Future-L1 interleaves latent visual spans with text in MLLM decoding, trained on a custom Future-L1-50K dataset via LA-DAPO RL, and reports SOTA gains on FutureBench (61.0 to 85.4) and TwiFF-Bench (2.44 to 3.04).
Videoauto-r1: Video auto reasoning via thinking once, answering twice
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
years
2026 8representative citing papers
AdaFocus achieves better accuracy on long-video benchmarks with roughly 33 times fewer visual tokens by combining query-aware adaptive sampling and zero-cache disk-based refinement.
Visual debiasing of omni-modal benchmarks combined with staged post-training lets a 3B model match or exceed a 30B model without a stronger teacher.
MER-R1 uses dual-objective RL to optimize fast-thinking recall and slow-thinking precision separately in multimodal emotion recognition, with calibration to align them, yielding SOTA results on two benchmarks.
CineCap combines structured reasoning and RL rewards to outperform baselines on cinematographic video captioning using a new 472-pair benchmark.
CRPO applies counterfactual videos and a cross-branch relation reward in RL post-training to reduce shortcut reliance in Video LLMs, with gains shown on the new DyBench paired benchmark.
STRIVE stabilizes RL for video QA by creating spatiotemporal video variants and using importance-aware sampling, yielding consistent gains over baselines on six benchmarks.
MindClaw adds belief memory and a skill-guided cognitive trigger so an embodied agent intervenes only when false beliefs or hidden goals block progress, beating direct VLMs on MindPower intervention metrics.
citing papers explorer
-
Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction
Future-L1 interleaves latent visual spans with text in MLLM decoding, trained on a custom Future-L1-50K dataset via LA-DAPO RL, and reports SOTA gains on FutureBench (61.0 to 85.4) and TwiFF-Bench (2.44 to 3.04).
-
AdaFocus: Adaptive Relevance-Diversity Sampling with Zero-Cache Look-back for Efficient Long Video Understanding
AdaFocus achieves better accuracy on long-video benchmarks with roughly 33 times fewer visual tokens by combining query-aware adaptive sampling and zero-cache disk-based refinement.
-
Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation
Visual debiasing of omni-modal benchmarks combined with staged post-training lets a 3B model match or exceed a 30B model without a stronger teacher.
-
MER-R1: Multimodal Emotion Reasoning via Slow-Fast Thinking Synergy
MER-R1 uses dual-objective RL to optimize fast-thinking recall and slow-thinking precision separately in multimodal emotion recognition, with calibration to align them, yielding SOTA results on two benchmarks.
-
CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning
CineCap combines structured reasoning and RL rewards to outperform baselines on cinematographic video captioning using a new 472-pair benchmark.
-
Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning
CRPO applies counterfactual videos and a cross-branch relation reward in RL post-training to reduce shortcut reliance in Video LLMs, with gains shown on the new DyBench paired benchmark.
-
STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering
STRIVE stabilizes RL for video QA by creating spatiotemporal video variants and using importance-aware sampling, yielding consistent gains over baselines on six benchmarks.
-
MindClaw: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
MindClaw adds belief memory and a skill-guided cognitive trigger so an embodied agent intervenes only when false beliefs or hidden goals block progress, beating direct VLMs on MindPower intervention metrics.