Trending Papers

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.

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.

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

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.

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

A framework called Visual Spatial Tuning (VST) enhances the spatial abilities of Vision-Language Models (VLMs) through progressive training with specialized datasets, achieving state-of-the-art results on spatial benchmarks.

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.

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.

The "Thinking with Video" paradigm enhances multimodal reasoning by integrating video generation models, demonstrated through the Video Thinking Benchmark and improved performance on both vision and text tasks.

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.

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.

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.

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.

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.

Diffusion language models outperform autoregressive models in low-data settings due to any-order modeling, iterative bidirectional denoising, and Monte Carlo augmentation, and maintain advantages even at scale.

Continuous Autoregressive Language Models (CALM) improve language model efficiency by predicting continuous vectors instead of discrete tokens, reducing computational cost while maintaining performance.

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.

AgentScope enhances agentic applications by providing flexible tool-based interactions, unified interfaces, and advanced infrastructure based on the ReAct paradigm, supporting efficient and safe development and deployment.

Reinforcement Learning with Adaptive Verifiable Environments (RLVE) improves language model reasoning by dynamically adjusting problem difficulty, outperforming static environments and traditional RL training.

ChronoEdit addresses physical consistency in image editing by reframing it as a video generation problem, leveraging pretrained video models and temporal reasoning tokens.

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

AgentFounder, a deep research agent model incorporating Agentic Continual Pre-training, achieves state-of-the-art performance in agentic tasks while maintaining strong tool-use ability.

PhysWorld integrates video generation and physical world modeling to enable accurate robotic manipulation from visual demonstrations without real robot data.

WebWatcher, a multimodal agent with enhanced visual-language reasoning, outperforms existing agents in complex visual and textual information retrieval tasks using synthetic trajectories and reinforcement learning.

WebShaper, a formalization-driven framework, synthesizes information-seeking datasets using set theory and Knowledge Projections to enhance reasoning structure and achieve top performance in open-sourced benchmarks.

The paper proposes a framework for building end-to-end agentic information seeking agents through a combination of data construction, trajectory sampling, supervised fine-tuning, and reinforcement learning, showcasing its effectiveness on information seeking benchmarks.

WebWeaver, a dual-agent framework, addresses open-ended deep research challenges by integrating adaptive planning and focused synthesis to produce high-quality, reliable reports.

ReSum, a novel paradigm with periodic context summarization, enhances web agents' performance on knowledge-intensive tasks by overcoming context window limitations, achieving significant improvements over ReAct.

WebSailor, a post-training methodology, enhances open-source models with systematic uncertainty reduction, matching proprietary agents' performance in complex information-seeking tasks.

WebSailor, a post-training methodology, enhances open-source LLMs with sophisticated reasoning to match proprietary systems in complex information-seeking tasks.

A generalist agent architecture combining multi-agent planning, hierarchical memory, and a refined tool suite outperforms existing systems in diverse tasks.

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.

Converting pretrained non-recurrent language models to depth-recurrent models improves performance at a given compute budget using a curriculum of recurrences.

Kimi Linear, a hybrid linear attention architecture, outperforms full attention in various scenarios with improved efficiency and performance, using Kimi Delta Attention and Multi-Head Latent Attention.

EdgeTAM optimizes SAM 2 for mobile devices by introducing a 2D Spatial Perceiver and a distillation pipeline, improving video segmentation performance while maintaining high frame rates.

RLinf-VLA is a unified framework for scalable reinforcement learning training of vision-language-action models, offering improved performance and generalization compared to supervised fine-tuning.

Behavior Best-of-N (bBoN) improves the reliability and success rates of computer-use agents by generating and selecting among multiple rollouts using behavior narratives, achieving state-of-the-art performance on OSWorld and strong generalization to different operating systems.

Agent S2, a compositional framework using Mixture-of-Grounding and Proactive Hierarchical Planning, achieves state-of-the-art performance in computer use automation across various benchmarks and operating systems.

V-Thinker, a multimodal reasoning assistant using reinforcement learning, enhances image-interactive thinking by synthesizing datasets and aligning perception for improved performance in vision-centric tasks.

Ovi is a unified audio-video generation model using twin-DiT modules with blockwise cross-modal fusion, enabling natural synchronization and high-quality multimodal outputs.

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.

olmOCR is an open-source toolkit using a fine-tuned vision language model to process PDFs into clean text while preserving structure, optimized for large-scale batch processing.

A framework called Easy Dataset synthesizes fine-tuning data from unstructured documents using a GUI and LLMs, improving domain-specific performance of LLMs while maintaining general knowledge.

DeepEyesV2, an agentic multimodal model, uses a two-stage training pipeline to effectively integrate tool use, demonstrating robust performance across real-world reasoning tasks.

GroundCUA, a large-scale desktop grounding dataset, enables the development of GroundNext models that achieve state-of-the-art performance in mapping instructions to UI elements with less training data.

A method using 3D Gaussians for scene representation and optimized rendering allows high-quality, real-time novel-view synthesis at 1080p resolution.

Emu3.5, a large-scale multimodal world model, predicts next states in vision and language, enhanced with reinforcement learning and Discrete Diffusion Adaptation for efficient inference, achieving strong performance in various multimodal tasks.