AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

BRIDGE: Bridging Gaps in Image Captioning Evaluation with Stronger Visual Cues

Sara Sarto, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 7 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
replace the transformer mapping module with a two-layer MLP (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 53.1 -> 54.4Improved
add template captions as mapping module input (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 53.7 -> 54.4Improved
add the regularization branch (ablation)BRIDGE · Flickr8k-CFKendall tau_b: 35.7 -> 36.1Improved
add textual similarity with the pseudo-caption to CLIP-Score (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 51.1 -> 54.4Improved
add visual similarity term to the score (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 53.8 -> 54.4Improved
use 3 pseudo tokens per noun chunk instead of 1, 2, 4 or 8 (ablation)BRIDGE · Flickr8k-Expert / CF / Pascal-50SKendall tau_b / accuracy: Expert 54.1 -> 54.4 (US=1 vs 3)Mixed
add the mapping module on top of BLIP template captions (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 48.6 -> 54.4Improved
use higher-quality template captions (BLIP-2 instead of Transformer captioner) (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 54.0 -> 54.4No clear effect
weight contrastive loss by number of noun chunks (ablation)BRIDGE · Flickr8k-CFKendall tau_b: 35.6 -> 36.1Improved
use grid-level visual features instead of global features (ablation)BRIDGE · Flickr8k-CFKendall tau_b: 35.1 -> 36.1Improved
keep only mapping outputs at masked positions instead of full output (ablation)BRIDGE · Flickr8k-ExpertKendall tau_b: 52.0 -> 54.4Improved
replace CLIP-Score with BRIDGE for hallucination detectionCLIP ViT-B/32 metric · FOILaccuracy: 87.2 -> 91.5Improved

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