Trending Papers

Z-Image, a 6B-parameter Scalable Single-Stream Diffusion Transformer (S3-DiT) model, achieves high-performance image generation with reduced computational cost, offering sub-second inference and compatibility with consumer hardware.

The study reveals that in text-to-image generation, CFG Augmentation is the primary driver of few-step distillation in Distribution Matching Distillation (DMD), while the distribution matching term acts as a regularizer.

WizardCoder, a Code LLM fine-tuned with complex instructions using Evol-Instruct, outperforms other open-source and closed LLMs on several code generation benchmarks.

Segment Anything Model 3 achieves state-of-the-art performance in promptable concept segmentation and tracking by leveraging a unified model architecture with decoupled recognition and localization.

LightRAG improves Retrieval-Augmented Generation by integrating graph structures for enhanced contextual awareness and efficient information retrieval, achieving better accuracy and response times.

HunyuanOCR, a lightweight Vision-Language Model, achieves state-of-the-art performance in OCR tasks through a unified end-to-end architecture combining Vision Transformer and lightweight LLM, supported by data-driven and RL strategies.

GAM, a novel framework that employs JIT compilation principles, improves memory efficiency and task completion by leveraging a lightweight memorizer and researcher in conjunction with reinforcement learning.

SAM 3D is a generative model that reconstructs 3D objects from single images using a multi-stage training framework that includes synthetic pretraining and real-world alignment, achieving high performance in human preference tests.

PaddleOCR-VL, a vision-language model combining NaViT-style visual encoder and ERNIE-4.5 language model, achieves state-of-the-art performance in document parsing with minimal resource consumption.

Flawed-Aware Policy Optimization (FAPO) enhances reinforcement learning with verifiable rewards by penalizing flawed-positive rollouts, improving reasoning capability and training stability in large language models.

PPTAgent, a two-stage approach, improves presentation generation by analyzing reference presentations and ensuring structural and content consistency, outperforming traditional methods across content, design, and coherence.

MobiAgent, a comprehensive mobile agent system, achieves state-of-the-art performance in real-world mobile scenarios through its MobiMind-series models, AgentRR framework, and MobiFlow benchmarking suite, while also reducing data annotation costs.

Agent0-VL, a self-evolving vision-language agent, incorporates tool usage into both reasoning and self-evaluation, enabling continual improvement through evidence-grounded analysis and reinforcement learning.

Agent0, a self-evolving framework utilizing multi-step co-evolution and tool integration, enhances LLM reasoning capabilities without human-curated data.

Vidi, a family of Large Multimodal Models, excels in temporal retrieval for video editing by processing long, multimodal video content and outperforming proprietary models on the VUE-TR benchmark.

The proposed AnyTalker framework generates high-quality multi-person talking videos by extending Diffusion Transformer with identity-aware attention, leveraging single-person videos for training, and using a specialized dataset for evaluation.

GigaWorld-0 is a unified world model framework that integrates video generation and 3D modeling to produce high-quality, diverse, and physically plausible VLA data, enabling strong real-world performance in embodied AI without real-world training.

LatentMAS enables efficient and effective collaboration among LLM agents using latent space representations, enhancing reasoning quality and reducing computational costs.

MinerU is an open-source tool that enhances document content extraction using fine-tuned models and pre/postprocessing rules across diverse document types.

LlamaFactory is a unified framework enabling efficient fine-tuning of large language models across various tasks using a web-based user interface.

A novel GPT-based model, OmniFlatten, enables real-time natural full-duplex spoken dialogue through a multi-stage post-training technique that integrates speech and text without altering the original model's architecture.

Agent Lightning is a flexible RL framework for training LLMs in various agents, using a hierarchical RL algorithm and decoupling execution from training to handle complex interactions.

MinerU2.5, a 1.2B-parameter document parsing vision-language model, achieves state-of-the-art recognition accuracy with computational efficiency through a coarse-to-fine parsing strategy.

Depth Anything 3 (DA3) uses a plain transformer for geometry prediction from visual inputs, achieving state-of-the-art results in camera pose estimation, any-view geometry, visual rendering, and monocular depth estimation.

ORS3D, a new task requiring language understanding, 3D grounding, and efficient scheduling, is introduced with a large dataset and an embodied multi-modal model named GRANT that uses a scheduling token mechanism for effective task management.

DeepAnalyze-8B, an agentic LLM, autonomously completes the data science pipeline from raw data to research reports using curriculum-based training and data-grounded trajectory synthesis.

A multi-agent framework using large language models for stock trading simulates real-world trading firms, improving performance metrics like cumulative returns and Sharpe ratio.

IndexTTS, an enhanced text-to-speech system combining XTTS and Tortoise models, offers improved naturalness, enhanced voice cloning, and controllable usage through hybrid character-pinyin modeling and optimized vector quantization.

Bridge Models, instantiated as Vision Bridge Transformer (ViBT), efficiently translate data through direct modeling of input-to-output trajectories, achieving robust performance in image and video editing tasks at large scales.

Zep, a memory layer service, outperforms MemGPT in the DMR benchmark and LongMemEval by excelling in dynamic knowledge integration and temporal reasoning, critical for enterprise use cases.

MIRIX, a modular multi-agent memory system, enhances language models' memory capabilities by integrating diverse memory types and a dynamic framework, achieving superior performance in multimodal and long-form conversation benchmarks.

Mem0, a memory-centric architecture with graph-based memory, enhances long-term conversational coherence in LLMs by efficiently extracting, consolidating, and retrieving information, outperforming existing memory systems in terms of accuracy and computational efficiency.

PaperTalker is a multi-agent framework that automates academic presentation video generation by integrating slide generation, layout refinement, subtitling, speech synthesis, and talking-head rendering, outperforming existing methods.

A simple head-specific sigmoid gate applied after Scaled Dot-Product Attention improves performance, stability, and scaling in large models, mitigating 'attention sink' and enhancing long-context extrapolation.

Today's denoising diffusion models do not "denoise" in the classical sense, i.e., they do not directly predict clean images. Rather, the neural networks predict noise or a noised quantity. In this paper, we suggest that predicting clean data and predicting noised quantities are fundamentally different. According to the manifold assumption, natural data should lie on a low-dimensional manifold, whereas noised quantities do not. With this assumption, we advocate for models that directly predict clean data, which allows apparently under-capacity networks to operate effectively in very high-dimensional spaces. We show that simple, large-patch Transformers on pixels can be strong generative models: using no tokenizer, no pre-training, and no extra loss. Our approach is conceptually nothing more than "Just image Transformers", or JiT, as we call it. We report competitive results using JiT with large patch sizes of 16 and 32 on ImageNet at resolutions of 256 and 512, where predicting high-dimensional noised quantities can fail catastrophically. With our networks mapping back to the basics of the manifold, our research goes back to basics and pursues a self-contained paradigm for Transformer-based diffusion on raw natural data.

Code2Video generates educational videos using a code-centric agent framework, improving coherence and interpretability compared to direct code generation.

InternVL3 is a multimodal pre-trained language model that jointly learns from both multimodal data and text, improving performance and scalability through advanced techniques and setting a new state-of-the-art in multimodal tasks.

LucidFlux, a caption-free UIR framework using a diffusion transformer, achieves robust image restoration through adaptive conditioning and SigLIP features without text prompts.

SteadyDancer, an Image-to-Video framework, ensures first-frame identity preservation and precise motion control through harmonized conditions, adaptive pose representation, and hierarchical training objectives.

G$^2$VLM integrates 3D geometry learning with vision-language models to enhance spatial understanding and reasoning, outperforming existing models in these tasks.

iMontage repurposes pre-trained video models to generate high-quality, diverse image sets with natural transitions and enhanced dynamics through a unified framework and tailored adaptation strategy.

GigaBrain-0, a VLA foundation model, uses world model-generated data to enhance cross-task generalization and policy robustness, improving real-world performance on complex manipulation tasks.

The PyTorch distributed data parallel module optimizes large-scale model training using techniques like gradient bucketing, computation-communication overlap, and selective synchronization to achieve near-linear scalability.

PyTorch Fully Sharded Data Parallel (FSDP) enables efficient and scalable training of large models across hardware configurations.

The proposed Attention Interaction Alignment (AIA) loss improves cross-modal attention and performance in unified multimodal models for image generation and understanding without decoupling.

Step-Audio-R1, using the Modality-Grounded Reasoning Distillation framework, achieves strong reasoning capabilities in audio, outperforming previous models and demonstrating the transferability of reasoning across modalities.

RAG-Anything is a unified framework that enhances multimodal knowledge retrieval by integrating cross-modal relationships and semantic matching, outperforming existing methods on complex benchmarks.

A unified Vision-Language-Action (VLA) and world model, RynnVLA-002, jointly learns environmental dynamics and action planning, outperforming individual models in both simulation and real-world tasks.

DINOv3, a self-supervised learning model, achieves superior performance across various vision tasks by scaling datasets and models, addressing dense feature degradation, and enhancing flexibility with post-hoc strategies.

GeoVista, an agentic model integrating tool invocation and reinforcement learning, achieves high geolocalization performance on GeoBench, outperforming open-source models and matching closed-source models.