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.

Residual connections with PreNorm are standard in modern LLMs, yet they accumulate all layer outputs with fixed unit weights. This uniform aggregation causes uncontrolled hidden-state growth with depth, progressively diluting each layer's contribution. We propose Attention Residuals (AttnRes), which replaces this fixed accumulation with softmax attention over preceding layer outputs, allowing each layer to selectively aggregate earlier representations with learned, input-dependent weights. To address the memory and communication overhead of attending over all preceding layer outputs for large-scale model training, we introduce Block AttnRes, which partitions layers into blocks and attends over block-level representations, reducing the memory footprint while preserving most of the gains of full AttnRes. Combined with cache-based pipeline communication and a two-phase computation strategy, Block AttnRes becomes a practical drop-in replacement for standard residual connections with minimal overhead. Scaling law experiments confirm that the improvement is consistent across model sizes, and ablations validate the benefit of content-dependent depth-wise selection. We further integrate AttnRes into the Kimi Linear architecture (48B total / 3B activated parameters) and pre-train on 1.4T tokens, where AttnRes mitigates PreNorm dilution, yielding more uniform output magnitudes and gradient distribution across depth, and improves downstream performance across all evaluated tasks.

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

Fish Audio S2 is an open-source text-to-speech system with multi-speaker capabilities, multi-turn generation, and instruction-following control through natural-language descriptions, utilizing a multi-stage training approach and production-ready inference engine.

A continual meta-learning framework for large language model agents that jointly evolves policies and reusable behavioral skills while minimizing downtime through opportunistic updates and skill-driven adaptation.

OpenClaw-RL framework enables policy learning from diverse next-state signals across multiple interaction modalities using asynchronous training with PRM judges and hindsight-guided distillation.

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

Great scientists have strong judgement and foresight, closely tied to what we call scientific taste. Here, we use the term to refer to the capacity to judge and propose research ideas with high potential impact. However, most relative research focuses on improving an AI scientist's executive capability, while enhancing an AI's scientific taste remains underexplored. In this work, we propose Reinforcement Learning from Community Feedback (RLCF), a training paradigm that uses large-scale community signals as supervision, and formulate scientific taste learning as a preference modeling and alignment problem. For preference modeling, we train Scientific Judge on 700K field- and time-matched pairs of high- vs. low-citation papers to judge ideas. For preference alignment, using Scientific Judge as a reward model, we train a policy model, Scientific Thinker, to propose research ideas with high potential impact. Experiments show Scientific Judge outperforms SOTA LLMs (e.g., GPT-5.2, Gemini 3 Pro) and generalizes to future-year test, unseen fields, and peer-review preference. Furthermore, Scientific Thinker proposes research ideas with higher potential impact than baselines. Our findings show that AI can learn scientific taste, marking a key step toward reaching human-level AI scientists.

We present MiroThinker v1.0, an open-source research agent designed to advance tool-augmented reasoning and information-seeking capabilities. Unlike previous agents that only scale up model size or context length, MiroThinker explores interaction scaling at the model level, systematically training the model to handle deeper and more frequent agent-environment interactions as a third dimension of performance improvement. Unlike LLM test-time scaling, which operates in isolation and risks degradation with longer reasoning chains, interactive scaling leverages environment feedback and external information acquisition to correct errors and refine trajectories. Through reinforcement learning, the model achieves efficient interaction scaling: with a 256K context window, it can perform up to 600 tool calls per task, enabling sustained multi-turn reasoning and complex real-world research workflows. Across four representative benchmarks-GAIA, HLE, BrowseComp, and BrowseComp-ZH-the 72B variant achieves up to 81.9%, 37.7%, 47.1%, and 55.6% accuracy respectively, surpassing previous open-source agents and approaching commercial counterparts such as GPT-5-high. Our analysis reveals that MiroThinker benefits from interactive scaling consistently: research performance improves predictably as the model engages in deeper and more frequent agent-environment interactions, demonstrating that interaction depth exhibits scaling behaviors analogous to model size and context length. These findings establish interaction scaling as a third critical dimension for building next-generation open research agents, complementing model capacity and context windows.

What if a world simulation model could render not an imagined environment but a city that actually exists? Prior generative world models synthesize visually plausible yet artificial environments by imagining all content. We present Seoul World Model (SWM), a city-scale world model grounded in the real city of Seoul. SWM anchors autoregressive video generation through retrieval-augmented conditioning on nearby street-view images. However, this design introduces several challenges, including temporal misalignment between retrieved references and the dynamic target scene, limited trajectory diversity and data sparsity from vehicle-mounted captures at sparse intervals. We address these challenges through cross-temporal pairing, a large-scale synthetic dataset enabling diverse camera trajectories, and a view interpolation pipeline that synthesizes coherent training videos from sparse street-view images. We further introduce a Virtual Lookahead Sink to stabilize long-horizon generation by continuously re-grounding each chunk to a retrieved image at a future location. We evaluate SWM against recent video world models across three cities: Seoul, Busan, and Ann Arbor. SWM outperforms existing methods in generating spatially faithful, temporally consistent, long-horizon videos grounded in actual urban environments over trajectories reaching hundreds of meters, while supporting diverse camera movements and text-prompted scenario variations.

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.

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

Deep search capabilities have become an indispensable competency for frontier Large Language Model (LLM) agents, yet the development of high-performance search agents remains dominated by industrial giants due to a lack of transparent, high-quality training data. This persistent data scarcity has fundamentally hindered the progress of the broader research community in developing and innovating within this domain. To bridge this gap, we introduce OpenSeeker, the first fully open-source search agent (i.e., model and data) that achieves frontier-level performance through two core technical innovations: (1) Fact-grounded scalable controllable QA synthesis, which reverse-engineers the web graph via topological expansion and entity obfuscation to generate complex, multi-hop reasoning tasks with controllable coverage and complexity. (2) Denoised trajectory synthesis, which employs a retrospective summarization mechanism to denoise the trajectory, therefore promoting the teacher LLMs to generate high-quality actions. Experimental results demonstrate that OpenSeeker, trained (a single training run) on only 11.7k synthesized samples, achieves state-of-the-art performance across multiple benchmarks including BrowseComp, BrowseComp-ZH, xbench-DeepSearch, and WideSearch. Notably, trained with simple SFT, OpenSeeker significantly outperforms the second-best fully open-source agent DeepDive (e.g., 29.5% v.s. 15.3% on BrowseComp), and even surpasses industrial competitors such as Tongyi DeepResearch (trained via extensive continual pre-training, SFT, and RL) on BrowseComp-ZH (48.4% v.s. 46.7%). We fully open-source the complete training dataset and the model weights to democratize frontier search agent research and foster a more transparent, collaborative ecosystem.

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.

MemOS, a memory operating system for Large Language Models, addresses memory management challenges by unifying plaintext, activation-based, and parameter-level memories, enabling efficient storage, retrieval, and continual learning.

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.

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.

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.

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.

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.

OASIS is a scalable and generalizable social media simulator that models large-scale user interactions and replicates complex social phenomena across platforms.

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

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.

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.

A video diffusion model is repurposed as a latent world simulator to enhance multimodal large language models with implicit 3D structural priors and physical laws through spatiotemporal feature extraction and semantic integration.

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.

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

Video diffusion transformers enhanced with camera pose representation enable precise action control and long-term 3D consistency in interactive gaming environments through physics-based action spaces and geometric grounding.

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.

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.

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

The Qwen3-TTS series presents advanced multilingual text-to-speech models with voice cloning and controllable speech generation capabilities, utilizing dual-track LM architecture and specialized speech tokenizers for efficient streaming synthesis.

CubiD is a discrete generation model for high-dimensional representations that enables fine-grained masking and learns rich correlations across spatial positions while maintaining fixed generation steps regardless of feature dimensionality.

Fast Action Sampling for ImmediaTE Reaction (FASTER) reduces real-time reaction latency in Vision-Language-Action models by adapting sampling schedules to prioritize immediate actions while maintaining long-horizon trajectory quality.

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.

Flash-kmeans enables efficient online k-means clustering on GPUs through novel kernel-level optimizations that eliminate I/O bottlenecks and reduce atomic write contention.

A benchmark and metric suite for poster generation evaluates visual quality, coherence, and content accuracy, leading to a multi-agent pipeline that outperforms existing models with reduced computational cost.

LTX-2 is an open-source audiovisual diffusion model that generates synchronized video and audio content using a dual-stream transformer architecture with cross-modal attention and classifier-free guidance.

A generalist language model agent system autonomously designs and improves task-specific agents through memory-based reinforcement learning with stateful prompts and skill libraries.

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.

Mamba, a novel SSM-based model, outperforms Transformers in inference speed and scalability across various modalities by selectively propagating information and using efficient hardware-aware algorithms.

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.

Large language models are shifting from passive information providers to active agents intended for complex workflows. However, their deployment as reliable AI workers in enterprise is stalled by benchmarks that fail to capture the intricacies of professional environments, specifically, the need for long-horizon planning amidst persistent state changes and strict access protocols. In this work, we introduce EnterpriseOps-Gym, a benchmark designed to evaluate agentic planning in realistic enterprise settings. Specifically, EnterpriseOps-Gym features a containerized sandbox with 164 database tables and 512 functional tools to mimic real-world search friction. Within this environment, agents are evaluated on 1,150 expert-curated tasks across eight mission-critical verticals (including Customer Service, HR, and IT). Our evaluation of 14 frontier models reveals critical limitations in state-of-the-art models: the top-performing Claude Opus 4.5 achieves only 37.4% success. Further analysis shows that providing oracle human plans improves performance by 14-35 percentage points, pinpointing strategic reasoning as the primary bottleneck. Additionally, agents frequently fail to refuse infeasible tasks (best model achieves 53.9%), leading to unintended and potentially harmful side effects. Our findings underscore that current agents are not yet ready for autonomous enterprise deployment. More broadly, EnterpriseOps-Gym provides a concrete testbed to advance the robustness of agentic planning in professional workflows.

AI-Researcher automates the scientific research process, achieving high implementation success and manuscript quality through a comprehensive benchmark system.

Latent Entropy-Aware Decoding (LEAD) enhances multimodal large reasoning models by using entropy-aware reasoning mode switching and prior-guided visual anchor injection to reduce hallucinations.

Qianfan-OCR is a 4B-parameter vision-language model that unifies document parsing, layout analysis, and understanding while maintaining strong performance across multiple OCR benchmarks through its Layout-as-Thought mechanism.

A novel tokenization scheme synchronizes acoustic features with text tokens in TTS systems, enabling unified modeling and reduced hallucinations through flow matching and text-only guidance.

Helios is a 14 billion parameter autoregressive diffusion model for video generation that achieves real-time performance and high-quality long-video synthesis without conventional optimization techniques.

Pretraining creates a parameter distribution where task-specific experts become more densely populated in large models, enabling effective ensemble methods for post-training adaptation.

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.