AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models

Sivan Doveh, Assaf Arbelle, Sivan Harary, Roei Herzig, Donghyun Kim, Paola Cascante-Bonilla, Amit Alfassy, Rameswar Panda, Raja Giryes, Rogerio Feris, Shimon Ullman, Leonid Karlinsky

NeurIPS 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 12 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
Fine-tune CLIP with quality-enhanced captions, LLM-expanded caption bags, negative text augmentation, and negative-enhanced MILOpenAI CLIP ViT-B/32 · VL-Checklist + AROAvg: 79.42 -> 85.18Improved
Fine-tune CLIP with quality-enhanced captions, SAM segment captions, negative text augmentation, and negative-enhanced MILOpenAI CLIP ViT-B/32 · VL-Checklist RelationRelation accuracy: 63.05 -> 89.75Improved
Add caption quality enhancement (ablation)CLIP fine-tuning with caption expansion · VL-Checklist + AROAvg: 61.16% -> 74.38%Improved
Add LLM caption density expansion to quality-enhanced captions with negative augmentation (ablation)CLIP fine-tuning without MIL · VL-Checklist + AROAvg: 60.20% -> 78.09%Improved
Add negative-enhanced MIL instead of randomly choosing one caption from the bag (ablation)Quality-enhanced CLIP fine-tuning with negative augmentation and LLM expansion · VL-Checklist RelationRelation accuracy: 81.97% -> 86.41%Improved
Add negative examples to the base MIL loss (ablation)LLM-expanded CLIP fine-tuning · VL-Checklist RelationRelation accuracy: 69.05% -> 86.41%Improved
Use MIL aggregation rather than MAX selection from the caption bag (ablation)LLM-expanded CLIP fine-tuning · VL-Checklist RelationRelation accuracy: 60.35% -> 69.05%Improved
Use MIL aggregation rather than averaging caption embeddings (ablation)LLM-expanded CLIP fine-tuning · VL-Checklist RelationRelation accuracy: 65.57% -> 69.05%Improved
Replace negative-enhanced MIL with MAX selection (ablation)LLM-expanded CLIP fine-tuning · VL-Checklist RelationRelation accuracy: 86.41% -> 60.35%Hurt
Replace negative-enhanced MIL with AVG selection (ablation)LLM-expanded CLIP fine-tuning · VL-Checklist RelationRelation accuracy: 86.41% -> 65.57%Hurt
Use LLM-based rather than SAM-based caption density expansion (ablation)DAC fine-tuning · VL-Checklist + AROAvg: 85.24% vs 83.83%Mixed
Use SAM-based rather than LLM-based caption density expansion (ablation)DAC fine-tuning · VL-Checklist + AROAvg: 83.83% vs 85.24%Mixed

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