AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Unlocking Few-Shot Capabilities in LVLMs via Prompt Conditioning and Head Selection

Adhémar de Senneville, Xavier Bou, Jérémy Anger, Rafael Grompone von Gioi, Gabriele Facciolo

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 1 citations (OpenAlex)

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 Task prompt conditioning (ablation)HEC-T · 10-way 4-shot zero-shotaccuracy: 82.34 -> 84.09Improved
add Domain prompt conditioning (ablation)HEC-T · 10-way 4-shot zero-shotaccuracy: 82.34 -> 87.08Improved
add Class prompt conditioning (ablation)HEC-T · 10-way 4-shot zero-shotaccuracy: 82.34 -> 88.45Improved
add Domain prompt conditioning (ablation)HEC-V · 10-way 4-shot few-shotaccuracy: 90.43 -> 92.78Improved
add Class prompt conditioning (ablation)HEC-V · 10-way 4-shot few-shotaccuracy: 90.43 -> 94.14Improved
use Domain prompt instead of Task prompt for probingQwen2-VL probing · 12 datasets 4-shotaccuracy: 75.6 -> 79.9Improved
use text-heads (HEC-T) instead of summary tokenQwen2-VL · 12 datasets 10-way 0-shotaccuracy: 81.1 -> 86.2Improved
use HEC-T on LLaVA-OV instead of baselineLLaVA-OV · 12 datasets 10-way 0-shotaccuracy: 79.7 -> 82.1Improved
add text-heads to vision-heads (HEC-VT over HEC-V)HEC · 12 datasets 4-shot vision-textaccuracy: 82.4 -> 83.0Improved
average probabilities instead of majority vote across heads (ablation)HEC-V · 10-way 4-shot ensembleaccuracy: 93.85 -> 94.00No clear effect
learn optimal weights instead of mean for head ensemble (ablation)HEC-V · 10-way 4-shot ensembleaccuracy: 94.00 -> 94.04No clear effect
select domain heads instead of task heads (ablation)HEC-V · 10-way 4-shot head selectionaccuracy: 93.79 -> 94.18Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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