Metraipolis › archive › dc_7b06648fa0e2

TriCLE: Tri-Modal Vision-Language Reasoning for Edge-Deployed Fine-Grained Clustering — arXiv:2608.04175

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.04175
License (as archived)
CC0-1.0
Archived
2026-10-31T11:47:47.304Z
SHA-256 of the archived text
e70918fe697d22bd2d1d8da47ca1bea4df5818cfad0bfaad93ea52c1af1c4e40
Stable URL of this page
https://metraipolis.com/documento/dc_7b06648fa0e2

How to cite

TriCLE: Tri-Modal Vision-Language Reasoning for Edge-Deployed Fine-Grained Clustering — arXiv:2608.04175 — arXiv (en), https://arxiv.org/abs/2608.04175 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7b06648fa0e2 (Metraipolis archive dc_7b06648fa0e2, sha256 e70918fe697d22bd2d1d8da47ca1bea4df5818cfad0bfaad93ea52c1af1c4e40)
BibTeX
@misc{dc7b06648fa0e2,
  title = {TriCLE: Tri-Modal Vision-Language Reasoning for Edge-Deployed Fine-Grained Clustering — arXiv:2608.04175},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.04175},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7b06648fa0e2; sha256 e70918fe697d22bd2d1d8da47ca1bea4df5818cfad0bfaad93ea52c1af1c4e40},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04175. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Edge platforms used for aerial observation must interpret aircraft imagery under limited memory, limited compute, and intermittent connectivity. This setting is difficult for standard RGB-only recognition models and general-purpose vision-language models, especially when calibrated thermal and LiDAR aircraft data are unavailable. We present TriCLE, an application-oriented tri-modal vision-language system for aircraft taxonomic grouping under edge constraints. From a single RGB aircraft image, TriCLE generates a structure-preserving FLIR-style thermal view and a pseudo-LiDAR depth projection, then fuses the aligned views with task instructions in a compact Qwen3-VL backbone. The model is aligned to an expert aircraft taxonomy based on propulsion, airframe family, size, design era, and configuration, so its outputs reflect engineering-relevant similarity rather than only surface appearance. We evaluate supervised fine-tuning, rotation-preserving SFT, and three policy-alignment strategies: GRPO, GSPO, and DAPO. Sequence-level GSPO gives the strongest validation performance, reaching 88.33\% validation accuracy and 0.91 weighted F1 on valid aircraft outputs. On a held-out aircraft test partition, GSPO achieves 78.00\% accuracy and 0.793 weighted F1 while preserving 94.00\% parseable output formatting. After 4-bit quantization and attention-memory optimization, the aligned 4B model fits an 8GB deployment target and processes each tri-modal triplet in 1.48 seconds. These results support TriCLE as a practical prototype for interpretable, edge-feasible aircraft grouping, while emphasizing the need for further validation on real aligned thermal and LiDAR sensor streams.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-06 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.04175 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T11:37:40.649Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).