Behavior Pack Optimization for Video MLLM Post-Training
cs.CV
Submitted: 2026-10-02
Updated: 2026-10-02
Comments: NeurIPS 2026 poster
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Video-R1: Reinforcing Video Reasoning in MLLMs
- Temporal Preference Optimization for Long-Form Video Understanding
- LLaVA-Video: Video Instruction Tuning With Synthetic Data
- CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models
- VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning
- VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning
- DeepVideo-R1: Video Reinforcement Fine-Tuning via Difficulty-aware Regressive GRPO
- SpaceR: Reinforcing MLLMs in Video Spatial Reasoning
- SAM 2: Segment Anything in Images and Videos
- TransNet V2: An effective deep network architecture for fast shot transition detection
- MixDPO: Modeling Preference Strength for Pluralistic Alignment
- Proximal Policy Optimization Algorithms
- SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models