AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

The All-Seeing Project V2: Towards General Relation Comprehension of the Open World

Weiyun Wang, yiming ren, Haowen Luo, Tiantong Li, Chenxiang Yan, Zhe Chen, Wenhai Wang, Qingyun Li, Lewei Lu, Xizhou Zhu, Yu Qiao, Jifeng Dai

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Benchmarks and analysis of MLLMs · 66 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 relation conversation (AS-V2) data to training (ablation)ASMv2 · CRPEoverall accuracy: 55.3 -> 64.5Improved
add relation conversation data to training (ablation)ASMv2 · MMEscore: 1554.6 -> 1621.0Improved
add relation conversation data to training (ablation)ASMv2 · MM-Vetscore: 42.2 -> 41.3Mixed
add relation conversation data to training (ablation)ASMv2 · VCRQ->AR accuracy: 77.0 -> 78.4Improved
add two-stage training (image-level stage 1 first) (ablation)ASMv2 · MMEscore: 1527.6 -> 1621.0Improved
add two-stage training (image-level stage 1 first) (ablation)ASMv2 · MM-Vetscore: 35.6 -> 41.3Improved
add two-stage training (image-level stage 1 first) (ablation)ASMv2 · CRPEoverall accuracy: 64.7 -> 64.5No clear effect
add two-stage training (ablation)ASMv2 · RefCOCO/+/g averageaccuracy: 86.1 -> 87.4Improved
include caption annotations in the GPT-4V prompt for data generationAS-V2 data construction · data qualityhallucination in generated dataImproved
single-task fine-tune on VCRASMv2 · VCR validationQ->AR accuracy: 78.4 -> 79.4Improved
train on relation conversation to predict specific predicatesASMv2 · PSG Predicate ClassificationmR@100 versus R@100: mR@100 44.5 vs VCTree 23.8; R@100 32.6 vs VCTree 53.1Mixed
open-ended generation of scene graphs with ASMv2 instead of TextPSGOpen-ended SGG · PSGRecall: 5.5 -> 14.2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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