PrecisionTrack: A Platform for Automated Long-Term Social Behavior Analysis in Naturalized Environments
Bibliographic record
Abstract
Abstract Large-scale ethological behavioral studies can provide insights into the neuronal processes underlying complex and social behaviors, potentially opening new avenues for mental health research. However, studying socially interacting animals in naturalistic environments remains technically challenging, as current approaches struggle to simultaneously maintain subject identity, extract behavior, and characterize social interactions over prolonged periods. Here, we present PrecisionTrack , an open-source and fully integrated framework designed for real-time multi-animal tracking, behavioral analysis, and social interaction inference in large groups of interacting animals. PrecisionTrack achieves high spatiotemporal accuracy in crowded and highly occlusive environments while maintaining robust long-term identity tracking and low-latency processing. To extend behavioral inference beyond pose estimation, we developed the Multi-animal Action Recognition Transformer ( MART ), a transformer-based architecture enabling real-time subject-level action recognition, and Graph-MART ( G-MART ), a graph neural network module that infers directed social interactions and interaction partners within groups. In addition, PrecisionTrack supports quantitative analysis of evolving social networks across time, enabling investigation of the temporal organization and stability of social dynamics in naturalistic settings. The entire framework is open source and accompanied by standardized workflows and documentation, enabling users to train, evaluate, and deploy custom behavioral analysis pipelines across species and experimental contexts. PrecisionTrack provides a scalable platform for quantitative investigation of complex social behaviors at a resolution and duration not accessible with existing methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".