AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Merlin: Empowering Multimodal LLMs with Foresight Minds

En Yu, Liang Zhao, YANA WEI, Jinrong Yang, Dongming Wu, Lingyu Kong, Haoran Wei, Tiancai Wang, Zheng Ge, Xiangyu Zhang, Wenbing Tao

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 36 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 Foresight Pre-Training data (ablation)Merlin · MMBench prediction reasoning (Dev)average score: 60.7 -> 64.4Improved
add Foresight Instruction-Tuning data (ablation)Merlin · MMBench prediction reasoning (Dev)average score: 61.2 -> 64.4Improved
include image-text pair data in pretraining (ablation)Merlin · MMBench prediction reasoning (Dev)average score: 52.8 -> 64.4Improved
add Trajectory Chain-of-Thought in FIT (ablation)Merlin · MMBench prediction reasoning (Dev)Prediction Reasoning: 63.0 -> 64.4Improved
use 448 instead of 336 input resolution (ablation)Merlin · Prediction Rea. / GOT10kPrediction Reasoning: 59.8 -> 64.4Improved
use Conv2d projector instead of MLP (fewer tokens) (ablation)Merlin · Prediction Rea. / GOT10kPrediction Reasoning: 58.1 -> 64.4Improved
unfreeze visual encoder during pre-training (ablation)Merlin · Prediction Rea. / GOT10kGOT10k AO: 28.4 -> 51.4Improved
add precise task description to prompts (ablation)Merlin · GOT10kAO: 28.4 -> 51.4Improved
add foresight learning (FPT and FIT)LLaVA-1.5-style MLLM · POPE COCO Randomaccuracy: 83.29 -> 91.58Improved
foresight learning versus LLaVA-1.5 on GQAMerlin · GQAaccuracy: 62.0 -> 60.5Hurt

Later papers that compared with this method

4 later accepted papers report Merlin as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.

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