Research map / Vision-language and multimodal / Multimodal LLMs
TWIST & SCOUT: Grounding Multimodal LLM-Experts by Forget-Free Tuning
Aritra Bhowmik, Mohammad Mahdi Derakhshani, Dennis Koelma, Yuki M. Asano, Martin R. Oswald, Cees G. M. Snoek
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add TWIST twin-expert module (ablation) | TWIST & SCOUT · MM-Vet | score: 0 -> 34.3 | Improved |
| add TWIST twin-expert module (ablation) | TWIST & SCOUT · RefCOCO | accuracy: 84.8 -> 85.6 | Improved |
| add TWIST twin-expert module (ablation) | TWIST & SCOUT · COCO grounded captioning | AP: 9.6 -> 11.4 | Improved |
| add SCOUT dataset (ablation) | TWIST & SCOUT · RefCOCO | accuracy: 85.6 -> 86.9 | Improved |
| add SCOUT dataset (ablation) | TWIST & SCOUT · COCO grounded captioning | AP: 11.4 -> 13.7 | Improved |
| add step-wise loss (ablation) | TWIST & SCOUT · COCO grounded captioning | AP: 13.7 -> 15.0 | Improved |
| add learned alpha gating (alpha=0.31 instead of 0) (ablation) | TWIST & SCOUT · COCO grounded captioning | mAP: 0 -> 15 | Improved |
| use 2 grounding experts instead of 1 (ablation) | TWIST & SCOUT · RefCOCO test-A | accuracy: 79.8 -> 90.2 | Improved |
| use 4 grounding experts instead of 2 (ablation) | TWIST & SCOUT · RefCOCO test-A | accuracy: 90.2 -> 90.3 | No clear effect |
| use MoE-LLaVA backbone instead of LLaVA (ablation) | TWIST & SCOUT · RefCOCO test-A | accuracy: 85.7 -> 90.2 | Improved |
| replace Visual Genome with SCOUT as fine-tuning data (ablation) | TWIST & SCOUT · COCO grounded captioning | AP50: 38.5 -> 49.3 | Improved |
| add negative samples to SCOUT (ablation) | TWIST & SCOUT · COCO grounded captioning | AP50: 48.1 -> 49.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Ferret-7B — Ferret: Refer and Ground Anything Anywhere at Any Granularity (ICLR 2024) · better on 1 of 2 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
- GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models (AAAI 2026 · similarity 0.91)
- Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning (AAAI 2026 · similarity 0.91)
- Grounded Chain-of-Thought for Multimodal Large Language Models (CVPR 2026 · similarity 0.90)
- GroundingME: Exposing the Visual Grounding Gap in MLLMs through Multi-Dimensional Evaluation (CVPR 2026 · similarity 0.90)
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.90)
- F-LMM: Grounding Frozen Large Multimodal Models (CVPR 2025 · similarity 0.90)
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.
