AI Paper Atlas

Research map / Video / Video-language understanding

Fine-Grained Text-to-Video Retrieval for Camera-Trap Data

Valentin Gabeff, Baptiste Maquignaz, Jiaxian Shan, Sepideh Mamooler, Gencer Sumbul, Blair Costelloe, Devis TUIA, Alexander Mathis

ECCV 2026 · Video › Video-language understanding · approach: Text-video retrieval

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add custom parsing library to the LLM code agent (ablation)Qwen3-8B agent · Prompting-MammAlps (oracle)F1: 0.44 -> 0.78Improved
add three in-context reasoning examples (ablation)Qwen3-8B agent · Prompting-MammAlps (oracle)F1: 0.78 -> 0.84Improved
add explicit label-space constraints with error feedback (ablation)Qwen3-8B agent · Prompting-MammAlps (oracle)F1: 0.84 -> 0.87Improved
use Qwen3-8B as code agent instead of other 7-8B LLMs (ablation)Agent + Oracle · Prompting-MammAlpsF1: 0.42 -> 0.87Improved
use structured-text LLM agent retrieval instead of zero-shot VLMsAgent + SALMA · Prompting-MammAlps (113 queries)F1: 0.18 -> 0.37Improved
fine-tune CLIP4Clip on MammAlps-S2 query-video pairsCLIP4Clip · Prompting-MammAlps (113 queries)F1: 0.17 -> 0.26Improved
replace SALMA predictions with ground-truth annotations (oracle)Agent · Prompting-MammAlpsF1: 0.34 -> 0.87Improved
remove InfoNCE contrastive loss on object queries (ablation)SALMA-MOT · MammAlps-S2 trackingIDF1: 84.5 -> 75.4Hurt
add second curriculum step (attribute heads) to SALMA-MOTSALMA · MammAlps-S2 trackingHOTA: 72.9 -> 69.0Hurt
use SALMA-MOT instead of MegaDetector+ByteTrackSALMA-MOT · MammAlps-S2 trackingHOTA / IDF1: HOTA 76.2 -> 72.9; IDF1 81.4 -> 84.5Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy