Accurately generating images across the Tree of Life is difficult: there are over 10M distinct species on Earth, many of which differ only by subtle visual traits. Despite the remarkable progress in text-to-image synthesis, existing models often fail to capture the fine-grained visual cues that define species identity, even when their outputs appear photo-realistic. To this end, we propose TaxaAdapter, a simple and lightweight approach that incorporates Vision Taxonomy Models (VTMs) such as BioCLIP to guide fine-grained species generation. Our method injects VTM embeddings into a frozen text-to-image diffusion model, improving species-level fidelity while preserving flexible text control over attributes such as pose, style, and background. Extensive experiments demonstrate that TaxaAdapter consistently improves morphology fidelity and species-identity accuracy over strong baselines, with a cleaner architecture and training recipe. To better evaluate these improvements, we also introduce a multimodal Large Language Model-based metric that summarizes trait-level descriptions from generated and real images, providing a more interpretable measure of morphological consistency. Beyond this, we observe that TaxaAdapter exhibits strong generalization capabilities, enabling species synthesis in challenging regimes such as few-shot species with only a handful of training images and even species unseen during training. Overall, our results highlight that VTMs are a key ingredient for scalable, fine-grained species generation.

Fine-grained wildlife classification in aerial imagery is limited not only by model performance, but also by unreliable labels: animals occupy few pixels, key visual cues vary seasonally, and modality-specific evidence can be ambiguous. We study adult-male identification in red deer, where the antler cycle defines predictable windows of reliable evidence for both annotation and prediction. Using 7,295 RGB-only, thermal-only, and matched RGB+thermal crop sets labeled by three annotators, we show that seasonal structure links (I) annotation quality, (II) downstream classification, and (III) selective prediction. Matched RGB+thermal review resolves more samples than either single modality, recovering majority-male labels otherwise missed by RGB or thermal alone, in human based as well as model based classification. Months with high annotator abstention also show lower classifier confidence, and soft seasonal priors mainly benefit the season-limited thermal view. Uncertainty-band abstention further improves covered accuracy up to 98.9

Humans have remarkable selective sensitivity to identities--they easily distinguish between highly-similar identities, even across significantly different contexts such as diverse viewpoints or lighting. Vision models have struggled to match this capability, and progress towards identity-focused tasks such as personalized image generation is slowed by a lack of identity-focused evaluation metrics. To help facilitate progress, we propose ID-Sim, a feed-forward metric designed to faithfully reflect human selective sensitivity. To build ID-Sim, we curate a high-quality training set of images spanning diverse real-world domains, augmented with generative synthetic data that provides controlled, fine-grained identity and contextual variations. We evaluate our metric on a new unified evaluation benchmark for assessing consistency with human annotations across identity-focused recognition, retrieval, and generative tasks.

Computer vision is increasingly used to automate recognition tasks in large ecological datasets, but more complex tasks such as multi-object tracking continue to pose challenges. As researchers seek to incorporate vision models in ecology workflows, various lines of research have explored how to make imperfect predictions useful through human-in-the-loop processes. We propose a new approach to working with imperfect tracking predictions through an interactive prediction correction workflow taking place as a conversation with a multimodal large language model, which we tailor to a sonar fish tracking dataset as an initial proof of concept. We investigate the performance of the tool, Molmo2Fish, across guided and unguided tasks, correcting its own predicted tracks and external tracks. We find that Molmo2Fish achieves high performance on fish tracking and track correction tasks, but there is still much room to improve on incorporating natural language guidance. The code and data are publicly available at https://github.com/tidalove/molmo2fish.

Movement ecology-the study of how and why animals move within their environments-stands to offer transformative insights into our rapidly changing world, with benefits for both nature and people. Here, we present the first global horizon scan for movement ecology, engaging leading experts to identify innovations likely to shape the field over the next two decades. These include: engineering breakthroughs, such as long-lived miniature tags with enhanced sensing capacities, non-invasive attachment mechanisms and real-time data processing; analytical advances to predict movement trajectories and scale individual data to population-level patterns; and targeted coordination to mobilize data, scale collaborative infrastructure and expand participation in underrepresented regions. Strategic investment in these priorities would advance understanding of wildlife biology and ecosystem functions, providing mechanistic insights that could help address planetary-scale challenges from biodiversity loss to global health. To highlight these opportunities, we map alignment between identified innovations, movement ecology applications and key multilateral environmental frameworks, including the Kunming-Montreal Global Biodiversity Framework and the Sustainable Development Goals. Our analyses fill a gap at a critical juncture in the evolution of movement ecology as a discipline, offering a community-driven agenda that calls attention to the wide-reaching implications of the advancements on the horizon today.

Abstract Ecology and artificial intelligence (AI) are becoming increasingly intertwined. Originally, the intersection between the two disciplines was driven by a critical need for AI to help process rapidly growing volumes of ecological data. Early applications primarily entailed applying AI methods to automate relatively basic tasks, such as detecting blank images from camera traps. However, researchers in both disciplines are beginning to recognize the potential for transformative advances when AI is fully integrated into ecological research and conservation practice. This special feature presents research at the cutting edge of the AI–ecology interface, focusing on work that advances the state of both fields beyond proof‐of‐concept to true interdisciplinary insight. The papers in this collection reveal a maturing field that balances technical advancement with ecological relevance. They address both methodological challenges and the critical need for meaningful integration between computer science innovations and fundamental ecological questions. As a whole, this collection demonstrates the potential for AI to enhance both fundamental ecological understanding and applied conservation efforts, as well as to bridge the gap between scientific discovery and policy implementation. The special feature underscores the importance of genuine interdisciplinary collaboration in developing technologies that not only showcase technical prowess, but also address pressing ecological challenges and support evidence‐based decision‐making in biodiversity conservation.

Climate information is advancing faster than the decision systems designed to use it. Emergency response operates on timescales of hours, whereas societal adaptation unfolds over decades. Yet climate science, impact assessment and policy remain poorly integrated, limiting coherent action across timescales. We argue that artificial intelligence should be developed not only as a domain-specific tool, but as an integration layer linking fragmented physical, social and institutional systems. This Perspective outlines a blueprint for Integrated Climate Intelligence built on three pillars: fairness, which prioritizes vulnerability alongside data availability; speed, which reduces the latency between assessment and action; and robustness, which quantifies uncertainty under extrapolation and high-stakes decision-making. We further propose a Climate AI Trust Index as an evaluation framework for assessing whether climate AI systems are sufficiently rigorous, equitable and decision-relevant for operational use.

Many ecological questions center on complex phenomena, such as species interactions, behaviors, phenology, and responses to disturbance, that are inherently difficult to observe and sparsely documented. Community science platforms such as iNaturalist contain hundreds of millions of biodiversity images, which often contain evidence of these complex phenomena. However, current workflows that seek to discover and analyze this evidence often rely on manual inspection, leaving this information largely inaccessible at scale. We introduce INQUIRE-Search, an open-source system that uses natural language to enable scientists to rapidly search within an ecological image database like iNaturalist for specific phenomena, verify and export relevant observations, and use these outputs for downstream scientific analysis. Compared to existing methods, INQUIRE-Search concentrates relevant observations 3–8 times more efficiently under comparable manual inspection budgets across five ecological case studies. This opening up new possibilities for scientific question answering. Through five case studies, we demonstrate how INQUIRE-Search can be used for ecological inference, from analyzing seasonal variation in behavior across species to forest regrowth after wildfires. These examples illustrate a new paradigm for interactive, efficient, and scalable scientific discovery that can begin to unlock previously inaccessible scientific value in large-scale biodiversity datasets. Finally, we highlight how AI-enabled discovery tools for science require reframing aspects of the scientific process, including experiment design, data collection, survey effort, and uncertainty analysis.

Autonomous scientific discovery systems offer the potential to accelerate research by automating the process of hypothesis generation and validation. However, current systems operate within constrained search spaces or require predefined research questions, limiting their capacity for true open-ended inquiry. Furthermore, while they generate hypotheses iteratively, they largely lack the ability to explicitly synthesize their own accumulated findings to uncover complex, interconnected phenomena. We introduce DiscoPER, an autonomous large language model-powered framework that conducts open-ended research by dynamically generating and executing code to explore datasets without pre-specified research objectives. To ensure rigorous scientific validity, every proposed discovery must pass statistical testing. To overcome the limitations of isolated search, our framework introduces a second-order reasoning mechanism that periodically analyzes its own accumulated discoveries. By treating prior discoveries as empirical data, DiscoPER identifies structural patterns, confounds, and epistemic gaps, actively redirecting hypothesis exploration toward uncharted regions of the search space. The search space is further expanded by incorporating tool use, enabling the system to explore hypotheses beyond structured metadata by seamlessly processing and extracting useful information from multimodal sources like images. Evaluated on iNatDisco, a new multimodal ecological knowledge benchmark with pattern-level ground truth obtained from peer-reviewed literature, DiscoPER recovers 8 of 9 known patterns with a 72.7

Recent advances in self-supervised visual representation learning have demonstrated the effectiveness of predictive latent-space objectives for learning transferable features. In particular, Image-based Joint-Embedding Predictive Architecture (I-JEPA) learns representations by predicting latent embeddings of masked target regions from visible context. However, it predicts target regions in parallel and all at once, lacking ability to order predictions meaningfully. Inspired by human visual perception, which attends selectively and progressively from primary to secondary cues, we propose DSeq-JEPA, a Discriminative Sequential Joint-Embedding Predictive Architecture that bridges latent predictive and autoregressive self-supervised learning. Specifically, DSeq-JEPA integrates a discriminatively ordered sequential process with JEPA-style learning objective. This is achieved by (i) identifying primary discriminative regions using an attention-derived saliency map that serves as a proxy for visual importance, and (ii) predicting subsequent regions in discriminative order, inducing a curriculum-like semantic progression from primary to secondary cues in pre-training. Extensive experiments across tasks – image classification (ImageNet), fine-grained visual categorization (iNaturalist21, CUB, Stanford Cars), detection/segmentation (MS-COCO, ADE20K), and low-level reasoning (CLEVR) – show that DSeq-JEPA consistently learns more discriminative and generalizable representations compared to I-JEPA variants. Project page: https://github.com/SkyShunsuke/DSeq-JEPA.

Scientific and environmental imagery often suffer from complex mixtures of noise related to the sensor and the environment. Existing restoration methods typically remove one degradation at a time, leading to cascading artifacts, overcorrection, or loss of meaningful signal. In scientific applications, restoration must be able to simultaneously handle compound degradations while allowing experts to selectively remove subsets of distortions without erasing important features. To address these challenges, we present PRISM (Precision Restoration with Interpretable Separation of Mixtures). PRISM is a prompted conditional diffusion framework which combines compound-aware supervision over mixed degradations with a weighted contrastive disentanglement objective that aligns primitives and their mixtures in the latent space. This compositional geometry enables high-fidelity joint removal of overlapping distortions while also allowing flexible, targeted fixes through natural language prompts. Across microscopy, wildlife monitoring, remote sensing, and urban weather datasets, PRISM outperforms state-of-the-art baselines on complex compound degradations, including zero-shot mixtures not seen during training. Importantly, we show that selective restoration significantly improves downstream scientific accuracy in several domains over standard "black-box" restoration. These results establish PRISM as a generalizable and controllable framework for high-fidelity restoration in domains where scientific utility is a priority.

The estimation of abundance and density in unmarked populations of great apes relies on statistical frameworks that require animal-to-camera distance measurements. In practice, acquiring these distances depends on labour-intensive manual interpretation of animal observations across large camera trap video corpora. This study introduces and evaluates an only sparsely explored alternative: the integration of computer vision–based monocular depth estimation (MDE) pipelines directly into ecological camera trap workflows for great ape conservation. Using a real-world dataset of 220 camera trap videos documenting a wild chimpanzee population, we combine two MDE models—Dense Prediction Transformers and Depth Anything—with multiple distance sampling strategies. These components are used to generate detection distance estimates, from which population density and abundance are inferred. Comparative analysis against manually derived ground-truth distances shows that calibrated Dense Prediction Transformer consistently outperforms DepthAnything. This advantage is observed in both distance estimation accuracy and downstream density and abundance inference. Nevertheless, both models exhibit systematic biases. We show that, given complex forest environments, they tend to overestimate detection distances and consequently underestimate density and abundance relative to conventional manual approaches. We further find that failures in animal detection across distance ranges are a primary factor limiting estimation accuracy. Overall, this work provides a case study beyond pure chimp distance estimation, demonstrating practically that MDE-driven camera trap distance sampling is a viable alternative to manual distance estimation via a case study. The proposed approach yields density and abundance estimates within 22% of those obtained using traditional methods. Taken together, these results indicate a realistic pathway toward scalable and automated population modelling for great apes.

Recognizing individual animals over time is central to many ecological and conservation questions, including estimating abundance, survival, movement, and social structure. Recent advances in automated identification from images and even acoustic data suggest that this process could be greatly accelerated, yet their promise has not translated well into ecological practice. We argue that the main barrier is not the performance of the automated methods themselves, but a mismatch between how those methods are typically developed and evaluated, and how ecological data is actually collected, processed, reviewed, and used. Future progress, therefore, will depend less on algorithmic gains alone than on recognizing that the usefulness of automated identification is grounded in ecological context: it depends on what question is being asked, what data are available, and what kinds of mistakes matter. Only by centering these questions can we move toward automated identification of individuals that is not only accurate but also ecologically useful, transparent, and trustworthy.

Aerial drone surveys increasingly support wildlife population estimation, yet a useful census is more than a count: population dynamics are defined by species composition, sex ratios and age structure, that is, by which species are present and how a herd splits into adult males, adult females and juveniles. We use red deer (Cervus elaphus) as a test case, because managers act on these dynamics and because the visible cue defining adult males, the antlers, is seasonally variable. Surveys are flown nadir, high enough not to disturb the animals, so each deer occupies only a small, low-resolution patch. The two recording modalities fail in opposite conditions: in color a deer under canopy blends into the ground, while in thermal it becomes a bright blob that loses fine detail. Rather than trust either modality alone, we fuse them at every stage using self-supervised DINOv3 features. Our pipeline tracks animals in both modalities, treats an animal as confirmed only when the two cameras agree, keeps only the clear, non-occluded frames, and assigns species and sex by a vote across them; life stage is read separately from geo-referenced body size, since at survey resolution a juvenile often only differs from an adult female in size. Across four flights spanning the antler season the fused pipeline correctly classifies 25 of the 26 detected individuals (7 of 8 adult males, all 16 adult females and 2 juveniles), against 20 of 26 for either sensor alone. Multimodal species classification reaches 96.0

Active learning is now standard practice in labeling ecological data, enabling ecologists to quickly process large volumes of field data to understand and monitor natural environments. Current practices evaluate active learning inductively, estimating predictive performance on a held-out test set. We argue that this evaluation is misaligned with most ecological tasks, where the goal is to transductively label an entire pool of data as efficiently as possible. We demonstrate that ignoring the human-in-the-loop underestimates the importance of continuing to label, particularly for classes in the long tail which may be of disproportionate ecological importance (rare species, uncommon behaviors, etc.). Our analysis shows that, for this long tail, the transductive objective shifts importance from prediction to discovery: the true challenge becomes finding "needles in the haystack," examples of rare classes that are embedded within dense regions of abundant classes in the latent geometry, which we quantify with a novel metric of sampling difficulty. Finally, to translate these insights to practical ecological workflows, we propose a conservative hybrid stopping criterion inspired by ecological rarefaction curves, and show that combining predictive performance with discovery criteria reduces premature stopping on long-tailed pools, improving rare-class recovery when discovery, not classification, is the limiting factor.

Abstract Monitoring fish movement is essential for understanding population dynamics, informing conservation efforts and supporting fisheries management. Traditional methods, such as visual observations by volunteers, are constrained by time limitations, environmental conditions and labour intensity. Recent advancements in computer vision (CV) and deep learning offer promising solutions for automating fish counting from underwater videos, improving efficiency and data resolution. In this study, we developed and applied a deep learning‐based CV system to monitor river herring (Alosa spp.) migration, covering all essential steps from field camera deployment, video annotation to model training and in‐season population counting. We assessed the labelling and training efforts required to achieve good model performance and explored the use of importance sampling to correct biases in CV‐based fish counts. Our results demonstrated that CV models trained on a single site and year showed limited generalization to sites or years unseen during training, while models trained on more diverse labelled data generalized better. We also found that the amount of annotations required is related to dataset complexity. When applied for in‐season fish counting, CV efficiently processed season‐long datasets and produced counts consistent with human review, with some moderate differences under migration pulses that can be adjusted by importance sampling. By providing continuous, high‐resolution monitoring throughout the entire migration season, CV counts offer more reliable run size estimates and greater insight into the spawning migration of river herring. This study demonstrates a scalable, cost‐effective and efficient approach with significant potential for addressing complex ecological questions and supporting conservation strategies and resource management.

The integration of artificial intelligence (AI) into biodiversity research and conservation is growing rapidly, demonstrating great potential in reducing the intensive human labour required for data preprocessing, thereby, facilitating larger data collections that offer ecological insights at unprecedented scales. However, most of these AI applications for biodiversity are still in the early stages of development, hindered by challenges inherent in real-world datasets and the limited accessibility of these technologies to practitioners without extensive programming knowledge. The recent advent of multimodal language models, which can process and generate multiple data modalities, has significantly expanded the realm of possible AI applications in biodiversity research. These models have demonstrated the ability to classify species and recognize more complex concepts, such as animal postures and orientations, without prior exposure during training. Multimodal language models can also provide explanations for their predictions and interact with humans in natural language, thereby making them more transparent, intuitive and accessible to non-specialists. Despite these advancements, the use of multimodal language models for biodiversity still needs to overcome unique barriers to application, including high computational and financial demands, reliance on prompt engineering for consistent model performance on large datasets and insufficient open-source sharing of state-of-the-art methods. This paper explores the transformative potential of multimodal language models for biodiversity research and discusses several possible applications in biodiversity research. We also discuss challenges to implement these models in real-world conservation scenarios and propose directions for future research to overcome these hurdles. Our goal is to encourage robust discussions and research into the integration of multimodal language models to advance AI for biodiversity research and conservation.

Fine-grained image retrieval in scientific domains such as ecology demands compositional and expert-level reasoning that general-purpose VLMs often lack. In this paper, we introduce a two-stage structured reranking pipeline that augments queries with web-sourced expert knowledge and decomposes them into verifiable subquestions using large multimodal models. Then, images are scored against these subqueries to produce a final relevance ranking. Our method consistently boosts retrieval accuracy, especially on behavioral and contextual queries, while also greatly reducing manual effort, improving interpretability, and advancing automated visual search for scientific research.

Large, well described gaps exist in both what we know and what we need to know to address the biodiversity crisis. Artificial intelligence (AI) offers new potential for filling these knowledge gaps, but where the biggest and most influential gains could be made remains unclear. To date, biodiversity-related uses of AI have largely focused on tracking and monitoring of wildlife populations. Rapid progress is being made in the use of AI to build phylogenetic trees and species distribution models. However, AI also has considerable unrealized potential in the re-evaluation of important ecological questions, especially those that require the integration of disparate and inherently complex data types, such as images, video, text, audio and DNA. This Review describes the current and potential future use of AI to address seven clearly defined shortfalls in biodiversity knowledge. Recommended steps for AI-based improvements include the re-use of existing image data and the development of novel paradigms, including the collaborative generation of new testable hypotheses. The resulting expansion of biodiversity knowledge could lead to science spanning from genes to ecosystems — advances that might represent our best hope for meeting the rapidly approaching 2030 targets of the Global Biodiversity Framework. Seven well defined shortfalls in global biodiversity data must be overcome to meet critical global biodiversity targets. Pollock et al. discuss the current and future roles of AI in bridging these knowledge gaps.