GIRL-DETR: Gradient-Isolated Reinforcement Learning for Video Moment Retrieval

• URL: http://arxiv.org/abs/2606.00775v1
• Date: Sat, 30 May 2026 15:40:00 GMT
• Title: GIRL-DETR: Gradient-Isolated Reinforcement Learning for Video Moment Retrieval
• Abstract summary: Video Moment Retrieval (VMR) task requires accurately localizing temporal boundaries aligned with natural language queries.<n>GIRL-DETR introduces RL post-training into a lightweight temporal localization framework for the first time.
• Score: 10.034595422596562
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Video Moment Retrieval (VMR) task requires accurately localizing temporal boundaries aligned with natural language queries, but many models suffer from a misalignment between continuous surrogate losses and non-differentiable metrics, leading to optimization stagnation during the late stages of training and trapping boundary predictions in suboptimal solutions. Although Reinforcement Learning (RL) post-training successfully optimizes localization results for large models, applying it directly to lightweight networks easily disrupts the fragile feature representations established during the supervised phase. To overcome this optimization bottleneck, we propose Gradient-Isolated Reinforcement Learning for DETR (GIRL-DETR), introducing RL post-training into a lightweight temporal localization framework for the first time. The input video and text features first establish early alignment through Cross-Modal Interaction (CMI) before entering the transformer encoder. Subsequently, a Text-Guided Gating (TGG) mechanism dynamically injects semantic priors into the queries before the transformer decoder generates candidate proposals, providing high signal-to-noise ratio inputs for temporal prediction. After the supervised training reaches convergence, the backbone network is frozen to protect the feature manifold, while the detection head directly optimizes the non-differentiable evaluation metric tIoU to enhance localization accuracy through a Three-stage Progressive Reinforcement Learning (TPRL) strategy. This approach achieves an orthogonal decoupling of state representation and metric optimization. Experiments on Charades-STA, QVHighlights, and TACoS demonstrate that GIRL-DETR effectively resolves surrogate loss degradation and achieves substantial accuracy improvements with minimal parameter updates, providing a robust new pathway for RL applications in lightweight VMR models.

Related papers

• LCPNet: Latent Consistent Proximal Unfolding Network for Infrared Small Target Detection [50.3892176822492]
  Infrared small target detection (IRSTD) aims to identify long distance small targets from complex infrared backgrounds.<n>Deep learning methods have improved IRSTD by learning discriminative image-to-mask mappings.<n>Deep unfolding methods partially address this issue by embedding model-driven iterations into neural networks.<n>We propose Latent Consistent Proximal unfolding network (LCPNet) to address these limitations.
  arXiv  Detail & Related papers  (2026-07-06T02:15:09Z)
• Explicit Critic Guidance for Aligning Diffusion Models [27.6704888666835]
  We propose a state-aligned latent actor-critic framework for diffusion post-training.<n>Our method consistently outperforms prior group-relative RL and actor-critic baselines on single-reward and multi-reward benchmarks.
  arXiv  Detail & Related papers  (2026-05-26T22:20:51Z)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• Efficient and Adaptive Human Activity Recognition via LLM Backbones [39.42078885809324]
  We propose a paradigm shift that reuses large pretrained language models (LLMs) as generic temporal backbones for sensor-based HAR.<n>We show that this approach enables rapid convergence, strong data efficiency, and robust cross-dataset transfer.
  arXiv  Detail & Related papers  (2026-05-12T12:06:39Z)
• Power Reinforcement Post-Training of Text-to-Image Models with Super-Linear Advantage Shaping [66.25536973294726]
  Post-training methods for text-to-image (T2I) models are prone to reward hacking.<n>Super-Linear Advantage Shaping (SLAS) reshapes the local policy space.<n>SLAS consistently surpasses the DanceGRPO baseline across multiple backbones and benchmarks.
  arXiv  Detail & Related papers  (2026-05-11T17:59:25Z)
• Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence [30.38509036690193]
  MOCHI is a multi-view 3D face prediction framework trained without requiring registered training data.<n>Standard point-to-surface distances induce training instabilities and visual artifacts in registration-free settings.<n>We propose pointmap- and normal-based losses instead, which provide smoother gradients and superior reconstruction fidelity.
  arXiv  Detail & Related papers  (2026-05-02T13:59:29Z)
• Warm-Started Reinforcement Learning for Iterative 3D/2D Liver Registration [13.711805805211247]
  Learning-based methods have recently achieved registration errors comparable to optimization-based approaches.<n>We present a discrete-action reinforcement learning framework that formulates CT-to-video registration as a sequential decision-making process.
  arXiv  Detail & Related papers  (2026-04-11T14:58:45Z)
• Astrolabe: Steering Forward-Process Reinforcement Learning for Distilled Autoregressive Video Models [58.3184497327891]
  Distilled autoregressive (AR) video models enable efficient streaming generation but frequently misalign with human visual preferences.<n>We present Astrolabe, an efficient online reinforcement learning framework tailored for distilled AR models.
  arXiv  Detail & Related papers  (2026-03-17T18:32:18Z)
• Transformer-based Scalable Beamforming Optimization via Deep Residual Learning [12.79709425087431]
  unsupervised deep learning framework for downlink beamforming in large-scale MU-MISO channels.<n>Model is trained offline, allowing real-time inference through lightweight feedforward computations in dynamic communication environments.
  arXiv  Detail & Related papers  (2025-10-15T01:43:51Z)
• ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving [64.42138266293202]
  ResAD is a Normalized Residual Trajectory Modeling framework.<n>It reframes the learning task to predict the residual deviation from an inertial reference.<n>On the NAVSIM benchmark, ResAD achieves a state-of-the-art PDMS of 88.6 using a vanilla diffusion policy.
  arXiv  Detail & Related papers  (2025-10-09T17:59:36Z)
• Enhancing Robustness of Vision-Language Models through Orthogonality Learning and Self-Regularization [77.62516752323207]
  We introduce an orthogonal fine-tuning method for efficiently fine-tuning pretrained weights and enabling enhanced robustness and generalization. A self-regularization strategy is further exploited to maintain the stability in terms of zero-shot generalization of VLMs, dubbed OrthSR. For the first time, we revisit the CLIP and CoOp with our method to effectively improve the model on few-shot image classficiation scenario.
  arXiv  Detail & Related papers  (2024-07-11T10:35:53Z)
• Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
  We show that a host of variations can be covered in a unified framework that we propose. We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
  arXiv  Detail & Related papers  (2020-06-10T08:22:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.