Molmo VLMs trained on newly collected PixMo open datasets achieve state-of-the-art performance among open-weight models and surpass multiple proprietary VLMs including Claude 3.5 Sonnet and Gemini 1.5 Pro.
hub Canonical reference
CogVLM2: Visual Language Models for Image and Video Understanding
Canonical reference. 83% of citing Pith papers cite this work as background.
32
Pith papers citing it
Background
83% of classified citations
abstract
Beginning with VisualGLM and CogVLM, we are continuously exploring VLMs in pursuit of enhanced vision-language fusion, efficient higher-resolution architecture, and broader modalities and applications. Here we propose the CogVLM2 family, a new generation of visual language models for image and video understanding including CogVLM2, CogVLM2-Video and GLM-4V. As an image understanding model, CogVLM2 inherits the visual expert architecture with improved training recipes in both pre-training and post-training stages, supporting input resolution up to $1344 \times 1344$ pixels. As a video understanding model, CogVLM2-Video integrates multi-frame input with timestamps and proposes automated temporal grounding data construction. Notably, CogVLM2 family has achieved state-of-the-art results on benchmarks like MMBench, MM-Vet, TextVQA, MVBench and VCGBench. All models are open-sourced in https://github.com/THUDM/CogVLM2 and https://github.com/THUDM/GLM-4, contributing to the advancement of the field.
hub tools
citation-role summary
background 6
citation-polarity summary
roles
background 6representative citing papers
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.
Introduces culture-aware humorous captioning task and staged alignment framework that improves contextual fit and balances image relevance with humor in multimodal LLMs.
Introduces the U-HOI task and shows MLLMs plus a language-to-graph pipeline can handle human-object interactions without any predefined vocabulary at training or inference time.
VSAS-Bench offers temporally dense annotations and synchronous/asynchronous protocols to evaluate streaming VLMs on timeliness, consistency, accuracy, and latency trade-offs, showing that adapted conventional VLMs can outperform specialized streaming models.
Video-Holmes benchmark shows top MLLMs achieve at most 45% accuracy on tasks needing integration of multiple clues from suspense films, unlike existing perception-focused tests.
EgoWalk supplies 50 hours of real-world multimodal human navigation data in varied indoor/outdoor settings together with open pipelines that auto-generate language goal annotations and traversability masks.
AdaMMS merges heterogeneous MLLMs via architecture mapping, linear weight interpolation, and unsupervised hyper-parameter search, outperforming prior methods on vision-language benchmarks as the first such approach without labeled data.
HumanVBench provides a 16-task benchmark for human-centric video understanding in MLLMs, created through automated annotation and distractor synthesis pipelines, and shows top models lag human performance on emotion perception and cross-modal alignment.
S⁴ST shows that dimensionally consistent scaling with low-redundancy complementary transforms achieves state-of-the-art data-free transferable targeted attacks by exploiting visual data's multi-scale nature.
LVBench is a new benchmark for extreme long video understanding that evaluates multimodal large language models on hour-scale videos using tasks designed to probe extended memory and comprehension.
Staged post-training that first solidifies visual perception before visual and textual reasoning improves VLM accuracy and shortens reasoning traces on visual math and perception benchmarks.
CAST reduces object hallucination in LVLMs by 6.03% on average across five models and five benchmarks by identifying caption-sensitive attention heads and applying optimized steering directions to their outputs, with negligible added inference cost.
VersaVogue unifies garment generation and virtual dressing via trait-routing attention with mixture-of-experts and an automated multi-perspective preference optimization pipeline that uses DPO without human labels.
MIRAGE introduces a benchmark for multi-instance image editing and a training-free framework that uses vision-language parsing and parallel regional denoising to achieve precise edits without altering backgrounds.
VideoGPA distills geometry priors via self-supervised DPO to enhance 3D consistency, temporal stability, and motion coherence in video diffusion models.
GLM-4.5V reaches state-of-the-art results on 42 multimodal benchmarks among open-source models of similar size by applying reinforcement learning with curriculum sampling to a strong vision foundation model.
Presents YesBut (V2) benchmark and shows state-of-the-art VLMs significantly underperform humans on tasks requiring comparative reasoning for contradictory humor in comics.
A human preference dataset and VideoReward model enable Flow-DPO and Flow-NRG to produce smoother, better-aligned videos from text prompts in flow-based generators.
MotionBench is a new benchmark showing poor fine-grained motion understanding in VLMs and proposes TE Fusion to improve performance with higher frame rates.
VisionReward learns multi-dimensional human preferences for image and video generation via hierarchical assessment and linear weighting, outperforming VideoScore by 17.2% in prediction accuracy and yielding 31.6% higher win rates in text-to-video models.
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.
CogVideoX generates coherent 10-second text-to-video outputs at high resolution using a 3D VAE, expert adaptive LayerNorm transformer, progressive training, and a custom data pipeline, claiming state-of-the-art results.
MONET is an open 104.9M image-text pair dataset created via safety filtering, deduplication, and multi-VLM recaptioning from 2.9B raw pairs, validated by training a competitive 4B-parameter latent diffusion model.
citing papers explorer
-
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models
Molmo VLMs trained on newly collected PixMo open datasets achieve state-of-the-art performance among open-weight models and surpass multiple proprietary VLMs including Claude 3.5 Sonnet and Gemini 1.5 Pro.
-
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.
-
Culture-Aware Humorous Captioning: Multimodal Humor Generation across Cultural Contexts
Introduces culture-aware humorous captioning task and staged alignment framework that improves contextual fit and balances image relevance with humor in multimodal LLMs.
-
Towards Unconstrained Human-Object Interaction
Introduces the U-HOI task and shows MLLMs plus a language-to-graph pipeline can handle human-object interactions without any predefined vocabulary at training or inference time.
-
VSAS-Bench: Real-Time Evaluation of Visual Streaming Assistant Models
VSAS-Bench offers temporally dense annotations and synchronous/asynchronous protocols to evaluate streaming VLMs on timeliness, consistency, accuracy, and latency trade-offs, showing that adapted conventional VLMs can outperform specialized streaming models.
-
Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?
Video-Holmes benchmark shows top MLLMs achieve at most 45% accuracy on tasks needing integration of multiple clues from suspense films, unlike existing perception-focused tests.
-
EgoWalk: A Multimodal Dataset for Robot Navigation in the Wild
EgoWalk supplies 50 hours of real-world multimodal human navigation data in varied indoor/outdoor settings together with open pipelines that auto-generate language goal annotations and traversability masks.
-
AdaMMS: Model Merging for Heterogeneous Multimodal Large Language Models with Unsupervised Coefficient Optimization
AdaMMS merges heterogeneous MLLMs via architecture mapping, linear weight interpolation, and unsupervised hyper-parameter search, outperforming prior methods on vision-language benchmarks as the first such approach without labeled data.
-
HumanVBench: Probing Human-Centric Video Understanding in MLLMs with Automatically Synthesized Benchmarks
HumanVBench provides a 16-task benchmark for human-centric video understanding in MLLMs, created through automated annotation and distractor synthesis pipelines, and shows top models lag human performance on emotion perception and cross-modal alignment.
-
S$^4$ST: A Strong, Self-transferable, faSt, and Simple Scale Transformation for Transferable Targeted Attack
S⁴ST shows that dimensionally consistent scaling with low-redundancy complementary transforms achieves state-of-the-art data-free transferable targeted attacks by exploiting visual data's multi-scale nature.
-
LVBench: An Extreme Long Video Understanding Benchmark
LVBench is a new benchmark for extreme long video understanding that evaluates multimodal large language models on hour-scale videos using tasks designed to probe extended memory and comprehension.
-
From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
Staged post-training that first solidifies visual perception before visual and textual reasoning improves VLM accuracy and shortens reasoning traces on visual math and perception benchmarks.
-
CAST: Mitigating Object Hallucination in Large Vision-Language Models via Caption-Guided Visual Attention Steering
CAST reduces object hallucination in LVLMs by 6.03% on average across five models and five benchmarks by identifying caption-sensitive attention heads and applying optimized steering directions to their outputs, with negligible added inference cost.
-
VersaVogue: Visual Expert Orchestration and Preference Alignment for Unified Fashion Synthesis
VersaVogue unifies garment generation and virtual dressing via trait-routing attention with mixture-of-experts and an automated multi-perspective preference optimization pipeline that uses DPO without human labels.
-
MIRAGE: Benchmarking and Aligning Multi-Instance Image Editing
MIRAGE introduces a benchmark for multi-instance image editing and a training-free framework that uses vision-language parsing and parallel regional denoising to achieve precise edits without altering backgrounds.
-
VideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generation
VideoGPA distills geometry priors via self-supervised DPO to enhance 3D consistency, temporal stability, and motion coherence in video diffusion models.
-
GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning
GLM-4.5V reaches state-of-the-art results on 42 multimodal benchmarks among open-source models of similar size by applying reinforcement learning with curriculum sampling to a strong vision foundation model.
-
When 'YES' Meets 'BUT': Can Large Models Comprehend Contradictory Humor Through Comparative Reasoning?
Presents YesBut (V2) benchmark and shows state-of-the-art VLMs significantly underperform humans on tasks requiring comparative reasoning for contradictory humor in comics.
-
Improving Video Generation with Human Feedback
A human preference dataset and VideoReward model enable Flow-DPO and Flow-NRG to produce smoother, better-aligned videos from text prompts in flow-based generators.
-
MotionBench: Benchmarking and Improving Fine-grained Video Motion Understanding for Vision Language Models
MotionBench is a new benchmark showing poor fine-grained motion understanding in VLMs and proposes TE Fusion to improve performance with higher frame rates.
-
VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation
VisionReward learns multi-dimensional human preferences for image and video generation via hierarchical assessment and linear weighting, outperforming VideoScore by 17.2% in prediction accuracy and yielding 31.6% higher win rates in text-to-video models.
-
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.
-
CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer
CogVideoX generates coherent 10-second text-to-video outputs at high resolution using a 3D VAE, expert adaptive LayerNorm transformer, progressive training, and a custom data pipeline, claiming state-of-the-art results.
-
MONET: A Massive, Open, Non-redundant and Enriched Text-to-image dataset
MONET is an open 104.9M image-text pair dataset created via safety filtering, deduplication, and multi-VLM recaptioning from 2.9B raw pairs, validated by training a competitive 4B-parameter latent diffusion model.
-
Illusion-Aware Visual Preprocessing and Anti-Illusion Prompting for Classic Illusion Understanding in Vision-Language Models
A combination of illusion-specific image transformations, anti-illusion prompts, and majority voting lets VLMs reach 90.48% accuracy on a 630-image illusion benchmark without any model training.
-
KD-CVG: A Knowledge-Driven Approach for Creative Video Generation
KD-CVG uses an Advertising Creative Knowledge Base plus Semantic-Aware Retrieval and Multimodal Knowledge Reference modules to improve semantic alignment and motion realism in text-to-video generation for advertising.
-
ReCAPA: Hierarchical Predictive Correction to Mitigate Cascading Failures
ReCAPA adds predictive correction and multi-level semantic alignment to VLA models, plus two new metrics for tracking error spread and recovery, yielding competitive benchmark results over LLM baselines.
-
EchoVLM: Dynamic Mixture-of-Experts Vision-Language Model for Universal Ultrasound Intelligence
EchoVLM applies a Mixture-of-Experts vision-language model to ultrasound imaging across seven body regions, reporting gains of 10.15 BLEU-1 and 4.77 ROUGE-1 over Qwen2-VL on report generation.
-
PDF-WuKong: A Large Multimodal Model for Efficient Long PDF Reading with End-to-End Sparse Sampling
PDF-WuKong adds a sparse sampler to an MLLM for efficient long-PDF multimodal QA and reports an 8.6% F1 gain over proprietary models on a new 1.1M-pair academic-paper dataset.
-
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.
- High-Entropy Tokens as Multimodal Failure Points in Vision-Language Models