pith:O5FPJH7I
TrajTok: Learning Trajectory Tokens enables better Video Understanding
TrajTok learns trajectory tokens end-to-end through implicit space-time clustering to improve video model accuracy and efficiency.
arxiv:2602.22779 v3 · 2026-02-26 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{O5FPJH7I2B4HNHMA5SHOEQKEEE}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
With TrajTok, we implement a video CLIP model trained from scratch (TrajViT2). It achieves the best accuracy at scale across both classification and retrieval benchmarks, while maintaining efficiency comparable to the best token-merging methods.
That implicit clustering of pixels in space and time will produce trajectories that are semantically useful for downstream video understanding tasks when the segmenter is co-trained only for adaptability rather than pixel-level fidelity.
TrajTok learns adaptive trajectory tokens for videos through a unified end-to-end segmenter, improving understanding performance and efficiency over patch-based or external-pipeline tokenizers.
Formal links
Receipt and verification
| First computed | 2026-06-04T01:08:46.329945Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
774af49fe8d078769d80ec8ee24144212348484984fc484b38ae6301fddf0def
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/O5FPJH7I2B4HNHMA5SHOEQKEEE \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 774af49fe8d078769d80ec8ee24144212348484984fc484b38ae6301fddf0def
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "7c063636c3c98fd927b376d4a70fb97379984d7bda9b537f2c7cdd773ad63a2d",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-02-26T09:15:34Z",
"title_canon_sha256": "72079029b17a646fb5973a5cdd367d3c100b454490f0e26fa12335cd158e1b76"
},
"schema_version": "1.0",
"source": {
"id": "2602.22779",
"kind": "arxiv",
"version": 3
}
}