ViMU is the first benchmark for evaluating video models on metaphorical and subtextual understanding using hint-free questions grounded in multimodal evidence.
hub Baseline reference
Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal llms in video analysis
Baseline reference. 60% of citing Pith papers use this work as a benchmark or comparison.
19
Pith papers citing it
Baseline
60% of classified citations
hub tools
citation-role summary
dataset 5
background 4
baseline 1
citation-polarity summary
years
2026 19representative citing papers
TraceAV-Bench is the first benchmark for multi-hop trajectory reasoning over long audio-visual videos, showing top models reach only 51-68% accuracy with substantial room for improvement.
DriveSpatial benchmark shows the best of 15 VLMs trails humans by 28.4 points on spatiotemporal driving tasks, with cognitive scene construction as the main failure mode.
VGenST-Bench is a new video benchmark for MLLM spatio-temporal reasoning built via generative synthesis, a multi-agent pipeline with human oversight, a 3x2x2 taxonomy, and hierarchical tasks separating perception from reasoning.
Introduces the Grounded Personality Reasoning task and MM-OCEAN dataset to show that MLLMs frequently produce correct Big Five personality ratings without grounding them in observable video evidence.
Omni-DuplexEval creates a new benchmark and LLM-as-a-Judge framework for real-time duplex omni-modal interaction, revealing that current models score below 40% overall and struggle especially with proactive responses.
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.
ReTool-Video uses a 134-tool meta-augmented library and recursive grounding to translate abstract video intents into fine-grained multimodal operations, outperforming baselines on MVBench, MLVU, and Video-MME.
TOC-Bench is a new diagnostic benchmark that reveals major weaknesses in temporal object consistency for Video-LLMs, including event counting, ordering, identity reasoning, and hallucination avoidance.
SYNCR benchmark shows leading MLLMs reach only 52.5% average accuracy on cross-video reasoning tasks against an 89.5% human baseline, with major weaknesses in physical and spatial reasoning.
Temporal information in Video-LLMs is encoded well by video-centric encoders but disrupted by standard projectors; time-preserved MLPs plus AoT supervision yield 98.1% accuracy on arrow-of-time and gains on other temporal tasks.
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.
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.
Continued pre-training with balanced long-document VQA data extends a 7B LVLM to 128K context, improving long-document VQA by 7.1% and generalizing to 512K without further training.
OTT-Vid uses optimal transport with non-uniform token mass and locality-aware costs to dynamically allocate compression budgets across video frames, retaining 95.8% VQA and 73.9% VTG performance at 10% token retention.
LDDR proposes a linear DPP-based dynamic-resolution frame sampler that achieves 3x speedup and up to 2.5-point gains on video MLLM benchmarks by selecting non-redundant frames and allocating tokens accordingly.
Video-MME-v2 is a new benchmark that applies progressive visual-to-reasoning levels and non-linear group scoring to expose gaps in video MLLM capabilities.
OmniSelect is a training-free, modality-adaptive token pruning framework that dynamically selects Audio-Centric, Video-Centric, or Uniform compression regimes using AudioCLIP cross-modal relevance scores and then applies adaptive fine-grained pruning within temporal groups.
OmniRefine introduces alignment-aware chunk refinement via similarity and dynamic programming followed by modality-cooperative token compression, achieving near-baseline accuracy at 44% token retention on WorldSense.
citing papers explorer
-
ViMU: Benchmarking Video Metaphorical Understanding
ViMU is the first benchmark for evaluating video models on metaphorical and subtextual understanding using hint-free questions grounded in multimodal evidence.
-
TraceAV-Bench: Benchmarking Multi-Hop Trajectory Reasoning over Long Audio-Visual Videos
TraceAV-Bench is the first benchmark for multi-hop trajectory reasoning over long audio-visual videos, showing top models reach only 51-68% accuracy with substantial room for improvement.
-
DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving
DriveSpatial benchmark shows the best of 15 VLMs trails humans by 28.4 points on spatiotemporal driving tasks, with cognitive scene construction as the main failure mode.
-
VGenST-Bench: A Benchmark for Spatio-Temporal Reasoning via Active Video Synthesis
VGenST-Bench is a new video benchmark for MLLM spatio-temporal reasoning built via generative synthesis, a multi-agent pipeline with human oversight, a 3x2x2 taxonomy, and hierarchical tasks separating perception from reasoning.
-
Perception or Prejudice: Can MLLMs Go Beyond First Impressions of Personality?
Introduces the Grounded Personality Reasoning task and MM-OCEAN dataset to show that MLLMs frequently produce correct Big Five personality ratings without grounding them in observable video evidence.
-
Omni-DuplexEval: Evaluating Real-time Duplex Omni-modal Interaction
Omni-DuplexEval creates a new benchmark and LLM-as-a-Judge framework for real-time duplex omni-modal interaction, revealing that current models score below 40% overall and struggle especially with proactive responses.
-
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.
-
ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding
ReTool-Video uses a 134-tool meta-augmented library and recursive grounding to translate abstract video intents into fine-grained multimodal operations, outperforming baselines on MVBench, MLVU, and Video-MME.
-
TOC-Bench: A Temporal Object Consistency Benchmark for Video Large Language Models
TOC-Bench is a new diagnostic benchmark that reveals major weaknesses in temporal object consistency for Video-LLMs, including event counting, ordering, identity reasoning, and hallucination avoidance.
-
SYNCR: A Cross-Video Reasoning Benchmark with Synthetic Grounding
SYNCR benchmark shows leading MLLMs reach only 52.5% average accuracy on cross-video reasoning tasks against an 89.5% human baseline, with major weaknesses in physical and spatial reasoning.
-
Tracing the Arrow of Time: Diagnosing Temporal Information Flow in Video-LLMs
Temporal information in Video-LLMs is encoded well by video-centric encoders but disrupted by standard projectors; time-preserved MLPs plus AoT supervision yield 98.1% accuracy on arrow-of-time and gains on other temporal tasks.
-
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.
-
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.
-
Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context
Continued pre-training with balanced long-document VQA data extends a 7B LVLM to 128K context, improving long-document VQA by 7.1% and generalizing to 512K without further training.
-
OTT-Vid: Optimal Transport Temporal Token Compression for Video Large Language Models
OTT-Vid uses optimal transport with non-uniform token mass and locality-aware costs to dynamically allocate compression budgets across video frames, retaining 95.8% VQA and 73.9% VTG performance at 10% token retention.
-
LDDR: Linear-DPP-Based Dynamic-Resolution Frame Sampling for Video MLLMs
LDDR proposes a linear DPP-based dynamic-resolution frame sampler that achieves 3x speedup and up to 2.5-point gains on video MLLM benchmarks by selecting non-redundant frames and allocating tokens accordingly.
-
Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding
Video-MME-v2 is a new benchmark that applies progressive visual-to-reasoning levels and non-linear group scoring to expose gaps in video MLLM capabilities.
-
OmniSelect: Dynamic Modality-Aware Token Compression for Efficient Omni-modal Large Language Models
OmniSelect is a training-free, modality-adaptive token pruning framework that dynamically selects Audio-Centric, Video-Centric, or Uniform compression regimes using AudioCLIP cross-modal relevance scores and then applies adaptive fine-grained pruning within temporal groups.
-
OmniRefine: Alignment-Aware Cooperative Compression for Efficient Omnimodal Large Language Models
OmniRefine introduces alignment-aware chunk refinement via similarity and dynamic programming followed by modality-cooperative token compression, achieving near-baseline accuracy at 44% token retention on WorldSense.