GRASP is a large-scale dataset and benchmark for social reasoning grounded in gaze and gesture events in multi-person videos, with Social Grounding Reward (SGR) proposed to improve model performance on GRASP-Bench.
hub Baseline reference
Scaling rl to long videos
Baseline reference. 50% of citing Pith papers use this work as a benchmark or comparison.
20
Pith papers citing it
Baseline
50% of classified citations
hub tools
citation-role summary
background 3
baseline 2
dataset 1
citation-polarity summary
representative citing papers
LMMs perceive videos but underexploit visual content for causal reasoning due to textual shortcuts; ProCauEval diagnoses this and ADPO training reduces reliance on priors.
TriAttention compresses KV cache by exploiting stable pre-RoPE Q/K concentration and trigonometric distance preferences to match full-attention reasoning accuracy with far lower memory and higher speed.
RLVR exhibits correct-set turnover where solved problems regress during training, and a periodic review mechanism exploiting a repair-window principle improves retention and performance over baselines.
An agentic three-stage video annotation pipeline with an MSTED intermediate and top-down domain adaptation produces CoT training data that lifts Cosmos-Reason2 past Gemini on traffic event reasoning.
ParaVT introduces the first multi-agent RL framework for parallel video tool calling in LMMs, using PARA-GRPO to resolve the Tool Prior Paradox and achieve +7.9% average improvement over Qwen3-VL baseline across six benchmarks.
VideoSeeker integrates agentic reasoning and visual prompts into LVLMs via automated data synthesis, cold-start supervision, and RL training, yielding +13.7% gains on instance-level video tasks over baselines including GPT-4o.
Video-Zero is an annotation-free Questioner-Solver co-evolution framework that centers self-evolution on temporally localized evidence to improve video VLMs.
Filtering post-training data to visually grounded questions improves VLM video understanding performance by up to 6.2 points using 69% of the data.
GraphThinker reduces temporal hallucinations in video reasoning by constructing event-based scene graphs and applying visual attention rewards in reinforcement finetuning.
Skyra is an MLLM that detects AI-generated videos by identifying and reasoning over grounded visual artifacts, supported by a new annotated dataset and benchmark.
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
Activation Replay boosts multimodal reasoning in post-trained LMMs by replaying low-entropy activations from base models to RLVR counterparts at test time via visual token manipulation.
REVISOR adds multimodal visual-text reflection and a Dual Attribution Decoupled Reward to improve long-form video reasoning in MLLMs without extra supervised fine-tuning.
TempR1 applies temporal-aware multi-task RL using GRPO and three types of localization rewards to achieve SOTA temporal understanding in MLLMs with synergistic gains from joint optimization.
OneThinker unifies image and video reasoning in one model across 10 tasks via a 600k corpus, CoT-annotated SFT, and EMA-GRPO reinforcement learning, reporting strong results on 31 benchmarks plus some cross-task transfer.
This is a survey that frames video MLLM research via a human-view formulation of perceptual representations, memory states, reasoning traces, and predictions, then reviews methods, datasets, benchmarks, and open problems.
EasyVideoR1 delivers an optimized RL pipeline for video understanding in large vision-language models, achieving 1.47x throughput gains and aligned results on 22 benchmarks.
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
citing papers explorer
-
GRASP: Learning to Ground Social Reasoning in Multi-Person Non-Verbal Interactions
GRASP is a large-scale dataset and benchmark for social reasoning grounded in gaze and gesture events in multi-person videos, with Social Grounding Reward (SGR) proposed to improve model performance on GRASP-Bench.
-
Perception Without Engagement: Dissecting the Causal Discovery Deficit in LMMs
LMMs perceive videos but underexploit visual content for causal reasoning due to textual shortcuts; ProCauEval diagnoses this and ADPO training reduces reliance on priors.
-
TriAttention: Efficient Long Reasoning with Trigonometric KV Compression
TriAttention compresses KV cache by exploiting stable pre-RoPE Q/K concentration and trigonometric distance preferences to match full-attention reasoning accuracy with far lower memory and higher speed.
-
Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR
RLVR exhibits correct-set turnover where solved problems regress during training, and a periodic review mechanism exploiting a repair-window principle improves retention and performance over baselines.
-
MAVEN: A Multi-stage Agentic Annotation Pipeline for Video Reasoning Tasks
An agentic three-stage video annotation pipeline with an MSTED intermediate and top-down domain adaptation produces CoT training data that lifts Cosmos-Reason2 past Gemini on traffic event reasoning.
-
ParaVT: Taming the Tool Prior Paradox for Parallel Tool Use in Agentic Video Reinforcement Learning
ParaVT introduces the first multi-agent RL framework for parallel video tool calling in LMMs, using PARA-GRPO to resolve the Tool Prior Paradox and achieve +7.9% average improvement over Qwen3-VL baseline across six benchmarks.
-
VideoSeeker: Incentivizing Instance-level Video Understanding via Native Agentic Tool Invocation
VideoSeeker integrates agentic reasoning and visual prompts into LVLMs via automated data synthesis, cold-start supervision, and RL training, yielding +13.7% gains on instance-level video tasks over baselines including GPT-4o.
-
Video-Zero: Self-Evolution Video Understanding
Video-Zero is an annotation-free Questioner-Solver co-evolution framework that centers self-evolution on temporally localized evidence to improve video VLMs.
-
Watch Before You Answer: Learning from Visually Grounded Post-Training
Filtering post-training data to visually grounded questions improves VLM video understanding performance by up to 6.2 points using 69% of the data.
-
GraphThinker: Reinforcing Temporally Grounded Video Reasoning with Event Graph Thinking
GraphThinker reduces temporal hallucinations in video reasoning by constructing event-based scene graphs and applying visual attention rewards in reinforcement finetuning.
-
Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning
Skyra is an MLLM that detects AI-generated videos by identifying and reasoning over grounded visual artifacts, supported by a new annotated dataset and benchmark.
-
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
-
Boosting Reasoning in Large Multimodal Models via Activation Replay
Activation Replay boosts multimodal reasoning in post-trained LMMs by replaying low-entropy activations from base models to RLVR counterparts at test time via visual token manipulation.
-
REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding
REVISOR adds multimodal visual-text reflection and a Dual Attribution Decoupled Reward to improve long-form video reasoning in MLLMs without extra supervised fine-tuning.
-
TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning
TempR1 applies temporal-aware multi-task RL using GRPO and three types of localization rewards to achieve SOTA temporal understanding in MLLMs with synergistic gains from joint optimization.
-
OneThinker: All-in-one Reasoning Model for Image and Video
OneThinker unifies image and video reasoning in one model across 10 tasks via a 600k corpus, CoT-annotated SFT, and EMA-GRPO reinforcement learning, reporting strong results on 31 benchmarks plus some cross-task transfer.
-
Watch, Remember, Reason: Human-View Video Understanding with MLLMs
This is a survey that frames video MLLM research via a human-view formulation of perceptual representations, memory states, reasoning traces, and predictions, then reviews methods, datasets, benchmarks, and open problems.
-
EasyVideoR1: Easier RL for Video Understanding
EasyVideoR1 delivers an optimized RL pipeline for video understanding in large vision-language models, achieving 1.47x throughput gains and aligned results on 22 benchmarks.
-
A Survey of Reinforcement Learning for Large Reasoning Models
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
- VISD: Enhancing Video Reasoning via Structured Self-Distillation