The widespread availability of off-the-shelf machine learning models poses a challenge: which model, of the many available candidates, should be chosen for a given data analysis task? This question of model selection is traditionally answered by collecting and annotating a validation dataset -- a costly and time-intensive process. We propose a method for active model selection, using predictions from candidate models to prioritize the labeling of test data points that efficiently differentiate the best candidate. Our method, CODA, performs consensus-driven active model selection by modeling relationships between classifiers, categories, and data points within a probabilistic framework. The framework uses the consensus and disagreement between models in the candidate pool to guide the label acquisition process, and Bayesian inference to update beliefs about which model is best as more information is collected. We validate our approach by curating a collection of 26 benchmark tasks capturing a range of model selection scenarios. CODA outperforms existing methods for active model selection significantly, reducing the annotation effort required to discover the best model by upwards of 70% compared to the previous state-of-the-art. Code and data are available at https://github.com/justinkay/coda.

Hierarchical classification predicts labels across multiple levels of a taxonomy, e.g., from coarse-level \textit{Bird} to mid-level \textit{Hummingbird} to fine-level \textit{Green hermit}, allowing flexible recognition under varying visual conditions. It is commonly framed as multiple single-level tasks, but each level may rely on different visual cues. Distinguishing \textit{Bird} from \textit{Plant} relies on {\it global features} like {\it feathers} or {\it leaves}, while separating \textit{Anna's hummingbird} from \textit{Green hermit} requires {\it local details} such as {\it head coloration}. Prior methods improve accuracy using external semantic supervision, but such statistical learning criteria fail to ensure consistent visual grounding at test time, resulting in incorrect hierarchical classification. We propose, for the first time, to enforce \textit{internal visual consistency} by aligning fine-to-coarse predictions through intra-image segmentation. Our method outperforms zero-shot CLIP and state-of-the-art baselines on hierarchical classification benchmarks, achieving both higher accuracy and more consistent predictions. It also improves internal image segmentation without requiring pixel-level annotations.

Modern vision models excel at general purpose downstream tasks. It is unclear, however, how they may be used for personalized vision tasks, which are both fine-grained and data-scarce. Recent works have successfully applied synthetic data to general-purpose representation learning, while advances in T2I diffusion models have enabled the generation of personalized images from just a few real examples. Here, we explore a potential connection between these ideas, and formalize the challenge of using personalized synthetic data to learn personalized representations, which encode knowledge about an object of interest and may be flexibly applied to any downstream task relating to the target object. We introduce an evaluation suite for this challenge, including reformulations of two existing datasets and a novel dataset explicitly constructed for this purpose, and propose a contrastive learning approach that makes creative use of image generators. We show that our method improves personalized representation learning for diverse downstream tasks, from recognition to segmentation, and analyze characteristics of image generation approaches that are key to this gain.

Global biodiversity is declining at an unprecedented rate, yet little information isknown about most species and how their populations are changing. Indeed, some90% Earth’s species are estimated to be completely unknown. Machine learning hasrecently emerged as a promising tool to facilitate long-term, large-scale biodiversitymonitoring, including algorithms for fine-grained classification of species fromimages. However, such algorithms typically are not designed to detect examplesfrom categories unseen during training – the problem of open-set recognition(OSR) – limiting their applicability for highly diverse, poorly studied taxa such asinsects. To address this gap, we introduce Open-Insect, a large-scale, fine-graineddataset to evaluate unknown species detection across different geographic regionswith varying difficulty. We benchmark 38 OSR algorithms across three categories:post-hoc, training-time regularization, and training with auxiliary data, finding thatsimple post-hoc approaches remain a strong baseline. We also demonstrate how toleverage auxiliary data to improve species discovery in regions with limited data.Our results provide timely insights to guide the development of computer visionmethods for biodiversity monitoring and species discovery.

Benchmarks for out-of-distribution (OOD) generalization often reveal a strong positive correlation between in-distribution (ID) and OOD accuracy across models, a phenomenon known as “accuracy-on-the-line.” This pattern is commonly interpreted as evidence that spurious correlations—relationships that improve ID but harm OOD performance—are rare in practice. We show that this positive correlation can be an artifact of aggregating heterogeneous OOD examples. Using a simple gradient-based method, OODSelect, we identify semantically coherent OOD subsets where accuracy-on-the-line breaks down. Across widely used distribution-shift benchmarks, OODSelect uncovers subsets—sometimes comprising more than half of the standard OOD set—where higher ID accuracy predicts lower OOD accuracy. These results suggest that aggregate metrics can mask critical failure modes in OOD robustness. We release code and the identified subsets to support further research.

ABSTRACT The Grévy's zebra, an endangered species native to Kenya and southern Ethiopia, has been the target of sustained conservation efforts in recent years. Accurately monitoring Grévy's zebra populations is essential for ecologists to evaluate ongoing conservation initiatives. Recently, in both 2016 and 2018, a full census of the Grévy's zebra population was enabled by the Great Grévy's Rally (GGR), a citizen science event that combines teams of volunteers to capture data with computer vision algorithms that help experts estimate the number of individuals in the population. A complementary, scalable, cost‐effective and long‐term Grévy's population monitoring approach involves deploying a network of camera traps, which we have done at the Mpala Research Centre in Laikipia County, Kenya. In both scenarios, a substantial majority of the images of zebras are not usable for individual identification due to ‘in‐the‐wild’ imaging conditions—occlusions from vegetation or other animals, oblique views, low image quality and animals that appear in the far background and are thus too small to identify. Camera trap images, without an intelligent human photographer to select the framing and focus on the animals of interest, are of even poorer quality, with high rates of occlusion and high spatiotemporal similarity within image bursts. We employ an image filtering pipeline incorporating animal detection, species identification, viewpoint estimation, quality evaluation and temporal subsampling to compensate for these factors and obtain individual crops from camera trap and GGR images of suitable quality for re‐ID. We then employ the local clusterings and their alternatives (LCA) algorithm, a hybrid computer vision and graph clustering method for animal re‐ID, on the resulting high‐quality crops. Our method processed images taken during GGR‐16 and GGR‐18 in Meru County, Kenya, into 4142 highly comparable annotations, requiring only 120 contrastive same‐vs‐different‐individual decisions from a human reviewer to produce a population estimate of 349 individuals (within 4.6 of the ground truth count in Meru County). Our method also efficiently processed 8.9M unlabelled camera trap images from 70 camera traps at Mpala over 2 years into 685 encounters of 173 unique individuals, requiring only 331 contrastive decisions from a human reviewer.

In this position paper, we argue that application-driven research has been systemically under-valued in the machine learning community. As applications of machine learning proliferate, innovative algorithms inspired by specific real-world challenges have become increasingly important. Such work offers the potential for significant impact not merely in domains of application but also in machine learning itself. In this paper, we describe the paradigm of application-driven research in machine learning, contrasting it with the more standard paradigm of methods-driven research. We illustrate the benefits of application-driven machine learning and how this approach can productively synergize with methods-driven work. Despite these benefits, we find that reviewing, hiring, and teaching practices in machine learning often hold back application-driven innovation. We outline how these processes may be improved.

We introduce INQUIRE, a text-to-image retrieval benchmark designed to challenge multimodal vision-language models on expert-level queries. INQUIRE includes iNaturalist 2024 (iNat24), a new dataset of five million natural world images, along with 250 expert-level retrieval queries. These queries are paired with all relevant images comprehensively labeled within iNat24, comprising 33,000 total matches. Queries span categories such as species identification, context, behavior, and appearance, emphasizing tasks that require nuanced image understanding and domain expertise. Our benchmark evaluates two core retrieval tasks: (1) INQUIRE-Fullrank, a full dataset ranking task, and (2) INQUIRE-Rerank, a reranking task for refining top-100 retrievals. Detailed evaluation of a range of recent multimodal models demonstrates that INQUIRE poses a significant challenge, with the best models failing to achieve an mAP@50 above 50%. In addition, we show that reranking with more powerful multimodal models can enhance retrieval performance, yet there remains a significant margin for improvement. By focusing on scientifically-motivated ecological challenges, INQUIRE aims to bridge the gap between AI capabilities and the needs of real-world scientific inquiry, encouraging the development of retrieval systems that can assist with accelerating ecological and biodiversity research.

Abstract Camera traps facilitate non‐invasive wildlife monitoring, but their widespread adoption has created a data processing bottleneck: a camera trap survey can create millions of images, and the labour required to review those images strains the resources of conservation organisations. AI is a promising approach for accelerating image review, but AI tools for camera trap data are imperfect; in particular, classifying small animals remains difficult, and accuracy falls off outside the ecosystems in which a model was trained. It has been proposed that incorporating an object detector into an image analysis pipeline may help address these challenges, but the benefit of object detection has not been systematically evaluated in the literature. In this work, the authors assess the hypothesis that classifying animals cropped from camera trap images using a species‐agnostic detector yields better accuracy than classifying whole images. We find that incorporating an object detection stage into an image classification pipeline yields a macro‐average F1 improvement of around 25% on a large, long‐tailed dataset; this improvement is reproducible on a large public dataset and a smaller public benchmark dataset. The authors describe a classification architecture that performs well for both whole and detector‐cropped images, and demonstrate that this architecture yields state‐of‐the‐art benchmark accuracy.

Monitoring animal behavior can facilitate conservation efforts by providing key insights into wildlife health, population status, and ecosystem function. Automatic recognition of animals and their behaviors is critical for capitalizing on the large unlabeled datasets generated by modern video devices and for accelerating monitoring efforts at scale. However, the development of automated recognition systems is currently hindered by a lack of appropriately labeled datasets. Existing video datasets 1) do not classify animals according to established biological taxonomies; 2) are too small to facilitate large-scale behavioral studies and are often limited to a single species; and 3) do not feature temporally localized annotations and therefore do not facilitate localization of targeted behaviors within longer video sequences. Thus, we propose MammalNet, a new large-scale animal behavior dataset with taxonomy-guided annotations of mammals and their common behaviors. MammalNet contains over 18K videos totaling 539 hours, which is ~10 times larger than the largest existing animal behavior dataset [36]. It covers 17 orders, 69 families, and 173 mammal categories for animal categorization and captures 12 high-level animal behaviors that received focus in previous animal behavior studies. We establish three benchmarks on MammalNet: standard animal and behavior recognition, compositional low-shot animal and behavior recognition, and behavior detection. Our dataset and code have been made available at: https://mammalnet.github.io.

Generalization to novel domains is a fundamental chal-lenge for computer vision. Near-perfect accuracy on bench-marks is common, but these models do not work as expected when deployed outside of the training distribution. To build computer vision systems that truly solve real-world prob-lems at global scale, we need benchmarks that fully capture real-world complexity, including geographic domain shift, long-tailed distributions, and data noise. We propose urban forest monitoring as an ideal testbed for studying and improving upon these computer vision challenges, while working towards filling a crucial environ-mental and societal need. Urban forests provide significant benefits to urban societies. However, planning and main-taining these forests is expensive. One particularly costly aspect of urban forest management is monitoring the ex-isting trees in a city: e.g., tracking tree locations, species, and health. Monitoring efforts are currently based on tree censuses built by human experts, costing cities millions of dollars per census and thus collected infrequently. Previous investigations into automating urban forest monitoring focused on small datasets from single cities, covering only common categories. To address these short-comings, we introduce a new large-scale dataset that joins public tree censuses from 23 cities with a large collection of street level and aerial imagery. Our Auto Arborist dataset contains over 2.5M trees and 344 genera and is >2 or-ders of magnitude larger than the closest dataset in the literature. We introduce baseline results on our dataset across modalities as well as metrics for the detailed analy-sis of generalization with respect to geographic distribution shifts, vital for such a system to be deployed at-scale.

Recent progress in self-supervised learning has resulted in models that are capable of extracting rich representations from image collections without requiring any explicit label supervision. However, to date the vast majority of these approaches have restricted themselves to training on standard benchmark datasets such as ImageNet. We argue that fine-grained visual categorization problems, such as plant and animal species classification, provide an informative testbed for self-supervised learning. In order to facilitate progress in this area we present two new natural world visual classification datasets, iNat2021 and NeWT. The former consists of 2.7M images from 10k different species up-loaded by users of the citizen science application iNaturalist. We designed the latter, NeWT, in collaboration with domain experts with the aim of benchmarking the performance of representation learning algorithms on a suite of challenging natural world binary classification tasks that go beyond standard species classification. These two new datasets allow us to explore questions related to large-scale representation and transfer learning in the context of fine-grained categories. We provide a comprehensive analysis of feature extractors trained with and without supervision on ImageNet and iNat2021, shedding light on the strengths and weaknesses of different learned features across a diverse set of tasks. We find that features produced by standard supervised methods still outperform those produced by self-supervised approaches such as SimCLR. However, improved self-supervised learning methods are constantly being released and the iNat2021 and NeWT datasets are a valuable resource for tracking their progress.

Conservation science depends on an accurate understanding of what’s happening in a given ecosystem. How many species live there? What is the makeup of the population? How is that changing over time? Species Distribution Modeling (SDM) seeks to predict the spatial (and sometimes temporal) patterns of species occurrence, i.e. where a species is likely to be found. The last few years have seen a surge of interest in applying powerful machine learning tools to challenging problems in ecology [2, 5, 8]. Despite its considerable importance, SDM has received relatively little attention from the computer science community. Our goal in this work is to provide computer scientists with the necessary background to read the SDM literature and develop ecologically useful ML-based SDM algorithms. In particular, we introduce key SDM concepts and terminology, review standard models, discuss data availability, and highlight technical challenges and pitfalls.

African elephants are vital to their ecosystems, but their populations are threatened by a rise in human-elephant conflict and poaching. Monitoring population dynamics is essential in conservation efforts; however, tracking elephants is a difficult task, usually relying on the invasive and sometimes dangerous placement of GPS collars. Although there have been many recent successes in the use of computer vision techniques for automated identification of other species, identification of elephants is extremely difficult and typically requires expertise as well as familiarity with elephants in the population. We have built and deployed a web-based platform and database for human-in-the-loop re-identification of elephants combining manual attribute labeling and state-of-the-art computer vision algorithms, known as ElephantBook. Our system is currently in use at the Mara Elephant Project, helping monitor the protected and at-risk population of elephants in the Greater Maasai Mara ecosystem. ElephantBook makes elephant re-identification usable by non-experts and scalable for use by multiple conservation NGOs.

Distribution shifts—where the training distribution differs from the test distribution—can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity in the real-world deployments, these distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated benchmark of 10 datasets reflecting a diverse range of distribution shifts that naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training yields substantially lower out-of-distribution than in-distribution performance. This gap remains even with models trained by existing methods for tackling distribution shifts, underscoring the need for new methods for training models that are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. The full paper, code, and leaderboards are available at https://wilds.stanford.edu.

In static monitoring cameras, useful contextual information can stretch far beyond the few seconds typical video understanding models might see: subjects may exhibit similar behavior over multiple days, and background objects remain static. Due to power and storage constraints, sampling frequencies are low, often no faster than one frame per second, and sometimes are irregular due to the use of a motion trigger. In order to perform well in this setting, models must be robust to irregular sampling rates. In this paper we propose a method that leverages temporal context from the unlabeled frames of a novel camera to improve performance at that camera. Specifically, we propose an attention-based approach that allows our model, Context R-CNN, to index into a long term memory bank constructed on a per-camera basis and aggregate contextual features from other frames to boost object detection performance on the current frame. We apply Context R-CNN to two settings: (1) species detection using camera traps, and (2) vehicle detection in traffic cameras, showing in both settings that Context R-CNN leads to performance gains over strong baselines. Moreover, we show that increasing the contextual time horizon leads to improved results. When applied to camera trap data from the Snapshot Serengeti dataset, Context R-CNN with context from up to a month of images outperforms a single-frame baseline by 17.9% mAP, and outperforms S3D (a 3d convolution based baseline) by 11.2% mAP.

The ability to detect and classify rare occurrences in images has important applications - for example, counting rare and endangered species when studying biodiversity, or detecting infrequent traffic scenarios that pose a danger to self-driving cars. Few-shot learning is an open problem: current computer vision systems struggle to categorize objects they have seen only rarely during training, and collecting a sufficient number of training examples of rare events is often challenging and expensive, and sometimes outright impossible. We explore in depth an approach to this problem: complementing the few available training images with ad-hoc simulated data.Our testbed is animal species classification, which has a real-world long-tailed distribution. We present two natural world simulators, and analyze the effect of different axes of variation in simulation, such as pose, lighting, model, and simulation method, and we prescribe best practices for efficiently incorporating simulated data for real-world performance gain. Our experiments reveal that synthetic data can considerably reduce error rates for classes that are rare, that as the amount of simulated data is increased, accuracy on the target class improves, and that high variation of simulated data provides maximum performance gain.

Camera trapping is used by conservation biologists to study snow leopards. In this research, we introduce techniques that find motion in camera trap images. Images are grouped into sets and a common background image is computed for each set. The background and superpixel-based features are then used to segment each image into objects that correspond to motion. The proposed methods are robust to changes in illumination due to time of day or the presence of camera flash.