Hasil untuk "Human ecology. Anthropogeography"

Carsten Meyer, Patrick Weigelt, H. Kreft

Olivia M. Smith, Kayla L. Davis, R. Pizza et al.

Tom Kouwenhoven, Max Peeperkorn, Roy de Kleijn et al.

Languages are shaped by the inductive biases of their users. Using a classical referential game, we investigate how artificial languages evolve when optimised for inductive biases in humans and large language models (LLMs) via Human-Human, LLM-LLM and Human-LLM experiments. We show that referentially grounded vocabularies emerge that enable reliable communication in all conditions, even when humans \textit{and} LLMs collaborate. Comparisons between conditions reveal that languages optimised for LLMs subtly differ from those optimised for humans. Interestingly, interactions between humans and LLMs alleviate these differences and result in vocabularies more human-like than LLM-like. These findings advance our understanding of the role inductive biases in LLMs play in the dynamic nature of human language and contribute to maintaining alignment in human and machine communication. In particular, our work underscores the need to think of new LLM training methods that include human interaction and shows that using communicative success as a reward signal can be a fruitful, novel direction.

Alex Cuellar, Ho Chit Siu, Julie A Shah

As robots' manipulation capabilities improve for pick-and-place tasks (e.g., object packing, sorting, and kitting), methods focused on understanding human-acceptable object configurations remain limited expressively with regard to capturing spatial relationships important to humans. To advance robotic understanding of human rules for object arrangement, we introduce positionally-augmented RCC (PARCC), a formal logic framework based on region connection calculus (RCC) for describing the relative position of objects in space. Additionally, we introduce an inference algorithm for learning PARCC specifications via demonstrations. Finally, we present the results from a human study, which demonstrate our framework's ability to capture a human's intended specification and the benefits of learning from demonstration approaches over human-provided specifications.

Xin Wang, Stephanie Tulk Jesso, Sadamori Kojaku et al.

Trust plays a fundamental role in shaping the willingness of users to engage and collaborate with artificial intelligence (AI) systems. Yet, measuring user trust remains challenging due to its complex and dynamic nature. While traditional survey methods provide trust levels for long conversations, they fail to capture its dynamic evolution during ongoing interactions. Here, we present VizTrust, which addresses this challenge by introducing a real-time visual analytics tool that leverages a multi-agent collaboration system to capture and analyze user trust dynamics in human-agent communication. Built on established human-computer trust scales-competence, integrity, benevolence, and predictability-, VizTrust enables stakeholders to observe trust formation as it happens, identify patterns in trust development, and pinpoint specific interaction elements that influence trust. Our tool offers actionable insights into human-agent trust formation and evolution in real time through a dashboard, supporting the design of adaptive conversational agents that responds effectively to user trust signals.

Georgios Hadjiantonis, Sarah Gillet, Marynel Vázquez et al.

Robot-moderated group discussions have the potential to facilitate engaging and productive interactions among human participants. Previous work on topic management in conversational agents has predominantly focused on human engagement and topic personalization, with the agent having an active role in the discussion. Also, studies have shown the usefulness of including robots in groups, yet further exploration is still needed for robots to learn when to change the topic while facilitating discussions. Accordingly, our work investigates the suitability of machine-learning models and audiovisual non-verbal features in predicting appropriate topic changes. We utilized interactions between a robot moderator and human participants, which we annotated and used for extracting acoustic and body language-related features. We provide a detailed analysis of the performance of machine learning approaches using sequential and non-sequential data with different sets of features. The results indicate promising performance in classifying inappropriate topic changes, outperforming rule-based approaches. Additionally, acoustic features exhibited comparable performance and robustness compared to the complete set of multimodal features. Our annotated data is publicly available at https://github.com/ghadj/topic-change-robot-discussions-data-2024.

Akshay K. Jagadish, Mirko Thalmann, Julian Coda-Forno et al.

Human cognition is profoundly shaped by the environments in which it unfolds. Yet, it remains an open question whether learning and decision making can be explained as a principled adaptation to the statistical structure of real-world tasks. We introduce ecologically rational analysis, a computational framework that unifies the normative foundations of rational analysis with ecological grounding. Leveraging large language models to generate ecologically valid cognitive tasks at scale, and using meta-learning to derive rational models optimized for these environments, we develop a new class of learning algorithms: Ecologically Rational Meta-learned Inference (ERMI). ERMI internalizes the statistical regularities of naturalistic problem spaces and adapts flexibly to novel situations, without requiring hand-crafted heuristics or explicit parameter updates. We show that ERMI captures human behavior across 15 experiments spanning function learning, category learning, and decision making, outperforming several established cognitive models in trial-by-trial prediction. Our results suggest that much of human cognition may reflect adaptive alignment to the ecological structure of the problems we encounter in everyday life.

Zhiyao Sun, Yu-Hui Wen, Ho-Jui Fang et al.

Creating detailed 3D human avatars with fitted garments traditionally requires specialized expertise and labor-intensive workflows. While recent advances in generative AI have enabled text-to-3D human and clothing synthesis, existing methods fall short in offering accessible, integrated pipelines for generating CG-ready 3D avatars with physically compatible outfits; here we use the term CG-ready for models following a technical aesthetic common in computer graphics (CG) and adopt standard CG polygonal meshes and strands representations (rather than neural representations like NeRF and 3DGS) that can be directly integrated into conventional CG pipelines and support downstream tasks such as physical simulation. To bridge this gap, we introduce Tailor, an integrated text-to-3D framework that generates high-fidelity, customizable 3D avatars dressed in simulation-ready garments. Tailor consists of three stages. (1) Seman tic Parsing: we employ a large language model to interpret textual descriptions and translate them into parameterized human avatars and semantically matched garment templates. (2) Geometry-Aware Garment Generation: we propose topology-preserving deformation with novel geometric losses to generate body-aligned garments under text control. (3) Consistent Texture Synthesis: we propose a novel multi-view diffusion process optimized for garment texturing, which enforces view consistency, preserves photorealistic details, and optionally supports symmetric texture generation common in garments. Through comprehensive quantitative and qualitative evaluations, we demonstrate that Tailor outperforms state-of-the-art methods in fidelity, usability, and diversity. Our code will be released for academic use. Project page: https://human-tailor.github.io

Morten Roed Frederiksen, Kasper Støy

In an effort to improve how robots function in social contexts, this paper investigates if a robot that actively shares a reaction to an event with a human alters how the human perceives the robot's affective impact. To verify this, we created two different test setups. One to highlight and isolate the reaction element of affective robot expressions, and one to investigate the effects of applying specific timing delays to a robot reacting to a physical encounter with a human. The first test was conducted with two different groups (n=84) of human observers, a test group and a control group both interacting with the robot. The second test was performed with 110 participants using increasingly longer reaction delays for the robot with every ten participants. The results show a statistically significant change (p$<$.05) in perceived affective impact for the robots when they react to an event shared with a human observer rather than reacting at random. The result also shows for shared physical interaction, the near-human reaction times from the robot are most appropriate for the scenario. The paper concludes that a delay time around 200ms may render the biggest impact on human observers for small-sized non-humanoid robots. It further concludes that a slightly shorter reaction time around 100ms is most effective when the goal is to make the human observers feel they made the biggest impact on the robot.

Alexander Scarlatos, Yusong Wu, Ian Simon et al.

Recent advances in generative artificial intelligence (AI) have created models capable of high-quality musical content generation. However, little consideration is given to how to use these models for real-time or cooperative jamming musical applications because of crucial required features: low latency, the ability to communicate planned actions, and the ability to adapt to user input in real-time. To support these needs, we introduce ReaLJam, an interface and protocol for live musical jamming sessions between a human and a Transformer-based AI agent trained with reinforcement learning. We enable real-time interactions using the concept of anticipation, where the agent continually predicts how the performance will unfold and visually conveys its plan to the user. We conduct a user study where experienced musicians jam in real-time with the agent through ReaLJam. Our results demonstrate that ReaLJam enables enjoyable and musically interesting sessions, and we uncover important takeaways for future work.

Alex Leitch, Celia Chen

As AI art generation becomes increasingly sophisticated, HCI research has focused primarily on questions of detection, authenticity, and automation. This paper argues that such approaches fundamentally misunderstand how artistic value emerges from the concerns that drive human image production. Through examination of historical precedents, we demonstrate that artistic style is not only visual appearance but the resolution of creative struggle, as artists wrestle with influence and technical constraints to develop unique ways of seeing. Current AI systems flatten these human choices into reproducible patterns without preserving their provenance. We propose that HCI's role lies not only in perfecting visual output, but in developing means to document the origins and evolution of artistic style as it appears within generated visual traces. This reframing suggests new technical directions for HCI research in generative AI, focused on automatic documentation of stylistic lineage and creative choice rather than simple reproduction of aesthetic effects.

Roberto Donato da Silva Júnior, José Diego Gobbo Alves, Álvaro de Oliveira D’Antona et al.

Abstract We analyzed the constitution and challenges of integrated management of protected areas through the “Mosaic” model, based on one of the most relevant experiences in Brazilian conservation: the Lower Rio Negro Mosaic (MBRN). Based on fieldwork data collected from 106 traditional communities within MBRN and an analysis of documentation from the 26 meetings held by the Mosaic Advisory Council, we identified the successes and challenges of the proposal. The results indicate a political and institutional maturation of MBRN since its establishment in 2010, especially given the complexity of forming a Council focused on participatory management. However, inequality in the distribution of infrastructure, services, and projects implemented by governmental and non-governmental institutions poses some challenges to be addressed. Such socio-spatial inequality and variations in environmental management attributes affect the recognition of the Mosaic by community leaders and hinder the enhancement of integrated, multilevel management.

Abigail Skwara, K. Gowda, M. Yousef et al.

A. B. Pedersen, A. Fenton

A. Bevan, B. Cutler, C. Hennig et al.

Quantifiable, spatially-resolved, large-scale evidence about traditional food storage facilities is extremely rare, and yet highly insightful for researchers across subjects such as human ecology, anthropology, agronomy, archaeology and economic history. This paper takes advantage of some unusually detailed French colonial era records of cereal storage and agro-pastoral practice in 19th century central Algeria that inventory the underground food stores of different sedentary and nomadic tribes at a moment of colonial confrontation in which these stores were central to ecological and political resilience. We consider how different aspects of these food stores relate to environmental, social and economic variables across the study area. The overall results suggest important north-south trends in agro-pastoral lifestyle and storage practice.

A. D’Agostino, G. Di Marco, M. Rolfo et al.

Abstract Plants have always represented a key element in landscape delineation. Indeed, plant diversity, whose distribution is influenced by geographic/climatic variability, has affected both environmental and human ecology. The present contribution represents a multi‐proxy study focused on the detection of starch, pollen and non‐pollen palynomorphs in ancient dental calculus collected from pre‐historical individuals buried at La Sassa and Pila archaeological sites (Central Italy). The collected record suggested the potential use of plant taxa by the people living in Central Italy during the Copper‐Middle Bronze Age and expanded the body of evidence reported by previous palynological and palaeoecological studies. The application of a microscopic approach provided information about domesticated crops and/or gathered wild plants and inferred considerations on ancient environments, water sources, and past health and diseases. Moreover, the research supplied data to define the natural resources (e.g., C4‐plant intake) and the social use of the space during that period. Another important aspect was the finding of plant clues referable to woody habitats, characterised by broad‐leaved deciduous taxa and generally indicative of a warm‐temperate climate and grassy vegetation. Other unusual records (e.g., diatoms, brachysclereids) participated in defining the prehistoric ecological framework. Thus, this work provides an overview on the potential of the human dental calculus analysis to delineate some features of the ancient plant ecology and biodiversity.

D. Schmitt, K. Lupo, Nicolette M. Edwards et al.

We repurpose multidecadal ethnoarchaeological investigations of human hunting, prey availability, and socioeconomics in a rural Central African Republic village in the service of human ecology. Focusing on forest foragers in the village of Grima, initial 1999–2005 (Old Grima) data collection included documentation of hunting technology and offtakes, identification of wild meat bone assemblages, inventories of household material goods, and measurements of horticultural fields. Similar datasets were collected in 2021–2022 (New Grima) and longitudinal comparisons of prey remains and material wealth detected many significant differences. Old Grima house middens contained larger numbers of bones representing an array of wild meat taxa and inventories recorded diverse and abundant collections of material goods. The New Grima comparative data showed a reduction in the consumption of wild meat, increases in guns and especially metal cable snares, and marked declines in local wild meat (notably duiker) populations and forager material wealth paired with increases in debt. In 2022 the New Grima inhabitants were actively pursuing escargot for food and income and house middens were dominated by tortoise remains. All the data point to resource depression from overhunting and a community in jeopardy. The comparisons are also important because they include the transition from traditional nets and spears to more efficient metal cable snares and guns and provide information on the effects of hunting technology. The complexities of evaluating ecological perturbations and sustainability require multidisciplinary datasets and we propose ethnoarchaeology as a valuable tool to help identify subtleties in human food webs and biodiversity loss.

Chi Ian Tang, Lorena Qendro, Dimitris Spathis et al.

Wearable-based Human Activity Recognition (HAR) is a key task in human-centric machine learning due to its fundamental understanding of human behaviours. Due to the dynamic nature of human behaviours, continual learning promises HAR systems that are tailored to users' needs. However, because of the difficulty in collecting labelled data with wearable sensors, existing approaches that focus on supervised continual learning have limited applicability, while unsupervised continual learning methods only handle representation learning while delaying classifier training to a later stage. This work explores the adoption and adaptation of CaSSLe, a continual self-supervised learning model, and Kaizen, a semi-supervised continual learning model that balances representation learning and down-stream classification, for the task of wearable-based HAR. These schemes re-purpose contrastive learning for knowledge retention and, Kaizen combines that with self-training in a unified scheme that can leverage unlabelled and labelled data for continual learning. In addition to comparing state-of-the-art self-supervised continual learning schemes, we further investigated the importance of different loss terms and explored the trade-off between knowledge retention and learning from new tasks. In particular, our extensive evaluation demonstrated that the use of a weighting factor that reflects the ratio between learned and new classes achieves the best overall trade-off in continual learning.

Yuchong Zhang, Miguel Vasco, Mårten Björkman et al.

This paper presents findings from an exploratory needfinding study investigating the research current status and potential participation of the competitions on the robotics community towards four human-centric topics: safety, privacy, explainability, and federated learning. We conducted a survey with 34 participants across three distinguished European robotics consortia, nearly 60% of whom possessed over five years of research experience in robotics. Our qualitative and quantitative analysis revealed that current mainstream robotic researchers prioritize safety and explainability, expressing a greater willingness to invest in further research in these areas. Conversely, our results indicate that privacy and federated learning garner less attention and are perceived to have lower potential. Additionally, the study suggests a lack of enthusiasm within the robotics community for participating in competitions related to these topics. Based on these findings, we recommend targeting other communities, such as the machine learning community, for future competitions related to these four human-centric topics.

Meghana Deodhar, Xiao Ma, Yixin Cai et al.

We deal with the problem of localized in-video taxonomic human annotation in the video content moderation domain, where the goal is to identify video segments that violate granular policies, e.g., community guidelines on an online video platform. High quality human labeling is critical for enforcement in content moderation. This is challenging due to the problem of information overload - raters need to apply a large taxonomy of granular policy violations with ambiguous definitions, within a limited review duration to relatively long videos. Our key contribution is a novel human-machine learning (ML) collaboration framework aimed at maximizing the quality and efficiency of human decisions in this setting - human labels are used to train segment-level models, the predictions of which are displayed as "hints" to human raters, indicating probable regions of the video with specific policy violations. The human verified/corrected segment labels can help refine the model further, hence creating a human-ML positive feedback loop. Experiments show improved human video moderation decision quality, and efficiency through more granular annotations submitted within a similar review duration, which enable a 5-8% AUC improvement in the hint generation models.