RobustSora benchmark demonstrates that current AI video detectors rely heavily on visible watermarks, with average accuracy drops of 6.6 percentage points when watermarks are erased and increased false alarms when watermarks are spoofed onto real videos.
hub Mixed citations
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
Mixed citation behavior. Most common role is background (64%).
89
Pith papers citing it
Background
64% of classified citations
abstract
The Large Vision-Language Model (LVLM) has enhanced the performance of various downstream tasks in visual-language understanding. Most existing approaches encode images and videos into separate feature spaces, which are then fed as inputs to large language models. However, due to the lack of unified tokenization for images and videos, namely misalignment before projection, it becomes challenging for a Large Language Model (LLM) to learn multi-modal interactions from several poor projection layers. In this work, we unify visual representation into the language feature space to advance the foundational LLM towards a unified LVLM. As a result, we establish a simple but robust LVLM baseline, Video-LLaVA, which learns from a mixed dataset of images and videos, mutually enhancing each other. Video-LLaVA achieves superior performances on a broad range of 9 image benchmarks across 5 image question-answering datasets and 4 image benchmark toolkits. Additionally, our Video-LLaVA also outperforms Video-ChatGPT by 5.8%, 9.9%, 18.6%, and 10.1% on MSRVTT, MSVD, TGIF, and ActivityNet, respectively. Notably, extensive experiments demonstrate that Video-LLaVA mutually benefits images and videos within a unified visual representation, outperforming models designed specifically for images or videos. We aim for this work to provide modest insights into the multi-modal inputs for the LLM. Code address: \href{https://github.com/PKU-YuanGroup/Video-LLaVA}
hub tools
citation-role summary
background 8
baseline 3
method 2
other 1
citation-polarity summary
claims ledger
- abstract The Large Vision-Language Model (LVLM) has enhanced the performance of various downstream tasks in visual-language understanding. Most existing approaches encode images and videos into separate feature spaces, which are then fed as inputs to large language models. However, due to the lack of unified tokenization for images and videos, namely misalignment before projection, it becomes challenging for a Large Language Model (LLM) to learn multi-modal interactions from several poor projection layers. In this work, we unify visual representation into the language feature space to advance the found
- baseline "×" indicates the model is incapable of performing the task. Model Understanding Image Generation Image Editing MMBV MMBI MMMU MM-Vet GenEval WISE Overall Add Adjust Extract Replace Remove Hybird Image Understanding LLaV A-1.5 [25] × 36.4 67.8 36.3 × × × × × × × × × LLaV A-NeXT [57] × 79.3 51.1 57.4 × × × × × × × × × Image & Video Understanding Video-LLaV A [22] 1.05 60.9 32.8 32.0 × × × × × × × × × LLaV A-OV [17] 0.94 80.8 48.8 57.5 × × × × × × × × × Text-to-Image Generation SDXL [34] × × × × 0
- method backbone [13], as illustrated in Figure 2. The edit instruction and the original image are jointly fed into VLM, while the image is processed simultaneously by the vision encoder. The hidden states of VLM and the visual feature of the vision encoder are separately projected by MLPs and then concatenated, forming the text-branch input to DiT. Training proceeds in two stages [41], first optimizing MLPs and then jointly fine-tuning FLUX and MLPs. 3.4 Dataset Statistics ImgEdit comprises 1.2 million
- background SigLIP outperforms the other two vision encoders, especially in fine-grained understanding tasks involving texts. Based on this ablation study, we choose the pretrained SigLIP as our base vision encoder, and then adapt it to taking dynamic resolutions as inputs. 5 Related Work Multimodal LLMs for Native Video Understanding. Early video MLLMs primarily relied on sparsely sampled frames and simple connectors, such as MLPs [12, 13, 139], discrete visual tokenizers [140], and Q-formers [141, 142], t
- background SpatialLadder: Progressive training for spatial reasoning in vision-language models.arXiv, abs/2510.08531, 2025. [31] Junnan Li, Dongxu Li, Silvio Savarese, and Steven C. H. Hoi. BLIP-2: Bootstrapping language- image pre-training with frozen image encoders and large language models. InInternational Conference on Machine Learning, volume 202, pages 19730-19742, 2023. [32] Bin Lin, Bin Zhu, Yang Ye, Munan Ning, Peng Jin, and Li Yuan. Video-LLaV A: Learning united visual representation by alignment
- method [74] Yitong Li, Zhe Gan, Yelong Shen, Jingjing Liu, Yu Cheng, Yuexin Wu, Lawrence Carin, David Carlson, and Jianfeng Gao. Storygan: A sequential conditional gan for story visualization, 2019. 40 [75] Zhuowan Li, Xingrui Wang, Elias Stengel-Eskin, Adam Kortylewski, Wufei Ma, Benjamin Van Durme, and Alan Yuille. Super-clevr: A virtual benchmark to diagnose domain robustness in visual reasoning, 2023. 39 [76] Bin Lin, Bin Zhu, Yang Ye, Munan Ning, Peng Jin, and Li Yuan. Video-llava: Learning united
- background language models. InECCV, 2024. 3 [37] Yucheng Li, Huiqiang Jiang, Chengruidong Zhang, Qianhui Wu, Xufang Luo, Surin Ahn, Amir H Abdi, Dongsheng Li, Jianfeng Gao, Yuqing Yang, et al. Mapsparse: Accelerating pre-filling for long-context visual language models via modality-aware permutation sparse attention. InICLR 2025 Workshop on Foundation Models in the Wild. 5 [38] Bin Lin, Yang Ye, Bin Zhu, Jiaxi Cui, Munan Ning, Peng Jin, and Li Yuan. Video-llava: Learning united visual representation by alig
co-cited works
representative citing papers
Introduces VidPair-Halluc benchmark of 1K background-controlled adversarial video pairs and 11K QA pairs generated via PairFlow pipeline to evaluate hallucination in LVMs.
SelfBootTok decomposes image tokens into global and local groups via self-bootstrapped learning, enabling generators to use only global tokens for ~40% less computation and a new SOTA gFID of 1.56 with 64 tokens.
EvoCut is a training-free visual token compression technique that identifies important tokens via multi-layer evolution deviation, retaining 11.1% tokens with 94.4% average performance preserved on LLaVA-1.5-7B.
AffectVerse improves multimodal emotion recognition by at least 2.57% on nine benchmarks through an Emotion World Module that performs short-horizon latent affective prediction via cross-modal temporal imagination and belief aggregation.
CoRDS selects a compact KV-cache subset via joint-space coreset coverage and log-det diversity to outperform token-wise heuristics on long-video VLM benchmarks.
WirelessSenseLLM bridges unsegmented Wi-Fi CSI signals to LLMs via a CSI-to-Language Adapter for zero-shot human activity understanding and reasoning.
A proposer-solver agent pair achieves supervised-level video temporal grounding and fine-grained captioning from 2.5K unlabeled videos via self-reinforcing evolution.
EyeCue detects driver cognitive distraction by modeling gaze-visual context interactions in egocentric videos and achieves 74.38% accuracy on the new CogDrive dataset, outperforming 11 baselines.
LearnPruner prunes vision tokens to 5.5% of the original count while retaining about 95% of VLM performance and delivering 3.2 times faster inference by fixing attention sink in encoders and using unbiased middle-layer attention in LLMs.
A new benchmark converts video clips into shared grounded event records and tests models across physics, semantic, and control prompts under original, shuffled, ablated, and masked conditions, finding selective robustness and weak spatial performance.
SAri-RFT applies GRPO-based reinforcement fine-tuning to LVLMs on novel two-term and three-term visual semantic arithmetic tasks, reaching SOTA on the new IRPD dataset and Visual7W-Telling.
SVAgent improves long video question answering by constructing storylines via multi-agent collaboration and aligning cross-modal predictions for more robust, human-like reasoning.
VideoThinker uses LLM-generated synthetic tool trajectories in caption space grounded to video frames to train agentic VideoLLMs that outperform baselines on long-video benchmarks.
LFS learns to select temporally diverse and event-aware frames for video captioning by using direct feedback from frozen video-LLMs, yielding gains up to 2% on VDC and over 4% on the new ICH-CC benchmark.
GAR-Font is a global-aware autoregressive framework for multimodal few-shot font generation that adds global tokenization, a language-style adapter, and post-refinement to improve style coherence over patch-based methods.
StreamGaze is a new benchmark and QA generation pipeline that measures how well MLLMs leverage gaze trajectories for temporal reasoning and proactive intention prediction in streaming egocentric videos.
Introduces the first dedicated benchmark for live multi-modal LLM task guidance with mistake detection and a streaming baseline model.
Introduces TennisTV benchmark for evaluating 17 MLLMs on tennis video understanding from stroke-level to rally-level tasks with automated pipelines and human verification.
WorldSense provides the first benchmark requiring synergistic audio-video-text understanding on 1,662 real-world videos and 3,172 QA pairs, where the best current multimodal LLM reaches only 65.1% accuracy.
PyramidDrop accelerates LVLMs by staged, similarity-based dropping of visual tokens that become redundant in deeper layers, delivering 40% faster training and 55% lower inference cost with comparable accuracy.
MLVU is a new benchmark for long video understanding that uses extended videos across diverse genres and multi-task evaluations, revealing that current MLLMs struggle significantly and degrade sharply with longer durations.
HPP decouples perception from reasoning in long-video VLMs by having an LLM run iterative programmatic probes on hierarchically segmented video, reporting gains on LongVideoBench, EgoSchema, VideoMME, and MLVU.
T-MOR is a multi-modal contrastive framework that pre-trains transferable skeleton motion representations using a new 1M video-skeleton-text dataset and shows gains on action classification and temporal detection benchmarks plus few/zero-shot settings.
citing papers explorer
-
ImgEdit: A Unified Image Editing Dataset and Benchmark
ImgEdit supplies 1.2 million curated edit pairs and a three-part benchmark that let a VLM-based model outperform prior open-source editors on adherence, quality, and detail preservation.
-
LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval
LiveVLM introduces VSB and PaR to compress and retrieve KV cache in streaming video LLMs, enabling LLaVA-OneVision to reach SOTA accuracy among training-free query-agnostic and training-based online models.
-
FaVChat: Hierarchical Prompt-Query Guided Facial Video Understanding with Data-Efficient GRPO
FaVChat proposes hierarchical prompt-query guided visual features and Data-Efficient GRPO for efficient training, plus the FaVChat-170K dataset, claiming consistent outperformance over prior VLLMs on facial video tasks.
-
Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Tasks
Uni-NaVid unifies diverse embodied navigation tasks into one video-based vision-language-action model trained on 3.6 million samples from four sub-tasks, achieving state-of-the-art performance on benchmarks and real-world tests.
-
LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
LongVU adaptively compresses long video tokens using DINOv2-based frame deduplication, text-guided cross-modal selection, and temporal spatial reduction to improve video-language understanding in MLLMs with minimal detail loss.
-
LongVILA: Scaling Long-Context Visual Language Models for Long Videos
LongVILA scales visual-language models from 8 to 2048 video frames with 99.8% needle-in-a-haystack accuracy using long-context extension, supervised fine-tuning, and multi-modal sequence parallelism on up to 256 GPUs.
-
What to Say and When to Say it: Live Fitness Coaching as a Testbed for Situated Interaction
Introduces the QEVD benchmark for asynchronous situated interaction in fitness coaching and proposes a streaming baseline to address limitations of existing vision-language models.
-
TempCompass: Do Video LLMs Really Understand Videos?
TempCompass benchmark reveals that state-of-the-art Video LLMs have poor ability to perceive temporal aspects such as speed, direction, and ordering in videos.
-
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
MoE-LLaVA applies mixture-of-experts sparsity to LVLMs via MoE-Tuning, delivering LLaVA-1.5-7B level visual understanding and better hallucination resistance with only ~3B active parameters.
-
Gemini: A Family of Highly Capable Multimodal Models
Gemini Ultra reaches human-expert performance on MMLU for the first time and sets new state-of-the-art results on 30 of 32 benchmarks, including all 20 multimodal ones tested.
-
LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams
LiveStarPro uses SVeD for response timing via perplexity, SCAM for incremental alignment, and TSHM for event-chain memory to achieve 28.9% better semantic correctness and 1.58x speedup on long video streams.
-
AudioX-Turbo: A Unified Framework for Efficient Anything-to-Audio Generation
AudioX-Turbo distills a Multimodal Diffusion Transformer into a 4-step student model for efficient multimodal anything-to-audio generation, trained on a new 9.2M-sample dataset IF-caps-Pro.
-
Audio-Visual Exchange-Aware Token Pruning for Efficient Audio-Visual Captioning
AVEX-Prune is an RL-based audio-visual token pruning method using modality exchange that maintains near-full performance at 40% token retention on VILA 1.5-8B and VideoLLaMA 2.
-
Question-Aware Evidence Ledgers for Video Relational Reasoning
A pipeline using question-aware evidence ledgers with GPT-5.5 achieves 92.95% overall and 93.79% macro accuracy on the VRR-QA video relational reasoning challenge.
-
Vision-language Models for Driver Monitoring Systems: A Driver Activity Description Dataset
Introduces Drive&Act description dataset of fine-grained driver activity text and reports fine-tuned VLM reaching 76 ACCR on DMD dataset versus 66 for zero-shot baseline.
-
ADMFormer: An Adaptive-Decomposition Transformer with Time-Varying Masked Spatial Attention for Traffic Forecasting
ADMFormer decouples traffic into regular and fluctuating components with time-node gating, processes them in dual temporal branches, and uses time-varying masked spatial attention to reach SOTA on four datasets.
-
MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering
MuKV adds multi-grained KV cache compression at patch-frame-segment levels plus semi-hierarchical retrieval to raise accuracy and cut memory in long video question-answering.
-
Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models
AOT reduces visual tokens in VLLMs via intra-frame and inter-frame anchors with local-global optimal transport, delivering competitive benchmark performance and efficiency gains in a training-free way.
-
Enhancing Speech Large Language Models through Reinforced Behavior Alignment
Reinforced Behavior Alignment (RBA) uses self-synthesized data from a teacher LLM and reinforcement learning to close the instruction-following gap in SpeechLMs, outperforming distillation and reaching SOTA on spoken QA and speech-to-text translation benchmarks.
-
UniWorld-V1: High-Resolution Semantic Encoders for Unified Visual Understanding and Generation
UniWorld-V1 shows that semantic features from large multimodal models enable unified visual understanding and generation, achieving strong results on perception and manipulation tasks with only 2.7 million training samples.
-
Growing a Multi-head Twig via Distillation and Reinforcement Learning to Accelerate Large Vision-Language Models
TwigVLM adds a twig module to VLMs for twig-guided token pruning and self-speculative decoding, retaining 96% performance after pruning 88.9% visual tokens and delivering 154% speedup on long responses for LLaVA-1.5-7B.
-
LLaVA-Octopus: Unlocking Instruction-Driven Adaptive Projector Fusion for Video Understanding
LLaVA-Octopus introduces instruction-driven adaptive fusion of multiple visual projectors in a multimodal LLM to improve video understanding performance.
-
mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models
mPLUG-Owl3 introduces hyper attention blocks to integrate vision and language for long image-sequence understanding and reports SOTA results on single-image, multi-image, and video benchmarks.
-
LLaVA-OneVision: Easy Visual Task Transfer
LLaVA-OneVision is the first single open LMM to simultaneously achieve strong performance in single-image, multi-image, and video scenarios with cross-scenario transfer capabilities.
-
InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output
InternLM-XComposer-2.5 is a 7B vision-language model supporting up to 96K context that reaches GPT-4V-level performance on image, video, and multi-turn tasks and adds LoRA-driven text-image composition capabilities.
-
PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning
A temporal pooling layer added to LLaVA smooths video feature distributions and lifts performance on dense video captioning and QA to new SOTA levels without extra parameters.
-
World Model on Million-Length Video And Language With Blockwise RingAttention
Presents open-source 7B models for million-token video and language understanding via Blockwise RingAttention, setting new benchmarks in retrieval and long video tasks.
-
InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning
InternVideo3 introduces Multimodal Contextual Reasoning and M^2LA attention to enable closed-loop evidence accumulation in long-video understanding and agentic tool use, reporting strong benchmark results.
-
Kwai Keye-VL-2.0 Technical Report
Kwai Keye-VL-2.0-30B-A3B is a 30B MoE model with 3B active parameters using DSA adaptation and MOPD distillation that reports SOTA results on video understanding and agent benchmarks.
-
UNIVID: Unified Vision-Language Model for Video Moderation
UNIVID generates policy-aware captions for video moderation, reducing violation leakage by 42.7% and overkill rate by 37.0% while replacing over 1,000 policy-specific models with a single backbone.
-
CREST: Curvature-Regulated Event-Centric Sampling for Efficient Long-Video Understanding
CREST uses local curvature of query-frame relevance over time to select informative frames, outperforming a lightweight baseline and approaching a costly pipeline at far lower preprocessing cost on long-video benchmarks.
-
AirVista-II: An Agentic System for Embodied UAVs Toward Dynamic Scene Semantic Understanding
AirVista-II integrates agent-based task identification and scheduling, multimodal perception, and scenario-tailored keyframe extraction to deliver high-quality zero-shot semantic understanding for embodied UAVs in dynamic environments.
-
VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
VideoLLaMA3 uses a vision-centric training paradigm and token-reduction design to reach competitive results on image and video benchmarks.
-
VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction
VITA-1.5 integrates vision and speech into a single LLM through multi-stage training, delivering competitive benchmark results on image, video, and speech tasks with near real-time response speed.
-
Open-Sora Plan: Open-Source Large Video Generation Model
Open-Sora Plan presents an open-source large video generation model that combines a Wavelet-Flow VAE, Joint Image-Video Skiparse Denoiser, and multi-dimensional data curation to achieve high-quality video outputs with public code and weights.
-
VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs
VideoLLaMA 2 improves video LLMs via a new STC connector for spatial-temporal dynamics and joint audio training, reaching competitive results on video QA and captioning benchmarks.
-
Toward Native Multimodal Modeling: A Roadmap
A roadmap that defines architectural nativity for multimodal models and categorizes them into Multi-to-Text, Multi-to-Target, and Multi-to-Multi types while outlining an industrial pipeline toward unified transformer-based native multimodal modeling.
-
A Survey on Hallucination in Large Vision-Language Models
This survey reviews the definition, symptoms, evaluation benchmarks, root causes, and mitigation methods for hallucinations in large vision-language models.
- Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark