Foundation models have enabled rapid progress across many specialized domains by leveraging large-scale pre-training on unlabeled data, demonstrating strong generalization to a variety of downstream tasks. While such models have gained significant attention in fields like Earth Observation, their application to Mars science remains limited. A key enabler of progress in other domains has been the availability of standardized benchmarks that support systematic evaluation. In contrast, Mars science lacks such benchmarks and standardized evaluation frameworks, which have limited progress toward developing foundation models for Martian tasks. To address this gap, we introduce Mars-Bench, the first benchmark designed to systematically evaluate models across a broad range of Mars-related tasks using both orbital and surface imagery. Mars-Bench comprises 20 datasets spanning classification, segmentation, and object detection, focused on key geologic features such as craters, cones, boulders, and frost. We provide standardized, ready-to-use datasets and baseline evaluations using models pre-trained on natural images, Earth satellite data, and state-of-the-art vision-language models. Results from all analyses suggest that Mars-specific foundation models may offer advantages over general-domain counterparts, motivating further exploration of domain-adapted pre-training. Mars-Bench aims to establish a standardized foundation for developing and comparing machine learning models for Mars science. Our data, models, and code are available at: https://mars-bench.github.io/.

Active Label Correction (ALC) is an interactive method that cleans an established training set of mislabeled examples in conjunction with a domain expert. ALC presumes that the expert who conducts this review is either more accurate than the original annotator or has access to additional resources that ensure a high quality label. A high-cost re-review is possible because ALC proceeds iteratively, scoring the full training set but selecting only small batches of examples that are likely mislabeled. The expert reviews each batch and corrects any mislabeled examples, after which the classifier is retrained and the process repeats until the expert terminates it. We compare several instantiations of ALC to fully-automated methods that attempt to discard or correct label noise in a single pass. Our empirical results show that ALC outperforms single-pass methods in terms of selection efficiency and classifier accuracy. We evaluate the best ALC instantiation on our motivating task of detecting mislabeled and poorly formulated sites within a land cover classification training set from the geography domain.

Orbital remote sensing provides a powerful way to efficiently survey targets such as the Earth and other planets and moons for features of interest. One such feature of astrobiological relevance is the presence of surface sulfur deposits. These deposits have been observed to be associated with microbial activity at the Borup Fiord glacial springs in Canada, a location that may provide an analogue to other icy environments such as Europa. This article evaluates automated classifiers for detecting sulfur in remote sensing observations by the hyperion spectrometer on the EO-1 spacecraft. We determined that a data-driven machine learning solution was needed because the sulfur could not be detected by simply matching observations to sulfur lab spectra. We also evaluated several methods (manual and automated) for identifying the most relevant attributes (spectral wavelengths) needed for successful sulfur detection. Our findings include (1) the Borup Fiord sulfur deposits were best modeled as containing two sub-populations: sulfur on ice and sulfur on rock; (2) as expected, classifiers using Gaussian kernels outperformed those based on linear kernels, and should be adopted when onboard computational constraints permit; and (3) Recursive Feature Elimination selected sensible and effective features for use in the computationally constrained environment onboard EO-1. This study helped guide the selection of algorithm parameters and configuration for the classification system currently operational on EO-1. Finally, we discuss implications for a similar onboard classification system for a future Europa orbiter.

This paper presents two machine learning techniques that greatly reduce the number of person-hours required to generate high-quality training data for land cover classification. The first technique uses active learning to guide the generation of training data by selecting only the most informative examples for labeling. The second technique identifies and mitigates the impact of mislabeled instances. Both techniques are tested on data from NASA's Moderate Resolution Imaging Spectroradiometer (MODIS), which has required thousands of person hours to label. Our results shows that the active learning method requires fewer labeled examples than random sampling to produce a high quality classifier. Our results on class noise mitigation show that if mislabelings occur, we can further improve classifier accuracy, and that weighting instances by their label confidence outperforms an analogous method that discards suspected mislabelings. If combined, these methods have the potential to make training data generation a more efficient and reliable process.

The problem of finding unusual time series has recently attracted much attention, and several promising methods are now in the literature. However, virtually all proposed methods assume that the data reside in main memory. For many real-world problems this is not be the case. For example, in astronomy, multi-terabyte time series datasets are the norm. Most current algorithms faced with data which cannot fit in main memory resort to multiple scans of the disk/tape and are thus intractable. In this work we show how one particular definition of unusual time series, the time series discord, can be discovered with a disk aware algorithm. The proposed algorithm is exact and requires only two linear scans of the disk with a tiny buffer of main memory. Furthermore, it is very simple to implement. We use the algorithm to provide further evidence of the effectiveness of the discord definition in areas as diverse as astronomy, Web query mining, video surveillance, etc., and show the efficiency of our method on datasets which are many orders of magnitude larger than anything else attempted in the literature.