ATLAS uses a single functional token to unify agentic and latent visual reasoning without image generation or external execution.
hub Canonical reference
Unified multimodal understanding and generation models: Advances, challenges, and opportunities.arXiv preprint arXiv:2505.02567
Canonical reference. 100% of citing Pith papers cite this work as background.
19
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 8
citation-polarity summary
roles
background 7polarities
background 7representative citing papers
UniPath adaptively models coordination-path diversity in unified multimodal models by training a path-conditioned executor and using a lightweight planner for input-dependent selection, improving performance over fixed strategies.
XTC-Bench reveals that strong performance on generation or understanding tasks in unified multimodal models does not guarantee cross-task semantic consistency, which instead depends on how tightly coupled the learning objectives are across modalities.
Large multimodal models display emerging but limited spatial action capabilities in goal-oriented urban 3D navigation, remaining far from human-level performance with errors diverging rapidly after critical decision points.
Lance presents a dual-stream mixture-of-experts model with modality-aware positional encoding and staged multi-task training that outperforms prior open-source unified models on image and video generation while keeping strong understanding performance.
LatentUMM proposes dual latent alignment at modality and capacity levels plus latent dynamics stabilization to reduce semantic drift and improve consistency in unified multimodal models.
PresentAgent-2 generates query-driven multimodal presentation videos with research grounding, supporting single-speaker, multi-speaker discussion, and interactive question-answering modes.
TorchUMM is the first unified codebase and benchmark suite for multimodal understanding, generation, and editing across varied UMM models and datasets.
One Tokenizer achieves zero-gap multimodal integration by mapping all inputs to a unified token vocabulary, allowing native LLMs to perform deep cross-modal reasoning without modular encoders or fusion layers, and outperforming encoder-based baselines on DNA-text tasks.
Mull-Tokens are modality-agnostic latent tokens that enable free-form multimodal thinking and deliver up to 16% gains on spatial reasoning benchmarks.
Speculative Coupled Decoding stabilizes draft sampling in Speculative Jacobi Decoding via an information-theoretic coupling step, delivering up to 4.2x image and 13.6x video speedups with no quality loss or training.
PaintBench provides a scalable deterministic benchmark for precise visual editing operations, revealing that even the best of 11 models achieves only 17.1% mIoU and that scores correlate strongly with applied data visualization editing performance.
UniGenDet unifies generative and discriminative models through symbiotic self-attention and detector-guided alignment to co-evolve image generation and authenticity detection.
UniRect-CoT is a training-free rectification chain-of-thought framework that treats diffusion denoising as visual reasoning and uses the model's inherent understanding to align and correct intermediate generation results.
OmniJigsaw is a self-supervised proxy task that reconstructs shuffled audio-visual clips via joint integration, sample-level selection, and clip-level masking strategies, yielding gains on 15 video, audio, and reasoning benchmarks.
A multi-dataset cross-domain knowledge distillation approach improves unified performance on medical image segmentation, classification, and detection by transferring domain-invariant features from a joint teacher model to task-specific students.
I2P adaptively selects the most discriminative layers from visual foundation models for synthetic image detection and constrains task updates to low-sensitivity parameter subspaces to improve specificity without harming generalization.
citing papers explorer
-
ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
ATLAS uses a single functional token to unify agentic and latent visual reasoning without image generation or external execution.
-
UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning
UniPath adaptively models coordination-path diversity in unified multimodal models by training a path-conditioned executor and using a lightweight planner for input-dependent selection, improving performance over fixed strategies.
-
Beyond Accuracy: Benchmarking Cross-Task Consistency in Unified Multimodal Models
XTC-Bench reveals that strong performance on generation or understanding tasks in unified multimodal models does not guarantee cross-task semantic consistency, which instead depends on how tightly coupled the learning objectives are across modalities.
-
How Far Are Large Multimodal Models from Human-Level Spatial Action? A Benchmark for Goal-Oriented Embodied Navigation in Urban Airspace
Large multimodal models display emerging but limited spatial action capabilities in goal-oriented urban 3D navigation, remaining far from human-level performance with errors diverging rapidly after critical decision points.
-
Lance: Unified Multimodal Modeling by Multi-Task Synergy
Lance presents a dual-stream mixture-of-experts model with modality-aware positional encoding and staged multi-task training that outperforms prior open-source unified models on image and video generation while keeping strong understanding performance.
-
LatentUMM: Dual Latent Alignment for Unified Multimodal Models
LatentUMM proposes dual latent alignment at modality and capacity levels plus latent dynamics stabilization to reduce semantic drift and improve consistency in unified multimodal models.
-
PresentAgent-2: Towards Generalist Multimodal Presentation Agents
PresentAgent-2 generates query-driven multimodal presentation videos with research grounding, supporting single-speaker, multi-speaker discussion, and interactive question-answering modes.
-
TorchUMM: A Unified Multimodal Model Codebase for Evaluation, Analysis, and Post-training
TorchUMM is the first unified codebase and benchmark suite for multimodal understanding, generation, and editing across varied UMM models and datasets.
-
Mind the Gap No More: Achieving Zero-Gap Multimodal Integration via One Tokenizer
One Tokenizer achieves zero-gap multimodal integration by mapping all inputs to a unified token vocabulary, allowing native LLMs to perform deep cross-modal reasoning without modular encoders or fusion layers, and outperforming encoder-based baselines on DNA-text tasks.
-
Mull-Tokens: Modality-Agnostic Latent Thinking
Mull-Tokens are modality-agnostic latent tokens that enable free-form multimodal thinking and deliver up to 16% gains on spatial reasoning benchmarks.
-
Speculative Coupled Decoding for Training-Free Lossless Acceleration of Autoregressive Visual Generation
Speculative Coupled Decoding stabilizes draft sampling in Speculative Jacobi Decoding via an information-theoretic coupling step, delivering up to 4.2x image and 13.6x video speedups with no quality loss or training.
-
PaintBench: Deterministic Evaluation of Precise Visual Editing
PaintBench provides a scalable deterministic benchmark for precise visual editing operations, revealing that even the best of 11 models achieves only 17.1% mIoU and that scores correlate strongly with applied data visualization editing performance.
-
UniGenDet: A Unified Generative-Discriminative Framework for Co-Evolutionary Image Generation and Generated Image Detection
UniGenDet unifies generative and discriminative models through symbiotic self-attention and detector-guided alignment to co-evolve image generation and authenticity detection.
-
Free Lunch for Unified Multimodal Models: Enhancing Generation via Reflective Rectification with Inherent Understanding
UniRect-CoT is a training-free rectification chain-of-thought framework that treats diffusion denoising as visual reasoning and uses the model's inherent understanding to align and correct intermediate generation results.
-
OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering
OmniJigsaw is a self-supervised proxy task that reconstructs shuffled audio-visual clips via joint integration, sample-level selection, and clip-level masking strategies, yielding gains on 15 video, audio, and reasoning benchmarks.
-
Multi-Dataset Cross-Domain Knowledge Distillation for Unified Medical Image Segmentation, Classification, and Detection
A multi-dataset cross-domain knowledge distillation approach improves unified performance on medical image segmentation, classification, and detection by transferring domain-invariant features from a joint teacher model to task-specific students.
-
Adaptive Forensic Feature Refinement via Intrinsic Importance Perception
I2P adaptively selects the most discriminative layers from visual foundation models for synthetic image detection and constrains task updates to low-sensitivity parameter subspaces to improve specificity without harming generalization.
- GRPO-TTA: Test-Time Visual Tuning for Vision-Language Models via GRPO-Driven Reinforcement Learning
- OpenWorldLib: A Unified Codebase and Definition of Advanced World Models