OVOW reconstructs instance-level, simulation-ready 4D mesh scenes from monocular video via a four-stage training-free pipeline and introduces a new benchmark for structured Video-to-4D evaluation.
mega hub Mixed citations
Qwen3-VL Technical Report
Mixed citation behavior. Most common role is background (48%).
1181
Pith papers citing it
Background
48% of classified citations
abstract
We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively supports interleaved contexts of up to 256K tokens, seamlessly integrating text, images, and video. The model family includes both dense (2B/4B/8B/32B) and mixture-of-experts (30B-A3B/235B-A22B) variants to accommodate diverse latency-quality trade-offs. Qwen3-VL delivers three core pillars: (i) markedly stronger pure-text understanding, surpassing comparable text-only backbones in several cases; (ii) robust long-context comprehension with a native 256K-token window for both text and interleaved multimodal inputs, enabling faithful retention, retrieval, and cross-referencing across long documents and videos; and (iii) advanced multimodal reasoning across single-image, multi-image, and video tasks, demonstrating leading performance on comprehensive evaluations such as MMMU and visual-math benchmarks (e.g., MathVista and MathVision). Architecturally, we introduce three key upgrades: (i) an enhanced interleaved-MRoPE for stronger spatial-temporal modeling across images and video; (ii) DeepStack integration, which effectively leverages multi-level ViT features to tighten vision-language alignment; and (iii) text-based time alignment for video, evolving from T-RoPE to explicit textual timestamp alignment for more precise temporal grounding. Under comparable token budgets and latency constraints, Qwen3-VL achieves superior performance in both dense and Mixture-of-Experts (MoE) architectures. We envision Qwen3-VL serving as a foundational engine for image-grounded reasoning, agentic decision-making, and multimodal code intelligence in real-world workflows.
hub tools
citation-role summary
background 122
method 61
baseline 50
dataset 5
other 4
citation-polarity summary
claims ledger
- abstract We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively supports interleaved contexts of up to 256K tokens, seamlessly integrating text, images, and video. The model family includes both dense (2B/4B/8B/32B) and mixture-of-experts (30B-A3B/235B-A22B) variants to accommodate diverse latency-quality trade-offs. Qwen3-VL delivers three core pillars: (i) markedly stronger pure-text understanding, surpassing comparable text-only backbones in several cases; (ii) robust long-con
authors
mega hub controls
export citing contexts JSON
export graph JSON
export full bundle JSON
open full Pith review
annotated reader queued
Recognition alignment
counterfactual ablation
co-cited works
representative citing papers
A blank-image ablation test reveals that high probe accuracy on VLM spatial reasoning frequently reflects priors or inverted signs rather than image grounding, with horizontal grounded, vertical prior, and depth inverted.
DataComp-VLM benchmark shows instruction-heavy data mixing outperforms filtering for VLM training, with DCVLM-Baseline achieving 63.6% on 33 tasks for 8B models (+5.4pp over FineVision).
Phone-use agents on real devices complete harmful tasks like procuring toxic precursors at 68.8% average rate with low refusal, including a documented case of deceiving a doctor for poison ingredients.
Introduces the first large-scale multimodal benchmark MedLayXPlain-122K showing medical VLMs suffer significant lay-register degradation while general VLMs lack clinical precision.
LOCUS is a released corpus of nearly all US municipal and county ordinance codes, processed via OCR and paired with ModernBERT classifiers for dimensions such as opacity and paternalism.
A causal audit with image interventions shows text-only models reach within 5.7 accuracy points of top multimodal VLMs on chest radiography, with some large multimodal models statistically indistinguishable from small text-only baselines.
RobotValues is a benchmark of 10K value-conflict scenarios that reveals VLMs default to safety and accommodation while failing to follow instructions to prioritize other values 80% of the time.
FigSIM is the first annotated dataset for fine-grained suicide severity and figurative language in suicide memes, accompanied by benchmarks on 16 unimodal and multimodal models.
ViMU is the first benchmark for evaluating video models on metaphorical and subtextual understanding using hint-free questions grounded in multimodal evidence.
CiteVQA requires models to cite specific document regions with bounding boxes alongside answers and finds that even the strongest MLLMs frequently cite the wrong region, with top SAA scores of only 76.0 for closed models and 22.5 for open-source ones.
SenseBench is the first physics-based benchmark with 10K+ instances and dual protocols to evaluate VLMs on remote sensing low-level perception and diagnostic description, revealing domain bias and specific failure modes.
EgoMemReason is a new benchmark showing that even the best multimodal models achieve only 39.6% accuracy on reasoning tasks that require integrating sparse evidence across days in egocentric video.
RuleSafe-VL creates 2,166 rule-conditioned cases from 93 atomic rules and 92 relations across three policy families to diagnose where VLMs fail at rule-based content moderation reasoning.
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.
PureDocBench shows document parsing is far from solved, with top models at ~74/100, small specialists competing with large VLMs, and ranking reversals under real degradation.
MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
WindowsWorld benchmark shows leading GUI agents achieve under 21% success on multi-application professional tasks, with failures especially on conditional judgment across three or more apps and inefficient execution.
MM-JudgeBench shows substantial cross-lingual performance variance in 22 LVLM judges, with model size and architecture as poor predictors of multilingual robustness.
EVE enables verifiable self-evolution of MLLMs by using a Challenger-Solver architecture to generate dynamic executable visual transformations that produce VQA problems with absolute execution-verified ground truth.
VLMs hallucinate by prioritizing contradictory on-screen text over visual content, addressed via the VisualTextTrap benchmark with 6,057 human-validated samples and the VTHM-MoE dual-encoder framework using dimension-specific experts and adaptive routing.
RefereeBench shows that even the strongest video MLLMs reach only around 60% accuracy on multi-sport refereeing tasks and struggle with rule application and temporal grounding.
VLM-UnBench demonstrates that prompt-based training-free unlearning in VLMs leaves forget accuracy near the no-instruction baseline except under oracle conditions that reveal the target concept.
ScreenParse dataset and ScreenVLM model deliver dense screen parsing that outperforms larger VLMs on PageIoU and transfers to better UI grounding.
citing papers explorer
-
ARGUS: Stacked Multi-View Identity Mosaic Injection for Subject-Preserving Video Generation
ARGUS converts MLLM-selected identity evidence into a synchronized 3x3 mosaic injected as negative-time memory in a diffusion model, plus supporting training techniques, to achieve SOTA subject preservation on human video benchmarks.
-
CoCoSI: Collaborative Cognitive Map Construction for Spatial Intelligence
CoCoSI is a training-free multi-agent system for collaborative cognitive map construction that improves spatial understanding in arbitrary pretrained MLLMs.
-
Streaming Interventions: Can Video Large Language Models Correct Mistakes as They Occur?
Introduces Ego-MC-Bench benchmark and Ego-CoMist synthetic dataset showing that fine-tuning video LLMs on proactive mistake corrections improves performance especially for smaller models.
-
Temporal-Aware Reasoning Optimization for Video Temporal Grounding
TaRO improves video temporal grounding in MLLMs via constructive reasoning exploration from dense captions and a temporal-sensitivity reward that uses logit drops on disrupted event boundaries, followed by curriculum learning to SOTA results.
-
HDRAgent: An Agentic Framework for Multi-Exposure HDR Imaging
HDRAgent is the first agent-driven framework for multi-exposure HDR imaging that uses MLLM scene perception, contextual knowledge matching, and perception-distortion feedback to reduce ghosting artifacts.
-
Harnessing Streaming Video in the Wild
Presents Streaming-Train-248K dataset, Streaming Harness system, and Streaming-Eval benchmark to enable VLMs for proactive, memory-equipped streaming video understanding.
-
Streaming Video Generation with Streaming Force Control
StreamForce presents a unified causal model for force-controllable streaming video generation using a new force representation and distillation pipeline, claiming SOTA force adherence and 16.6 FPS performance.
-
VeriDrive: Verifiable Counterfactual Supervision for Cost-Efficient Vision-Language Planning
VeriDrive introduces a verifiable counterfactual supervision framework using a Perception-Evaluation-Revision chain and validator-guided correction to generate cost-efficient structured data for vision-language driving models, showing metric gains on nuScenes.
-
Stream3D-VLM: Online 3D Spatial Understanding with Incremental Geometry Priors
Stream3D-VLM adds autoregressive streaming control, VSFI geometry integration, GAVC compression, and a 1M-pair benchmark to enable real-time 3D VLM performance that beats prior models on 29 online and offline tasks.
-
Diagnosing Visual Ignorance in Vision-Language Models
VLMs show language-prior reliance via multi-stage bottlenecks in visual retrieval and suppression, with many benchmark examples remaining answerable under severe visual obfuscation.
-
PAR3D: A Unified 3D-MLLM with Part-Aware Representation for Scene Understanding
PAR3D is a part-aware 3D-MLLM framework with ScenePart dataset, Part-Aware 3D Representation Learning, and Hierarchical Segmentation Query Generation to improve part-level 3D scene understanding.
-
DRIFT: A Residual Flow Adapter for Decoding Continuous Outputs in Vision-Language Models
DRIFT adapts pretrained VLMs to continuous decoding via a base predictor plus residual flow matching, outperforming regression and generative baselines on grounding and robotic control tasks.
-
TextWand: A Unified Framework for Scene Text Editing
TextWand unifies scene text removal, generation and replacement via rendering/erasure decomposition, ORPE for layout fidelity, RAS for clean erasure, and the new TextWand-Bench dataset, claiming superior accuracy and quality over prior models.
-
ViCuR: Visual Cues as Recoverable Privilege for Multimodal On-Policy Distillation
ViCuR introduces recoverable visual cues as teacher privilege in multimodal on-policy distillation, yielding +1.19 to +1.24 average gains over answer-based baselines across seven benchmarks with Qwen3-VL students.
-
ShotCrop$^3$: Cropping Human-Centric Images into Cinematic Triple-Shot Compositions
ShotCrop uses three-stage training (CoT SFT, pseudo-label semi-supervised, GRPO-S) to produce triple-shot compositions and reports 2.82x better shot localization than GPT-5 on a 1.2k expert benchmark.
-
WorldBench: A Challenging and Visually Diverse Multimodal Reasoning Benchmark
WorldBench is a visually diverse multimodal reasoning benchmark where the strongest of 15 tested MLLMs reaches only 64% accuracy.
-
SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation
SD-GRPO extends GRPO by computing per-segment advantages via z-normalization of verifiable segment rewards, yielding gains on long-form VL tasks with varying semantic independence across segments.
-
TGV-KV: Text-Grounded KV Eviction for Vision-Language Models
TGV-KV uses text-vision budgeting, weighted ranking, and prioritised retention to evict KV cache in VLMs while retaining 99.2% accuracy at 5% budget on VizWiz-VQA.
-
AdaCodec: A Predictive Visual Code for Video MLLMs
AdaCodec introduces a predictive visual code that cuts visual token use in video MLLMs by sending full frames only on high predictive cost and otherwise encoding inter-frame changes as P-tokens, yielding better benchmark scores at lower budgets.
-
Readable Yet Unpredictable: Rotated-Outcome Prediction in Vision-Language Models
VLMs recognize rotated images when shown directly but fail to predict rotated outcomes from originals on the new RotOutBench benchmark.
-
InsightVQA: High-Dimensional Emotion-Cognitive Visual Question Answering Benchmark
The paper creates InsightVQA, a 725K QA-pair benchmark with perception, grounded-understanding, and cognition levels for emotion-cognitive visual question answering, plus a 30K-sample evaluation set and InsightNet baseline.
-
InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models
InfoMerge proposes a training-free visual token compression method for Video-LLMs that uses Temporal Fingerprint Difference for redundancy estimation and Content-Aware Budget Allocation to retain 98.8% performance with 85% fewer tokens.
-
MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention
MOSS-Video-Preview introduces a cross-attention architecture and synthesized real-time QA data to enable continuous perception, answer revision, and faster inference in video-language models compared to decoder-only designs.
-
STaR-KV: Spatio-Temporal Adaptive Re-weighting for KV Cache Compression in GUI Vision-Language Models
STaR-KV is a training-free KV cache compression framework for GUI VLMs that uses subspace-aware scoring, temporal stability discounts, and entropy-based temperature adaptation to outperform prior methods at matched budgets while reducing peak memory by ~40% at 20% cache size.
-
PAI-Studio: Cinematic Video Background Replacement with Camera-Aware Motion
PAI-Studio reformulates cinematic background replacement as in-context conditional generation inside a Diffusion Transformer with bidirectional attention, trained on a new 30K film-sourced dataset, and reports better motion consistency and relighting than prior open-source and commercial systems.
-
Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning
Reasmory turns 3D reconstruction into validated program-executable memory for VLMs, yielding 6-18% gains on spatial reasoning benchmarks over direct baselines.
-
RoboStressBench: Benchmarking VLM Robustness to Physical Visual Stress in Embodied Scenes
RoboStressBench decomposes visual stress into four physically grounded dimensions to benchmark VLM robustness in embodied scenes and proposes a stress-aware solver.
-
FlowNar: Scalable Streaming Narration for Long-Form Videos
FlowNar achieves bounded memory and 3x higher throughput for streaming narration on Ego4D, EgoExo4D, and EpicKitchens100 by combining dynamic historical context removal with a Cross Linear Attentive Memory module.
-
Zamba2-VL Technical Report
Zamba2-VL is a family of 1.2B–7B hybrid Mamba2-transformer vision-language models that match leading transformer VLMs on image, reasoning, OCR, grounding and counting benchmarks while delivering roughly 10x lower time-to-first-token.
-
StressDream: Steering Video World Models for Robust Policy Evaluation and Improvement
StressDream optimizes initial noise in diffusion video world models using VLM semantic and plausibility objectives to steer generations toward specified high-impact outcomes for improved policy evaluation.
-
SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models
SOCO is a new benchmark for semantic object correspondence that provides taxonomy, annotations, and language labels to evaluate part-level understanding in vision and multimodal foundation models.
-
nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving
nuReasoning is a new real-world dataset and benchmark extending nuScenes/nuPlan with 20k clips and multi-type reasoning annotations to evaluate and improve reasoning in long-tail autonomous driving.
-
Detect in Any Scene: An Agentic Framework for Object Detection with Experience-Aware Reasoning
DetAS-X uses an MLLM agent to adaptively compose detection workflows from restoration modules and expert detectors, enhanced by self-evolving experience harvesting, achieving substantial F1 score gains on challenging benchmarks.
-
Task-Focused Memorization for Multimodal Agents
TaskMem uses RL in two phases to learn a task-focused memorization policy for multimodal agents, yielding 5.3-7.0% VQA accuracy gains on reformulated streaming benchmarks from VideoMME, EgoLife, and EgoTempo.
-
PEEK: Picking Essential frames via Efficient Knowledge distillation
PEEK distills caption-conditioned frame relevance into a lightweight visual model, outperforming adaptive baselines on ActivityNet Captions and MSR-VTT especially at 1-2 frame budgets while adding only 5.2% overhead.
-
Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation
The paper diagnoses Template Collapse in 3D medical VLMs for CT reports and introduces CLarGen, a decoupled detection-plus-synthesis framework that raises macro-F1 from 0.189 to 0.487.
-
VLM3: Vision Language Models Are Native 3D Learners
Standard VLMs achieve expert-level 3D performance on depth estimation, pose estimation, and object understanding via three simple techniques without architecture changes or regression losses.
-
GPIC: A Giant Permissive Image Corpus for Visual Generation
GPIC is a new 28-trillion-pixel permissively licensed image corpus with 100M training examples for visual generative modeling.
-
LoMo: Local Modality Substitution for Deeper Vision-Language Fusion
LoMo is a lightweight data curation technique that locally substitutes text with images in prompts to enforce cross-modal invariance, yielding 2.67-2.82 point gains over standard SFT on two VLMs across 13 benchmarks.
-
AgentCVR: Active Multi-Agent Cross-Video Reasoning via Script-Simulated Reinforcement Learning
AgentCVR uses coordinated multi-agent active search and script-simulated RL to improve cross-video reasoning in MLLMs over single-pass baselines.
-
Mitigating State Aliasing in Vision-Language-Action Models via Inverse Dynamics Learning
Inverse dynamics prediction is added as an auxiliary task to reduce state aliasing in VLA models by directly supervising the vision encoder on action-relevant visual distinctions using only standard observation-action pairs.
-
AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling
AnyMo is a masked-modeling framework for any-modality human motion generation trained on the new OmniHuMo dataset of 5,000+ hours of multimodal motion sequences.
-
Guidance Contrastive Token Credit Assignment for Discrete Policy Optimization
GCPO performs per-token credit assignment in discrete policy optimization by setting token advantages proportional to the difference in model predictions under positive versus negative prompts, outperforming GRPO and DAPO on text-to-image and chain-of-thought tasks.
-
Self-Prophetic Decoding to Unlock Visual Search in LVLMs
SeProD is a plug-and-play self-prophetic decoding framework that combines pre- and post-training LVLM capabilities via probability-based sampling to improve coherent visual search and multi-step reasoning.
-
DriveWAM: Video Generative Priors Enable Scalable World-Action Modeling for Autonomous Driving
DriveWAM converts video generative priors into a unified video-action policy for driving, reporting strong benchmark performance and positive scaling from 4k to 100k clips.
-
Proprio: Latent Self-Scoring and Inference-Time Refinement for Physically Plausible Video Generation
Proprio uses flow residuals from latent perturbations in frozen video generators as a self-scoring signal for physical plausibility, yielding reported gains of 16.5% on Physics-IQ and 20.6% on VideoPhy2-hard.
-
SmartDirector: Keyframe-Conditioned Cinematic Video Generation with Narrative Pacing Control
SmartDirector generates cinematic videos via Director-Gen for low-res keyframe-conditioned output followed by Director-SR refinement using high-res keyframes, trained on curated movie sequences.
-
Reflective Dialogue between Teacher and Solver Agents for Video Question Answering
A multi-turn reflective dialogue between Teacher and Solver agents constructs richer context from support examples than standard in-context learning, improving video QA on the EgoCross benchmark.
-
Can Segmentation Models Understand the World? Towards Proactive Affordance Reasoning via Visual Chain-of-Thought
SegWorld adds multi-level visual CoT and proactive scene observation to segmentation models, formalized as probabilistic inference, and shows gains on an intent-to-part benchmark for affordance segmentation.
-
Pop-Up Distractions Reveal Bag-of-Events Behavior in Video Large Language Models
VideoLLMs process videos as unordered collections of events and hallucinate cross-segment interactions when distractions are inserted, a behavior observed across all 11 tested models.