AI Paper Atlas

Research map / Video / Video-language understanding

SpaceVLLM: Endowing Multimodal Large Language Model with Spatio-Temporal Video Grounding Capability

Jiankang Wang, Zhihan Zhang, Zhihang Liu, Yang Li, Jiannan Ge, Hongtao Xie, Yongdong Zhang

AAAI 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 9 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
replace DETR-like decoder with lightweight Query-Guided Space Head (ablation)SpaceVLLM · HCSTVG-v1 testm_vIoU: 35.5 -> 39.5Improved
replace Historical-Assisted Decoder with simple space head (ablation)SpaceVLLM · HCSTVG-v1 testm_vIoU: 38.5 -> 39.5Improved
add historical-frame fusion via gating (ablation)SpaceVLLM · HCSTVG-v1 testm_vIoU: 38.5 vs 39.5No clear effect
add synthetic V-STG data (ablation)SpaceVLLM · HCSTVG-v1 testm_vIoU: 34.4 -> 39.5Improved
add synthetic V-STG data (ablation)SpaceVLLM · HCSTVG-v1 testm_tIoU: 54.3 -> 57.0Improved
decoupled space head with STA queries vs text coordinate generation (LLaVA-Video-SFT)LLaVA-Video-7B · HCSTVG-v1 testm_vIoU: 27.7 -> 39.5Improved
batched space head vs sequential coordinate text generationSpaceVLLM vs LLaVA-ST · VidSTG declarative testavg time per sample (s): 17.63 -> 2.89Improved
add STA queries and STVG trainingLLaVA-Video-SFT · MVBenchaccuracy: 61.6 -> 65.5Improved
add STA queriesLLaVA-Video-SFT · Charades-STAR1@0.7: 38.4 -> 41.9Improved
add STA queries and space headLLaVA-Video-7B · EgoSchemaaccuracy: 57.3 -> 57.4No clear effect
SpaceVLLM vs task-specific CG-STVGSpaceVLLM · HCSTVG-v2 valm_vIoU: 39.5 -> 34.3Hurt

Later papers that compared with this method

2 later accepted papers report SpaceVLLM-7B as a baseline in their results tables; 1 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