Trending Papers

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.

SANA-Video, a small diffusion model, efficiently generates high-resolution, high-quality videos with strong text-video alignment using linear attention and a constant-memory KV cache, achieving competitive performance at a lower cost and faster speed.

Lance is a unified multimodal model that combines understanding, generation, and editing capabilities for images and videos through collaborative multi-task training and a dual-stream architecture.

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.

AI-Trader presents the first fully automated live benchmark for evaluating large language models in financial decision-making across multiple markets with autonomous information processing.

ARIS is an open-source research harness that uses cross-model adversarial collaboration to ensure reliable long-term research outcomes through coordinated execution, orchestration, and assurance layers.

TideGS enables training 3D Gaussian Splatting with over one billion primitives on a single GPU by managing parameters across SSD-CPU-GPU hierarchy through block-virtualization, asynchronous pipeline, and differential streaming techniques.

OpenDevin is a platform for developing AI agents that interact with the world by writing code, using command lines, and browsing the web, with support for multiple agents and evaluation benchmarks.

EverMemOS presents a self-organizing memory system for large language models that processes dialogue streams into structured memory cells and scenes to enhance long-term interaction capabilities.

LongLive-2.0 presents an NVFP4-based parallel infrastructure for long video generation that addresses training and inference bottlenecks through sequence-parallel autoregressive training and diffusion model tuning.

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.

Pixal3D introduces a pixel-aligned 3D generation approach that addresses fidelity issues in 3D asset creation by establishing direct pixel-to-3D correspondences through back-projection conditioning.

Kronos, a specialized pre-training framework for financial K-line data, outperforms existing models in forecasting and synthetic data generation through a unique tokenizer and autoregressive pre-training on a large dataset.

PagedAttention algorithm and vLLM system enhance the throughput of large language models by efficiently managing memory and reducing waste in the key-value cache.

GenericAgent is a self-evolving large language model agent system that maximizes context information density through hierarchical memory, reusable SOPs, and efficient compression to overcome long-horizon limitations.

Multimodal procedural knowledge frameworks enable visual agents to leverage external reusable skills through structured representations combining text, state cards, and visual keyframes, improving decision-making in complex environments.

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.

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.

PaddleOCR-VL, a vision-language model combining NaViT-style dynamic resolution and ERNIE, achieves state-of-the-art performance in document parsing and element recognition with high efficiency.

Unified vision-language models treat understanding and generation as integrated processes rather than separate tasks, demonstrating strong performance across multiple multimodal capabilities including image synthesis and action reasoning.

Enhancements to the AgentScope platform improve scalability, efficiency, and ease of use for large-scale multi-agent simulations through distributed mechanisms, flexible environments, and user-friendly tools.

EnvFactory automates the creation of executable tool environments and natural multi-turn trajectories for training LLMs with agentic reinforcement learning, achieving superior performance with fewer resources.

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.

SmolDocling is a compact vision-language model that performs end-to-end document conversion with robust performance across various document types using 256M parameters and a new markup format.

AI systems demonstrate varying reliability across research stages, excelling in structured tasks but struggling with novel ideas and scientific judgment, necessitating human oversight for credible outcomes.

Adaptive chunking framework uses intrinsic document metrics to select optimal segmentation strategies for retrieval-augmented generation, significantly improving answer correctness and question resolution rates.

MiniCPM-V 4.5, a 8B parameter multimodal large language model, achieves high performance and efficiency through a unified 3D-Resampler architecture, a unified learning paradigm, and a hybrid reinforcement learning strategy.

MiniCPM-o 4.5 enables real-time full-duplex multimodal interaction through Omni-Flow, a unified streaming framework that aligns inputs and outputs temporally for simultaneous perception and response.

World Action Models unify predictive state modeling with action generation for embodied policy learning, forming a cohesive framework for understanding environment dynamics and action prediction.

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.

FashionChameleon enables real-time interactive multi-garment video customization through teacher-student distillation and in-context learning techniques while maintaining motion coherence.

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.

AstraFlow is a dataflow-oriented reinforcement learning system that enables efficient multi-policy collaborative training and elastic scaling across diverse compute resources for large language model agents.

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

Visual typesetting optimization addresses the gap between compilable LaTeX documents and publication-ready PDFs through vision-in-the-loop agents that iteratively diagnose and repair layout defects.

Generative post-training with semantic segmentation as a proxy enhances multimodal alignment and performance in unified models.

DataFlow is an LLM-driven data preparation framework that enhances data quality and reproducibility for various tasks, improving LLM performance with automatically generated pipelines.

AnyFlow introduces a novel any-step video diffusion distillation framework that improves upon consistency distillation by optimizing full ODE sampling trajectories through flow-map transition learning and backward simulation techniques.

AutoDev is an AI-driven software development framework that automates complex engineering tasks within a secure Docker environment, achieving high performance in code and test generation.

VibeVoice synthesizes long-form multi-speaker speech using next-token diffusion and a highly efficient continuous speech tokenizer, achieving superior performance and fidelity.

GoLongRL presents an open-source approach for long-context reinforcement learning with diverse reward optimization through capability-oriented data construction and TMN-Reweight methodology.

Agentic coding tools receive goals written in natural language as input, break them down into specific tasks, and write or execute the actual code with minimal human intervention. Central to this process are agent context files ("READMEs for agents") that provide persistent, project-level instructions. In this paper, we conduct the first large-scale empirical study of 2,303 agent context files from 1,925 repositories to characterize their structure, maintenance, and content. We find that these files are not static documentation but complex, difficult-to-read artifacts that evolve like configuration code, maintained through frequent, small additions. Our content analysis of 16 instruction types shows that developers prioritize functional context, such as build and run commands (62.3%), implementation details (69.9%), and architecture (67.7%). We also identify a significant gap: non-functional requirements like security (14.5%) and performance (14.5%) are rarely specified. These findings indicate that while developers use context files to make agents functional, they provide few guardrails to ensure that agent-written code is secure or performant, highlighting the need for improved tooling and practices.

A novel MLLM-based agentic framework called Code-as-Room generates 3D indoor rooms by converting top-down images into executable Blender code through a structured execution harness with cross-stage memory to maintain context.

LingBot-Map is a feed-forward 3D foundation model that reconstructs scenes from video streams using a geometric context transformer architecture with specialized attention mechanisms for coordinate grounding, dense geometric cues, and long-range drift correction, achieving stable real-time performance at 20 FPS.

Large language models are increasingly used as operational substrates for agent reasoning and execution in agentic systems, with code serving as a unified infrastructure layer across multiple domains and applications.

LLM agents executing within execution harnesses can produce correct outputs while violating safety constraints during execution, necessitating trajectory-level auditing to ensure proper resource access and information flow across multi-agent systems.

Orthrus is a dual-architecture framework that combines autoregressive LLMs with diffusion models to achieve fast parallel token generation while maintaining exact inference fidelity through shared KV caches and consensus mechanisms.

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

HY-World 2.0 is a multi-modal world model framework that generates high-fidelity 3D Gaussian Splatting scenes from diverse inputs using specialized modules for panorama generation, trajectory planning, world expansion, and composition, along with an enhanced rendering platform for interactive 3D exploration.

A self-supervised framework optimizes prompts for both closed and open-ended tasks by evaluating LLM outputs without external references, reducing costs and required data.