AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

New Dataset and Methods for Fine-Grained Compositional Referring Expression Comprehension via Specialist-MLLM Collaboration

Xuzheng Yang, Junzhuo Liu, Peng Wang, Guoqing Wang, Yang Yang, Heng Tao Shen

TPAMI 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Benchmarks and analysis of MLLMs · 4 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 Slow-Fast Adaptation routing between specialist and MLLMCogVLM · FineCops-RefPrecision@1 (All): 69.46 -> 73.57Improved
add focus-enhancement strategy to SFACogVLM(SFA) · FineCops-RefPrecision@1 (All): 73.57 -> 75.03Improved
add SFA routingQwen2-VL-7B · Ref-AdvPrecision@1 (All): 79.83 -> 79.62No clear effect
add Candidate Region Selection (CRS) tuningInternVL2.5-8B · FineCops-RefPrecision@1 (All): 66.05 -> 78.58Improved
add CRS tuningQwen2-VL-7B · Ref-ReasoningPrecision@1 (All): 35.77 -> 40.45Improved
add focus-enhancement strategyMDETR · Ref-AdvPrecision@1: 73.73 -> 72.12Hurt
add focus-enhancement strategyInternVL2.5-8B · Ref-AdvPrecision@1: 75.92 -> 80.29Improved
add focus-enhancement strategyMM-GDINO-L · Ref-AdvPrecision@1: 74.95 -> 74.55Hurt
use Grounding DINO instead of YOLO-World as difficulty router (ablation)SFA · FineCops-RefPrecision@1 (All): 71.53 -> 74.06Improved
use Grounding DINO instead of OV-DINO as difficulty router (ablation)SFA · FineCops-RefPrecision@1 (All): 70.39 -> 74.06Improved
train CRS on FineCops-Ref data instead of RefCOCO/+/g (ablation)CRS Qwen2-VL-7B · FineCops-RefPrecision@1 (All): 78.33 -> 81.38Improved
increase CRS instruction tuning data from 2k to 12k (ablation)CRS Qwen2-VL-7B · FineCops-RefPrecision@1 (All): 74.95 -> 78.84Improved

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