AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

LCEval: Learned Composite Metric for Caption Evaluation

Naeha Sharif, Lyndon White, Mohammed Bennamoun, Wei Liu, Syed Afaq Ali Shah

IJCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 8 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 learning (one-layer NN) over equal-weight feature combination (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.273 -> 0.375Improved
use two-layer NN instead of one-layer (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.375 -> 0.418Improved
add a second hidden layer (three-layer NN) (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.418 -> 0.417No clear effect
learn SPIDEr combination with network (ablation)SPIDEr · COMPOSITE (caption-level)Kendall tau: 0.357 -> 0.367Improved
use FastText instead of GloVe for MOWE (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.414 -> 0.418Improved
add lexical features (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.366 -> 0.418Improved
add semantic features (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.387 -> 0.418Improved
add syntactic features (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.413 -> 0.418Improved
train on all three captioning models instead of Adaptive Attention only (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.403 -> 0.418Improved
train on 132,984 examples instead of 88,656 from Show and Tell only (ablation)LCEval · COMPOSITE (caption-level)Kendall tau: 0.421 -> 0.418No clear effect
add more machine caption variety to training (ablation)LCEval · PASCAL50S HHCaccuracy: 58.7 -> 64.0Improved
increase number of reference captions from 1 to 12 (ablation)LCEval · PASCAL50Saccuracy: 0.740 -> 0.804Improved

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