# VideoTG-R1: Boosting Video Temporal Grounding via Curriculum Reinforcement Learning on Reflected Boundary Annotations

arXiv: https://arxiv.org/abs/2510.23397 · HTML: https://mlsanity.com/paper/2510.23397/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 47 of its 82 references that are in the map.
- First posted on arXiv 27 Oct 2025.

## Placed in

- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 30 of 47)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [VideoChat: Chat-Centric Video Understanding](https://mlsanity.com/paper/2305.06355/) (2023)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [Localizing Moments in Video with Natural Language](https://mlsanity.com/paper/1708.01641/) (2017)
- [DataComp: In search of the next generation of multimodal datasets](https://mlsanity.com/paper/2304.14108/) (2023)
- [TALL: Temporal Activity Localization via Language Query](https://mlsanity.com/paper/1705.02101/) (2017)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding](https://mlsanity.com/paper/2312.02051/) (2023)
- [Learning 2D Temporal Adjacent Networks for Moment Localization with Natural Language](https://mlsanity.com/paper/1912.03590/) (2019)
- [VTimeLLM: Empower LLM to Grasp Video Moments](https://mlsanity.com/paper/2311.18445/) (2023)
- [VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2504.06958/) (2025)
- [Zero-Shot Video Question Answering via Frozen Bidirectional Language Models](https://mlsanity.com/paper/2206.08155/) (2022)
- [Self-Chained Image-Language Model for Video Localization and Question Answering](https://mlsanity.com/paper/2305.06988/) (2023)
- [UniVTG: Towards Unified Video-Language Temporal Grounding](https://mlsanity.com/paper/2307.16715/) (2023)
- [Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding](https://mlsanity.com/paper/2503.13377/) (2025)
- [SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement](https://mlsanity.com/paper/2504.07934/) (2025)
- [Can I Trust Your Answer? Visually Grounded Video Question Answering](https://mlsanity.com/paper/2309.01327/) (2023)
- [LITA: Language Instructed Temporal-Localization Assistant](https://mlsanity.com/paper/2403.19046/) (2024)
- [Momentor: Advancing Video Large Language Model with Fine-Grained Temporal Reasoning](https://mlsanity.com/paper/2402.11435/) (2024)
- [TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning](https://mlsanity.com/paper/2410.19702/) (2024)
- [EgoLife: Towards Egocentric Life Assistant](https://mlsanity.com/paper/2503.03803/) (2025)
- [Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding](https://mlsanity.com/paper/2109.04872/) (2021)
- [TRACE: Temporal Grounding Video LLM via Causal Event Modeling](https://mlsanity.com/paper/2410.05643/) (2024)
- [HawkEye: Training Video-Text LLMs for Grounding Text in Videos](https://mlsanity.com/paper/2403.10228/) (2024)
- [Hierarchical Video-Moment Retrieval and Step-Captioning](https://mlsanity.com/paper/2303.16406/) (2023)
- [Knowing Where to Focus: Event-aware Transformer for Video Grounding](https://mlsanity.com/paper/2308.06947/) (2023)

## Built on by (most cited first, 7 of 7)

- [TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs](https://mlsanity.com/paper/2607.17423/) (2026)
- [Towards Fine-grained Temporal Perception: Post-Training Large Audio-Language Models with Audio-Side Time Prompt](https://mlsanity.com/paper/2604.13715/) (2026)
- [CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoning](https://mlsanity.com/paper/2606.08436/) (2026)
- [SER: Learning to Ground Video Reasoning with Semantic Evidence Rewards](https://mlsanity.com/paper/2606.24726/) (2026)
- [Parallelized Autoregressive Decoding for Omni-Modal Dense Video Captioning](https://mlsanity.com/paper/2607.02963/) (2026)
- [TimePLE: Rethinking Temporal Representation for Video Temporal Grounding](https://mlsanity.com/paper/2607.23951/) (2026)
- [Locate Anything in Videos: Rethinking Efficient Generative Spatio-Temporal Video Grounding](https://mlsanity.com/paper/2608.28192/) (2026)

## Abstract (arXiv)

Video temporal grounding (VTG) aims to locate precise segments in videos based on language queries, which is a fundamental challenge in video understanding. While recent Multimodal Large Language Models (MLLMs) have shown promise in tackling VTG through reinforcement learning (RL), they overlook the challenges arising from both the quality and difficulty of training samples. (1) Partially annotated samples. Many samples contain relevant segments beyond the annotated interval, introducing ambiguous supervision. (2) Hard-to-ground samples. Samples with poor zero-shot performance produce consistently low and indistinguishable rewards during RL training, exhibiting no clear preference among multiple outputs and thus hindering learning efficiency. To address these challenges, we propose VideoTG-R1, a novel curriculum RL framework with reflected boundary annotations, enabling data-efficient training. Specifically, we propose a Boundary Reflection Agent that utilizes MLLMs to predict query-relevant timestamps outside the annotated intervals, allowing us to identify and filter out partially annotated samples, thereby reducing ambiguity. Furthermore, we introduce a Difficulty Estimation Agent to assess the training difficulty of each sample and design a curriculum RL strategy that dynamically masks the videos of hard-to-ground samples according to the training steps, easing the training difficulty and providing clearer preference. Experiments on the VTG and grounded VideoQA tasks demonstrate the effectiveness of our method. Remarkably, with only 10% of the training samples and 21% of the computational budget, VideoTG-R1 outperforms full-data counterparts under both group relative policy optimization (GRPO) and supervised fine-tuning (SFT). The code is available at https://github.com/ldong1111/VideoTG-R1.