Trending Papers

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

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.

A large language model adapted for time-series forecasting achieves near-optimal zero-shot performance on diverse datasets across different time scales and granularities.

Video-MME-v2 presents a comprehensive benchmark for evaluating video understanding models through a progressive hierarchy and group-based evaluation to assess robustness and faithfulness.

VOID is a video object removal framework that uses vision-language models and video diffusion models to generate physically plausible scenes by leveraging causal reasoning and counterfactual reasoning.

TriAttention addresses KV cache memory bottlenecks in LLMs by leveraging Q/K vector concentration in pre-RoPE space to improve key importance estimation and enable efficient long-context generation.

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.

Bitnet.cpp enhances edge inference for ternary LLMs using a novel mixed-precision matrix multiplication library, achieving significant speed improvements over baselines.

DeepScientist autonomously conducts scientific discovery through Bayesian Optimization, surpassing human state-of-the-art methods on multiple AI tasks.

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

MegaTrain enables efficient training of large language models with over 100 billion parameters on a single GPU by utilizing host memory storage and optimized data streaming techniques.

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.

The AI Scientist-v2 autonomously proposes hypotheses, performs experiments, analyzes data, and writes peer-reviewed scientific papers, marking the first fully AI-generated paper accepted by a conference.

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

A large-scale dataset collection, The Well, provides diverse numerical simulations for benchmarking machine learning models in physical systems simulation.

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.

OpenWorldLib presents a standardized framework for advanced world models that integrate perception, interaction, and long-term memory capabilities for comprehensive world understanding and prediction.

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.

A large-scale dynamic dataset derived from AAA games is introduced to improve generative inverse and forward rendering, featuring high-resolution synchronized RGB and G-buffer data alongside a novel VLM-based evaluation method that correlates well with human judgment.

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.

GLM-5 advances foundation models with DSA for cost reduction, asynchronous reinforcement learning for improved alignment, and enhanced coding capabilities for real-world software engineering.

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.

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.

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.

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.

Memory Sparse Attention (MSA) enables large language models to process extremely long contexts with linear complexity and high efficiency through innovations like sparse attention and document-wise RoPE.

EvoScientist is an adaptive multi-agent framework that enhances scientific discovery by continuously learning from past interactions through persistent memory modules.

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.

Simple self-distillation improves code generation in large language models by fine-tuning on model-generated samples, effectively addressing precision-exploration trade-offs in decoding.

Autonomous multi-agent evolution framework enables open-ended discovery through persistent memory, asynchronous execution, and collaborative problem-solving, achieving superior performance on mathematical and optimization tasks.

CutClaw is an autonomous multi-agent framework that uses multimodal language models to automatically edit long video footage into rhythmic, narratively consistent short videos with synchronized audio and visual elements.

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.

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.

MetaChain, a fully-automated natural language-based framework, enables non-technical users to create and deploy LLM agents efficiently, demonstrating superior performance on multi-agent tasks and retrieval-augmented generation.

Memory Intelligence Agent framework integrates non-parametric and parametric memory systems with reinforcement learning to enable efficient reasoning and autonomous evolution in open-world environments.

A suite of open-source music foundation models is introduced, featuring components for audio-text alignment, lyric recognition, music coding, and large language model-based song generation with controllable attributes and scalable parameterization.

DFlash is a speculative decoding framework that uses a lightweight block diffusion model for parallel token drafting, achieving significant speedup over existing autoregressive methods while maintaining high-quality outputs.

daVinci-MagiHuman is an open-source audio-video generative model that synchronizes text, video, and audio through a single-stream Transformer architecture, achieving high-quality human-centric content generation with efficient inference capabilities.

Researchers introduce PTE (Prefill Token Equivalents), a hardware-aware metric for measuring efficiency in Tool-Integrated Reasoning scenarios, which better correlates with actual inference latency than traditional token counts by accounting for KV-Cache inefficiencies and long tool responses.

Latent space is emerging as a fundamental computational substrate for language-based models, offering advantages over explicit token-level approaches through continuous representation that mitigates linguistic redundancy and sequential inefficiency.

AutoFigure-Edit is an end-to-end system that generates editable scientific illustrations from text descriptions and reference images, supporting flexible style adaptation and efficient refinement.

LeWorldModel presents a stable end-to-end JEPA framework that trains efficiently from raw pixels using minimal loss terms while maintaining competitive performance in control tasks and encoding meaningful physical structures.

AURA is an end-to-end streaming visual interaction framework that enables continuous video stream processing with real-time question answering and proactive responses through integrated context management and optimized deployment.

Efficient Universal Perception Encoder (EUPE) improves edge device performance by distilling knowledge from multiple vision encoders through a two-stage scaling approach, achieving superior representation quality compared to previous methods.

GigaWorld-Policy introduces an action-centered World-Action Model that improves robotic policy learning by decoupling visual and motion representations, enabling faster inference and better task performance through dual supervision from action prediction and video generation.

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.

In-Place Test-Time Training enables large language models to adapt parameters during inference by modifying the final projection matrix in MLP blocks with a task-aligned objective and efficient update mechanism.

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.

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.