Research map / Video / Video-language understanding
Factorized Learning for Temporally Grounded Video-Language Models
Wenzheng Zeng, Difei Gao, Mike Zheng Shou, Hwee Tou Ng
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 |
|---|---|---|---|
| decompose grounding and answering into grounding-then-answer objective (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 21.2 -> 28.9 | Improved |
| decompose grounding and answering into grounding-then-answer objective (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg Sim: 11.3 -> 16.0 | Improved |
| use interleaved text-evi token generation in answering (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 28.9 -> 35.6 | Improved |
| use interleaved text-evi token generation in answering (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg F1: 23.1 -> 34.3 | Improved |
| add consistency constraint between the two evi stages (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 35.6 -> 39.5 | Improved |
| add consistency constraint between the two evi stages (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg Sim: 19.8 -> 21.2 | Improved |
| add event-level modeling of evi token (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 26.1 -> 39.5 | Improved |
| add event-level modeling of evi token (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg Sim: 16.2 -> 21.2 | Improved |
| add visual semantic capture into evi token (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg F1: 27.5 -> 35.0 | Improved |
| add visual semantic capture into evi token (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 37.1 -> 39.5 | Improved |
| add factorized preference optimization (FPO) (ablation) | D2VLM · E.T. Bench Grounding | Avg F1: 39.5 -> 42.3 | Improved |
| add factorized preference optimization (FPO) (ablation) | D2VLM · E.T. Bench Dense Captioning | Avg F1: 35.0 -> 37.5 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- TRACE-7B — TRACE: Temporal Grounding Video LLM via Causal Event Modeling (ICLR 2025) · better on 3 of 3 shared results
- VTG-LLM-7B — VTG-LLM: Integrating Timestamp Knowledge into Video LLMs for Enhanced Video Temporal Grounding (AAAI 2025) · better on 1 of 1 shared result
- TimeChat-7B — TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding (CVPR 2024) · better on 1 of 1 shared result
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
- Keeping the Evidence Chain: Semantic Evidence Allocation for Training-Free Token Pruning in Video Temporal Grounding (ECCV 2026 · similarity 0.91)
- Local-Global Video-Text Interactions for Temporal Grounding (CVPR 2020 · similarity 0.91)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.90)
- Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Video Temporal Grounding (NeurIPS 2025 · similarity 0.89)
- Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding (NeurIPS 2025 · similarity 0.89)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.89)
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.
