wings and compound eyes. Besides these exceptional features, these fossil insects are of extreme evolutionary importance, since they represent the first or the last occurrence for their lineage. In this regard, their use to calibrate nodes in a phylogenomic dating analysis led to backdating the origin of many insect lineages, including Diptera and Heteroptera. Up to now, a total of five species from Monte San Giorgio have been formally described, belonging to the orders Archaeognatha (†Monura and Machilidae), Ephemeroptera, Hemiptera (Tingidae) and Coleoptera (Adephaga). A further species, Merithone laetitiae (†Permithonidae) gen. et sp. nov., whose fossil is included among the recent findings in Val Mara site VM 12, is described in the present work.
Abstract Our new study of the species originally included in the genus Allolepidotus led to the taxonomic revision of the halecomorph species from the Triassic of Perledo, Italy. The morphological variation revealed by the analysis of the type material is sufficient to confirm four different taxa represented in the Perledo Formation. We correct the misunderstanding about the type species of Allolepidotus, which is A. ruppelii and not “A.” bellottii as considered in the literature over the past two decades. The latter species was originally placed in the genus Semionotus. Fossils from the Kalkschieferzone of Besnasca/Ca' del Frate (Viggiù-Varese, Italy) and Meride (Ticino, Switzerland) which were referred to Allolepidotus, rather represent a species of Eoeugnathus. Therefore, we transfer the species Semionotus bellottii to that genus and propose the new combination E. bellottii. The second and only other species originally included in the genus Allolepidotus is transferred here to the new genus Perledovatus. The holotype of P. nothosomoides new comb. has been mechanically prepared, revealing additional anatomical information that allows to place this taxon in the halecomorph family Subortichthyidae. The other halecomorph species named from the Perledo Formation, Pholidophorus oblongus and Pholidophorus curionii, have been treated as junior synonyms of E. bellottii, but our analysis indicates that they represent distinct separate taxa. However, due to the loss of the type specimens, it is not possible to decide whether they might have been conspecific with other ray-finned fishes from the Middle Triassic of the Alps.
GUILLERMO J. WINDHOLZ, JUAN D. PORFIRI, DOMENICA DOS SANTOS
et al.
Rebbachisauridae is a clade of sauropod dinosaurs whose maximum diversification and abundance are known from the
Cretaceous of South America. We describe an anterior caudal vertebra, MDPA-Pv 007, from the Upper Cretaceous of
Argentine Patagonia, whose characters allow it to be referred to this clade. Also, two phylogenetic analyses reinforce the
referral of the new material more exclusively to Rebbachisaurinae. We analyze pneumatic structures using the first CT
scans of a caudal element of a rebbachisaurid. The excellent preservation of MDPA-Pv 007, combined with CT images,
allows us to document external fossae and foramina that connect to larger internal chambers, constituting unambiguous
evidence of pneumaticity. The centrum of MDPA-Pv 007 is camerate, with large interconnected internal chambers; this is
accompanied by a neural arch with wide and deep fossae. Caudal pneumaticity has a complex phylogenetic distribution
among neosauropods. This feature may have evolved independently in diplodocoids and titanosauriforms, or it could
be ancestral for Neosauropoda but secondarily lost in a few lineages. Future investigations, taking advantage of new
technologies, will provide insights into the phylogenetic distribution and paleobiological implications of pneumaticity
in sauropod dinosaurs and other fossil archosaurs.
Shachar Don-Yehiya, Ben Burtenshaw, Ramon Fernandez Astudillo
et al.
Human feedback on conversations with language language models (LLMs) is central to how these systems learn about the world, improve their capabilities, and are steered toward desirable and safe behaviors. However, this feedback is mostly collected by frontier AI labs and kept behind closed doors. In this work, we bring together interdisciplinary experts to assess the opportunities and challenges to realizing an open ecosystem of human feedback for AI. We first look for successful practices in peer production, open source, and citizen science communities. We then characterize the main challenges for open human feedback. For each, we survey current approaches and offer recommendations. We end by envisioning the components needed to underpin a sustainable and open human feedback ecosystem. In the center of this ecosystem are mutually beneficial feedback loops, between users and specialized models, incentivizing a diverse stakeholders community of model trainers and feedback providers to support a general open feedback pool.
Sudipto Ghosh, Devanshu Verma, Balaji Ganesan
et al.
Legal research is a crucial task in the practice of law. It requires intense human effort and intellectual prudence to research a legal case and prepare arguments. Recent boom in generative AI has not translated to proportionate rise in impactful legal applications, because of low trustworthiness and and the scarcity of specialized datasets for training Large Language Models (LLMs). This position paper explores the potential of LLMs within Legal Text Analytics (LTA), highlighting specific areas where the integration of human expertise can significantly enhance their performance to match that of experts. We introduce a novel dataset and describe a human centered, compound AI system that principally incorporates human inputs for performing LTA tasks with LLMs.
In this work, we propose an LLM-based BT generation framework to leverage the strengths of both for sequential manipulation planning. To enable human-robot collaborative task planning and enhance intuitive robot programming by nonexperts, the framework takes human instructions to initiate the generation of action sequences and human feedback to refine BT generation in runtime. All presented methods within the framework are tested on a real robotic assembly example, which uses a gear set model from the Siemens Robot Assembly Challenge. We use a single manipulator with a tool-changing mechanism, a common practice in flexible manufacturing, to facilitate robust grasping of a large variety of objects. Experimental results are evaluated regarding success rate, logical coherence, executability, time consumption, and token consumption. To our knowledge, this is the first human-guided LLM-based BT generation framework that unifies various plausible ways of using LLMs to fully generate BTs that are executable on the real testbed and take into account granular knowledge of tool use.
Jinkyung Katie Park, Rahul Dev Ellezhuthil, Pamela Wisniewski
et al.
Collaborative human-AI annotation is a promising approach for various tasks with large-scale and complex data. Tools and methods to support effective human-AI collaboration for data annotation are an important direction for research. In this paper, we present CHAIRA: a Collaborative Human-AI Risk Annotation tool that enables human and AI agents to collaboratively annotate online incivility. We leveraged Large Language Models (LLMs) to facilitate the interaction between human and AI annotators and examine four different prompting strategies. The developed CHAIRA system combines multiple prompting approaches with human-AI collaboration for online incivility data annotation. We evaluated CHAIRA on 457 user comments with ground truth labels based on the inter-rater agreement between human and AI coders. We found that the most collaborative prompt supported a high level of agreement between a human agent and AI, comparable to that of two human coders. While the AI missed some implicit incivility that human coders easily identified, it also spotted politically nuanced incivility that human coders overlooked. Our study reveals the benefits and challenges of using AI agents for incivility annotation and provides design implications and best practices for human-AI collaboration in subjective data annotation.
With the increasing deployment of artificial intelligence (AI) technologies, the potential of humans working with AI agents has been growing at a great speed. Human-AI teaming is an important paradigm for studying various aspects when humans and AI agents work together. The unique aspect of Human-AI teaming research is the need to jointly study humans and AI agents, demanding multidisciplinary research efforts from machine learning to human-computer interaction, robotics, cognitive science, neuroscience, psychology, social science, and complex systems. However, existing platforms for Human-AI teaming research are limited, often supporting oversimplified scenarios and a single task, or specifically focusing on either human-teaming research or multi-agent AI algorithms. We introduce CREW, a platform to facilitate Human-AI teaming research in real-time decision-making scenarios and engage collaborations from multiple scientific disciplines, with a strong emphasis on human involvement. It includes pre-built tasks for cognitive studies and Human-AI teaming with expandable potentials from our modular design. Following conventional cognitive neuroscience research, CREW also supports multimodal human physiological signal recording for behavior analysis. Moreover, CREW benchmarks real-time human-guided reinforcement learning agents using state-of-the-art algorithms and well-tuned baselines. With CREW, we were able to conduct 50 human subject studies within a week to verify the effectiveness of our benchmark.
We examined the effectiveness of various human-AI collaboration designs on creative work. Through a human subjects experiment set in the context of creative writing, we found that while AI assistance improved productivity across all models, collaboration design significantly influenced output quality, user satisfaction, and content characteristics. Models incorporating human creative input delivered higher content interestingness and overall quality as well as greater task performer satisfaction compared to conditions where humans were limited to confirming AI's output. Increased AI involvement encouraged creators to explore beyond personal experience but also led to lower aggregate diversity in stories and genres among participants. However, this effect was mitigated through human participation in early creative tasks. These findings underscore the importance of preserving the human creative role to ensure quality, satisfaction, and creative diversity in human-AI collaboration.