Pith. sign in

REVIEW 5 major objections 4 minor 41 references

TerraIncognita: A Dynamic Benchmark for Species Discovery Using Frontier Models

T0 review · 5 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper introduces TerraIncognita, a benchmark that pairs images of well-known insect species with field-collected images of rare, possibly undescribed species, and asks twelve frontier vision-language models to classify each image down…

desk verdict Useful new benchmark with a robust coarse-to-fine performance drop, but the discovery-accuracy metric needs precise definition and released code before the headline spread is credible. read the letter →

arxiv 2506.03182 v1 pith:6QB6LDC7 submitted 2025-05-29 cs.CV cs.LG

Shivani Chiranjeevi , Hossein Zaremehrjerdi , Zi K. Deng , Talukder Z. Jubery , Ari Grele , Arti Singh , Asheesh K Singh , Soumik Sarkar
show 4 more authors
Nirav Merchant Harold F. Greeney Baskar Ganapathysubramanian Chinmay Hegde
This is my paper
classification cs.CVcs.LG
keywords insectbiodiversitybenchmarkdatasetvision-languagemodelsopen-setrecognitionzero-shotclassificationtaxonomichierarchyout-of-distributiondetectionspeciesdiscovery
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces TerraIncognita, a benchmark that pairs images of well-known insect species with field-collected images of rare, possibly undescribed species, and asks twelve frontier vision-language models to classify each image down the taxonomic hierarchy (Order, Family, Genus, Species) while abstaining when uncertain. It finds a steep competence gradient: on known species, top models exceed 90 percent F1 at the Order level but fall below 2 percent at the Species level, and on novel specimens, the ability to correctly abstain (discovery accuracy) ranges from 55 to 88 percent across models. The benchmark is designed to test whether AI can move from classifying the known to flagging the unknown, something central to biodiversity conservation.

What carries the argument

The central object is the TerraIncognita dataset itself: 200 species (100 known, 100 novel) with roughly 450 high-resolution images, where known images come from iNaturalist research-grade entries and novel images come from light-trap field collection in Central and South America with partial expert labels. Evaluation uses a fixed zero-shot prompt asking for the four-level hierarchy with 'Unknown' for uncertain levels; metrics include discovery accuracy (correct abstention on novel and confident species-level on known), hierarchical F1, and expert-reviewed explanation categories such as morphological hallucination and taxonomic overreach.

What would settle it

Conduct a reverse image search of the 237 novel images against public and commercial image corpora likely used in model training; if a substantial fraction return near-duplicate matches, the discovery accuracy results would measure memorization. Alternatively, run the identical prompt on an open-weight model whose training data is known to exclude these species and compare its discovery accuracy to the API models; if the open model performs similarly, the claim of open-world generalization is strengthened, but if it performs much worse, the API models' scores may reflect training contamination.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that current frontier vision-language models possess coarse taxonomic competence but not fine-grained species identification under zero-shot conditions. Across 100 known species and 100 novel, field-collected species, the same models that exceed 90 percent Order-level F1 score below 2 percent Species-level F1, and on novel specimens they either overcommit to unsupported genus or species labels or abstain, with no consistent strategy across model families. The paper interprets this as evidence that open-world species discovery is beyond current models, while coarse-level triage is feasible.

Load-bearing premise

The load-bearing premise is that the 100 field-collected Novel specimens are effectively unseen by the evaluated models; the paper acknowledges it cannot verify the training data of closed API models, so the discovery accuracy results could partly reflect familiarity instead of open-world generalization.

Editorial extensions

If this is right

  • Frontier models could serve as automatic pre-screeners that flag likely novel specimens for expert review, but cannot yet replace taxonomists at the species level.
  • The sharp order-to-species drop suggests model representations encode coarse morphological or contextual cues but not diagnostic fine-grained traits.
  • The wide variance in discovery accuracy (55-88 percent) shows abstention behavior is not reliable and needs targeted training or calibration.
  • Quarterly dataset refresh with new field specimens provides a way to longitudinally track whether future models improve or simply memorize familiar images.
  • The qualitative failure modes (hallucination, overreach) argue for explanation-aware evaluation in addition to label accuracy.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the novel images are indeed absent from training corpora, the 55-88 percent discovery accuracy reflects genuine open-set detection; for closed API models that assumption remains unverifiable.
  • The contrast between near-perfect Order accuracy and near-zero Species accuracy on the same images suggests models rely on global shape or background cues rather than diagnostic anatomy, a hypothesis that could be tested with occlusion or cropping experiments.
  • The benchmark's design could be extended beyond insects to other hyperdiverse taxa, and its quarterly updates could double as a contamination monitor for future model releases.
  • A testable extension is to fine-tune an open model on iNaturalist species and re-evaluate on the same known set; if the Species-level gap does not close, the 2 percent result points to dataset bias rather than model limits.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper introduces TerraIncognita, a benchmark for evaluating vision-language models on hierarchical taxonomic classification of insects, using two subsets: Known species (iNaturalist images with full labels) and Novel species (field-collected images of rare, potentially undescribed taxa with only partial labels). The authors evaluate 12 frontier multimodal models under a unified zero-shot prompting protocol, reporting order-level F1 above 90% for some models, species-level F1 below 2% for most, and a discovery-accuracy range of 55–88% on novel specimens. They also provide a qualitative expert review of model explanations, identifying failure modes such as morphological hallucination and taxonomic overreach, and commit to quarterly dataset updates.

Significance. If the reported numbers hold, TerraIncognita is a valuable addition to biodiversity AI evaluation: it uses real field-collected images of rare taxa rather than simulated novelty, involves expert entomological curation, evaluates a broad set of frontier models under a controlled prompt, and includes a structured qualitative analysis of model reasoning. The sharp drop from coarse- to fine-grained taxonomic accuracy is a clinically clear finding, and the public dataset plus longitudinal update plan are concrete strengths. However, the central quantitative claims currently rest on an underspecified metric and an unverifiable novelty assumption, so the benchmark's immediate usefulness depends on whether these issues are resolved.

major comments (5)
  1. [§4.1, Table 3] The discovery-accuracy metric is not precisely defined. The text says that for Novel images, 'the correct behavior is abstention (i.e., predicting “Unknown”) at the Species or higher levels, depending on label availability.' Because Table 1 reports that the Novel set has 43 distinct genera and only 12 species-level labels, it is unclear whether a model must abstain at the finest available label or only at Species. Without a per-image scoring rule, the reported values (55.27–87.76) cannot be reproduced. Please provide the exact algorithm, the raw per-image predictions, and code that maps predictions to the reported numbers.
  2. [§3.1, §5] The claim that the Novel specimens are 'unseen by the model' is asserted rather than verified. Since the evaluated models are closed API systems, the paper cannot rule out that the 100 Novel images or near-duplicates appeared in training data. The Section 3.1 clarification appropriately softens the language to 'very rare, and for which almost surely few (or no) labeled images exist,' but the abstract and Section 5 still interpret the 55–88% numbers as open-world discovery accuracy. This is a load-bearing assumption for the paper's central message. I recommend reframing the results as measuring behavior on rare, expert-curated specimens, and adding a control experiment, such as testing on iNaturalist species that were collected after model training cutoffs, to bound the contamination risk.
  3. [Abstract, Table 4] The abstract's statement that species-level F1 drops 'below 2%' is contradicted by Table 4, where Gemini-2.5-Flash achieves 3.00% Species F1. The accurate claim is that all models are below 3%, and only one exceeds 2%. Please correct the abstract and the corresponding sentence in Section 5 so that the quantitative summary matches the reported data.
  4. [Tables 3 and 4] The evaluation reports point estimates without confidence intervals or significance tests. Given that the Novel subset has only 237 images and the Known subset 200, the spread among mid-range discovery-accuracy values (e.g., 75.11 vs. 75.53) is within binomial noise, and even the reported range of 55.27–87.76 would benefit from interval estimates. The paper states that 'All TerraIncognita data, results, and future updates are available,' but the current supplement does not include the raw per-image model outputs, the scoring script, or code to reproduce Tables 3 and 4. Please release these artifacts to make the benchmark reproducible.
  5. [Table 1, §7.3] There is an internal inconsistency about the number of Novel specimens with genus-level labels. Table 1 reports 43 distinct genera and 12 species-level labels in the Unknown category, while Section 7.3 states that 'we include five examples labeled up to Genus.' Since the discovery-accuracy rubric depends on available label depth, this discrepancy affects the denominator of the metric and must be reconciled.
minor comments (4)
  1. [Table 3] The column header 'F1.' is unexplained; please define how the F1 value combines Known and Novel discovery accuracy, or rename it to something unambiguous.
  2. [Figure 3] The figure caption includes the stray text 'Loading [MathJax]/extensions/MathMenu.js', which appears to be a rendering artifact and should be removed.
  3. [§5, Comparative trends] The observation 'Gemini 2.5 models' is inaccurate because only one Gemini model (Gemini-2.5-Flash) is evaluated.
  4. [§4 and §7.2] The prompt description in Figure 4 asks for explanations of 50–60 words per level, while Prompt 1 in Section 7.2 requests no explanations and Prompt 2 requests a comma-separated format. Please clarify which prompt was used for which quantitative result, since both are referenced as the evaluation framework.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an empirical benchmark evaluation with no fitted parameters or derived predictions; the acknowledged unverifiable novelty assumption is a construct-validity caveat, not a circular derivation.

full rationale

This paper introduces a dataset and reports measured performance of frontier VLMs. There is no derivation chain, no fitted parameter renamed as a prediction, and no equation whose output is defined by its input. The only potentially circular-looking move is the definition of 'novel' in Section 3.1 as 'species that are unseen by the model — that is, species that are very rare, and for which almost surely few (or no) labeled images exist in the model’s training data,' followed by interpreting abstention as discovery accuracy. However, this is not a formal reduction: the benchmark's claim is conditional on that assumption, and the paper explicitly disclaims 'new to science' status and acknowledges that rigorous verification is outside its scope. If the assumption fails, the results would be less informative, but the evaluation is not constructed so that the desired outcome is forced by definition. A few citations involve overlapping authors (e.g., BioTrove [10] and OOD work [23]), but they are used as related work and are not load-bearing for the headline results, which come from newly collected field images and live API queries. No self-definitional, fitted-input, or self-citation circularity is exhibited, so the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The benchmark introduces no fitted parameters or new physical entities. Its numerical claims rest on five unverified domain assumptions, most importantly that novel images are absent from API training sets.

assumptions (5)
  • domain assumption Novel species are almost surely absent from VLM training data.
    Section 2 and Section 3.1 define novel as unseen by models and admit this cannot be verified for closed API models; discovery accuracy depends on this.
  • domain assumption iNaturalist research-grade labels are accurate ground truth for known species.
    Section 3 uses community-vetted iNaturalist entries; no independent taxonomic verification is reported.
  • domain assumption Expert Order and Family labels for novel specimens are reliable enough for scoring.
    Section 2 states labels are typically limited to Order or Family and were assigned by field experts; no formal validation is given.
  • domain assumption Closed API model outputs are zero-shot and untainted by benchmark-specific fine-tuning.
    Section 4 states no few-shot exemplars or retrieval augmentation were used, but the training sets of the API models cannot be inspected.
  • domain assumption The small, Lepidoptera-skewed sample supports reported aggregate F1 differences.
    Section 6 and supplementary Section 7.1 note limited scale and taxonomic skew; the paper provides no confidence intervals or stratified results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TerraIncognita: A Dynamic Benchmark for Species Discovery Using Frontier Models." pith.science (2026). https://pith.science/paper/6QB6LDC7

@misc{pith2026250603182,
  author       = {Pith},
  title        = {Pith review of: TerraIncognita: A Dynamic Benchmark for Species Discovery Using Frontier Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6QB6LDC7}},
  note         = {Machine review of arXiv:2506.03182}
}
abstract

The rapid global loss of biodiversity, particularly among insects, represents an urgent ecological crisis. Current methods for insect species discovery are manual, slow, and severely constrained by taxonomic expertise, hindering timely conservation actions. We introduce TerraIncognita, a dynamic benchmark designed to evaluate state-of-the-art multimodal models for the challenging problem of identifying unknown, potentially undescribed insect species from image data. Our benchmark dataset combines a mix of expertly annotated images of insect species likely known to frontier AI models, and images of rare and poorly known species, for which few/no publicly available images exist. These images were collected from underexplored biodiversity hotspots, realistically mimicking open-world discovery scenarios faced by ecologists. The benchmark assesses models' proficiency in hierarchical taxonomic classification, their capability to detect and abstain from out-of-distribution (OOD) samples representing novel species, and their ability to generate explanations aligned with expert taxonomic knowledge. Notably, top-performing models achieve over 90\% F1 at the Order level on known species, but drop below 2\% at the Species level, highlighting the sharp difficulty gradient from coarse to fine taxonomic prediction (Order $\rightarrow$ Family $\rightarrow$ Genus $\rightarrow$ Species). TerraIncognita will be updated regularly, and by committing to quarterly dataset expansions (of both known and novel species), will provide an evolving platform for longitudinal benchmarking of frontier AI methods. All TerraIncognita data, results, and future updates are available \href{https://baskargroup.github.io/TerraIncognita/}{here}.

Figures

Figures reproduced from arXiv: 2506.03182 by the authors.

Figure 1
Figure 1. A few example images from the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the TerraIncognita data collection pipeline. Insect specimens are first captured [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Treemap showing the hierarchical taxonomic distribution of [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Evaluation Dimensions for Open-World Insect Classification. We evaluate models along both quantitative and qualitative dimensions using a shared entomology-based prompt. This prompt asks the model to identify insect taxonomy at four levels — Order, Family, Genus, Speci…
Figure 5
Figure 5. Figure 5: Distribution of species by order and family for [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Taxonomic classification accuracy across models for both [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

41 extracted references · 34 canonical work pages

  1. [1]

    Numbers of insects (species and individuals)

    Smithsonian Institution. Numbers of insects (species and individuals). https://www.si.edu/ spotlight/buginfo/bugnos. Accessed: 2025-05-06

    work page 2025

  2. [2]

    Nigel E Stork, Michael JW Boyle, Carl Wardhaugh, and Roger A Beaver. What can an analysis of australian tropical rainforest bark beetles suggest about the missing millions of earth’s insect species? Insect Conservation and Diversity, 17(6):1156–1166, 2024

    work page 2024

  3. [3]

    Terry L. Erwin. Tropical forests: Their richness in coleoptera and other arthropod species. The Coleopterists Bulletin, 36(1):74–75, 1982

    work page 1982

  4. [4]

    Forces of change

    Smithsonian Institution. Forces of change. Website, 2025. Accessed: May 14, 2025

    work page 2025

  5. [5]

    Xprize rainforest names limelight rainforest winner of biodi- versity tech competition

    XPRIZE Foundation. Xprize rainforest names limelight rainforest winner of biodi- versity tech competition. https://www.xprize.org/prizes/rainforest/articles/ xprize-rainforest-names-limelight-rainforest-winner-of-biodiversity-tech-competition ,

    work page

  6. [6]

    Our team | limelight rainforest

    Limelight Rainforest Team. Our team | limelight rainforest. https://www. limelightrainforest.org/ourteam, 2024. Accessed: 2025-05-15

    work page 2024

  7. [7]

    From bugs and birds to drones and dna

    Wake Forest Magazine. From bugs and birds to drones and dna. https://magazine.wfu. edu/2025/01/27/from-bugs-and-birds-to-drones-and-dna , 2025. Accessed: 2025- 05-15

    work page 2025

  8. [8]

    Experts in insect taxonomy threat- ened with extinction, reveals first european red list

    International Union for Conservation of Nature. Experts in insect taxonomy threat- ened with extinction, reveals first european red list. https://iucn.org/news/202212/ experts-insect-taxonomy-threatened-extinction-reveals-first-european-red-list , 12 2022. IUCN news article. Accessed: 2025-05-15

    work page 2022

Show all 41 references
  1. [9]

    BioCLIP: A vision foundation model for the tree of life

    Samuel Stevens, Jiaman Wu, Matthew J Thompson, Elizabeth G Campolongo, Chan Hee Song, David Edward Carlyn, Li Dong, Wasila M Dahdul, Charles Stewart, Tanya Berger-Wolf, Wei- Lun Chao, and Yu Su. BioCLIP: A vision foundation model for the tree of life. In Proceedings of the IEE...

    2024

  2. [10]

    Chih-Hsuan Yang, Benjamin Feuer, Zaki Jubery, Zi K. Deng, Andre Nakkab, Md Zahid Hasan, Shivani Chiranjeevi, Kelly Marshall, Nirmal Baishnab, Asheesh K Singh, Arti Singh, Soumik Sarkar, Nirav Merchant, Chinmay Hegde, and Baskar Ganapathysubramanian. Biotrove: A large curated i...

    2024

  3. [11]

    Wang, Xiaoliang Huo, Joakim Bruslund Haurum, Scott C

    ZeMing Gong, Austin T. Wang, Xiaoliang Huo, Joakim Bruslund Haurum, Scott C. Lowe, Graham W. Taylor, and Angel X. Chang. CLIBD: Bridging vision and genomics for biodiversity monitoring at scale. arXiv preprint, 2024

    2024

  4. [12]

    The inaturalist species classification and detection dataset

    Grant Van Horn, Oisin Mac Aodha, Yang Song, Yin Cui, Chen Sun, Alex Shepard, Hartwig Adam, Pietro Perona, and Serge Belongie. The inaturalist species classification and detection dataset. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8769–8778, 2018

    2018

  5. [13]

    What is gbif? https://www.gbif.org/ what-is-gbif

    GBIF: The Global Biodiversity Information Facility. What is gbif? https://www.gbif.org/ what-is-gbif. Accessed: 2025-05-05

    2025

  6. [14]

    Benchmarking Representation Learning for Natural World Image Collections

    Grant Van Horn, Elijah Cole, Sara Beery, Kimberly Wilber, Serge Belongie, and Oisin MacAodha. Benchmarking Representation Learning for Natural World Image Collections . In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 12879–12888, Los Alamit...

    2021

  7. [15]

    Scheirer, Anderson de Rezende Rocha, Archana Sapkota, and Terrance E

    Walter J. Scheirer, Anderson de Rezende Rocha, Archana Sapkota, and Terrance E. Boult. Toward open set recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(7):1757–1772, 2013. 10

    2013

  8. [16]

    Yanayacu biological station and center for creative studies.https: //www.facebook.com/Yanayacustation/, 2025

    Yanayacu Biological Station. Yanayacu biological station and center for creative studies.https: //www.facebook.com/Yanayacustation/, 2025. Biological research station located in Cosanga, Napo, Ecuador. Accessed: 2025-05-15

    2025

  9. [17]

    Pitkin, Nadine Hilt, and Konrad Fiedler

    Gunnar Brehm, Linda M. Pitkin, Nadine Hilt, and Konrad Fiedler. Montane andean rain forests are a global diversity hotspot of geometrid moths. Journal of Biogeography, 32(9):1621–1627, 2005

    2005

  10. [18]

    A new led lamp for the collection of nocturnal lepidoptera and a spectral comparison of light-trapping lamps

    Gunnar Brehm. A new led lamp for the collection of nocturnal lepidoptera and a spectral comparison of light-trapping lamps. Nota Lepidopterologica, 40:87–108, 2017

    2017

  11. [19]

    Moths are strongly attracted to ultraviolet and blue radiation

    Gunnar Brehm, Julia Niermann, Luisa Maria Jaimes Nino, David Enseling, Thomas Jüstel, Jan Christoph Axmacher, Eric Warrant, and Konrad Fiedler. Moths are strongly attracted to ultraviolet and blue radiation. Insect Conservation and Diversity, 14(2):188–198, 2021

    2021

  12. [20]

    Lowe, Jaclyn T.A

    Zahra Gharaee, ZeMing Gong, Nicholas Pellegrino, Iuliia Zarubiieva, Joakim Bruslund Haurum, Scott C. Lowe, Jaclyn T.A. McKeown, Chris C.Y . Ho, Joschka McLeod, Yi-Yun C Wei, Jireh Agda, Sujeevan Ratnasingham, Dirk Steinke, Angel X. Chang, Graham W. Taylor, and Paul Fieguth. A ...

    2023

  13. [21]

    Lowe, ZeMing Gong, Pablo Millan Arias, Nicholas Pellegrino, Austin T

    Zahra Gharaee, Scott C. Lowe, ZeMing Gong, Pablo Millan Arias, Nicholas Pellegrino, Austin T. Wang, Joakim Bruslund Haurum, Iuliia Zarubiieva, Lila Kari, Dirk Steinke, Graham W. Taylor, Paul Fieguth, and Angel X. Chang. BIOSCAN-5M: A multimodal dataset for insect biodiversity....

    2024

  14. [22]

    A baseline for detecting misclassified and out-of-distribution examples in neural networks

    Dan Hendrycks and Kevin Gimpel. A baseline for detecting misclassified and out-of-distribution examples in neural networks. CoRR, abs/1610.02136, 2016

    2016 arXiv

  15. [23]

    Singh, Soumik Sarkar, Arti Singh, and Baskar Ganapathysubramanian

    Mojdeh Saadati, Aditya Balu, Shivani Chiranjeevi, Talukder Zaki Jubery, Asheesh K. Singh, Soumik Sarkar, Arti Singh, and Baskar Ganapathysubramanian. Out-of-distribution detection algorithms for robust insect classification. Plant Phenomics, 6:0170, 2024

    2024

  16. [24]

    Abhijit Bendale and Terrance E. Boult. Towards Open Set Deep Networks . In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , pages 1563–1572, Los Alamitos, CA, USA, June 2016. IEEE Computer Society

    2016

  17. [25]

    Out-of-distribution detection via deep multi-comprehension ensemble, 2024

    Chenhui Xu, Fuxun Yu, Zirui Xu, Nathan Inkawhich, and Xiang Chen. Out-of-distribution detection via deep multi-comprehension ensemble, 2024

    2024

  18. [26]

    Dietterich

    Dan Hendrycks, Mantas Mazeika, and Thomas G. Dietterich. Deep anomaly detection with outlier exposure. CoRR, abs/1812.04606, 2018

    2018 arXiv

  19. [27]

    Feed two birds with one scone: Exploiting wild data for both out-of-distribution generalization and detection, 2023

    Haoyue Bai, Gregory Canal, Xuefeng Du, Jeongyeol Kwon, Robert Nowak, and Yixuan Li. Feed two birds with one scone: Exploiting wild data for both out-of-distribution generalization and detection, 2023

    2023

  20. [28]

    Neco: Neural collapse based out-of-distribution detection, 2024

    Mouïn Ben Ammar, Nacim Belkhir, Sebastian Popescu, Antoine Manzanera, and Gianni Franchi. Neco: Neural collapse based out-of-distribution detection, 2024

    2024

  21. [29]

    Christian Schmidt, Aditya Jain, Yves Basset, Sara Beery, Maxim Larrivée, and David Rolnick

    Yuyan Chen, Nico Lang, B. Christian Schmidt, Aditya Jain, Yves Basset, Sara Beery, Maxim Larrivée, and David Rolnick. Open-set recognition of novel species in biodiversity monitoring, 2025

    2025

  22. [30]

    Insectset459: an open dataset of insect sounds for bioacoustic machine learning, 2025

    Marius Faiß, Burooj Ghani, and Dan Stowell. Insectset459: an open dataset of insect sounds for bioacoustic machine learning, 2025

    2025

  23. [31]

    Taxonomic bias in biodiversity data and societal preferences

    Julien Troudet, Philippe Grandcolas, Amandine Blin, Régine Vignes Lebbe, and Frédéric Legendre. Taxonomic bias in biodiversity data and societal preferences. Scientific Reports, 7, 12 2017. 11

    2017

  24. [32]

    Shiyu Liang, Yixuan Li, and R. Srikant. Principled detection of out-of-distribution examples in neural networks. CoRR, abs/1706.02690, 2017

    2017 arXiv

  25. [33]

    Owens, and Yixuan Li

    Weitang Liu, Xiaoyun Wang, John D. Owens, and Yixuan Li. Energy-based out-of-distribution detection. CoRR, abs/2010.03759, 2020

    2010 arXiv

  26. [34]

    Long-tailed out-of-distribution detection via normalized outlier distribution adaptation, 2024

    Wenjun Miao, Guansong Pang, Jin Zheng, and Xiao Bai. Long-tailed out-of-distribution detection via normalized outlier distribution adaptation, 2024

    2024

  27. [35]

    Learning multiple layers of features from tiny images

    Alex Krizhevsky. Learning multiple layers of features from tiny images. University of Toronto, 05 2012

    2012

  28. [36]

    Open-set recognition in the age of vision-language models

    Dimity Miller, Niko Sünderhauf, Alex Kenna, and Keita Mason. Open-set recognition in the age of vision-language models. In ECCV, page 1–18, Berlin, Heidelberg, 2024. Springer-Verlag

    2024

  29. [37]

    Benchmarking out-of-distribution detection in visual question answering

    Xiangxi Shi and Stefan Lee. Benchmarking out-of-distribution detection in visual question answering. In 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pages 5473–5483, 2024

    2024

  30. [38]

    Classifying the unknown: Insect identification with deep hierarchical bayesian learning

    Sarkhan Badirli, Christine Johanna Picard, George Mohler, Frannie Richert, Zeynep Akata, and Murat Dundar. Classifying the unknown: Insect identification with deep hierarchical bayesian learning. Methods in Ecology and Evolution, 14(6):1515–1530, 2023

    2023

  31. [39]

    Open-set recognition: a good closed-set classifier is all you need?, 2022

    Sagar Vaze, Kai Han, Andrea Vedaldi, and Andrew Zisserman. Open-set recognition: a good closed-set classifier is all you need?, 2022

    2022

  32. [40]

    From coarse to fine-grained open-set recognition

    Nico Lang, Vésteinn Snæbjarnarson, Elijah Cole, Oisin Mac Aodha, Christian Igel, and Serge Belongie. From coarse to fine-grained open-set recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 17804–17814, June 2024. 7 S...

    2024

  33. [2024]

    Accessed: 2025-05-15

    2025

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.