Trending Papers

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.

Harness² improves agent execution harnesses through recursive task-level refinement and continual harness-level learning, boosting performance across benchmarks without ground-truth verifiers.

Edge0 enables efficient mixture-of-experts inference on consumer hardware by predicting expert routing ahead of time and using recovery adapters to maintain quality with minimal active memory.

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.

ATAS improves long-form video consistency by compiling screenplays into referenced shot requests, retrieving spatial and temporal evidence from scene and event memories, and conditionally updating memory after generation.

RSI-Anything leverages predictive world models to guide action selection and adaptive learning, enabling recursive self-improvement across coding, interface navigation, laboratory planning, and scientific optimization tasks.

Unlimited OCR introduces Reference Sliding Window Attention to eliminate growing memory consumption during long-sequence OCR tasks, enabling efficient transcription of multiple pages in a single forward pass.

Player-1 improves game control by combining supervised fine-tuning on gameplay trajectories with reinforcement learning across synthetic environments, achieving substantial gains over demonstration-only baselines.

A framework for self-evolving agent harnesses uses feedback-driven revisions and evolutionary optimization with theoretical generalization guarantees while keeping model weights fixed.

Assistants from 2011 answered and waited, and agents from 2023 did a task and stopped. In September 2026 Meta's Muse showed an agent for one person, with accounts, devices, memory and a conversation that lasts, closed, in a vendor's cloud, in one country. Such an agent is expected to act on a person's accounts and devices, remember them across weeks, speak first when it is worth it, and answer for what it did. It is a kind of software, not a model, and until now had no open counterpart. This report defines the personal agent in five questions and three horizons. It reads how Muse is built from Meta's public record and a copy of its production prompt, each statement marked by its source. It then presents nanoMuse, the open-source counterpart under the GPL-3.0, one agent on every device a person owns, with hands on the phone's screen and the computer's. They share one conversation over a relay anyone can run; every action goes through a Sentinel, memory is files the person can read, and the model is their choice. Its size and cost are given as estimates. What is open, memory with provenance, an evaluation suite for the hands and an open model for them, is set out as a roadmap.

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

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.

Agent-Reflex trains interactive agents to reflect and recover from noisy tool feedback and user corrections through skill discovery, supervised fine-tuning, and on-policy optimization.

YuE2 unifies symbolic and audio music generation via symbolic planning, producing readable scores and full-song audio with state-of-the-art quality and agentic editing capabilities.

Frontier large language models exhibit Internal Safety Collapse, where they generate harmful content under specific task conditions, revealing inherent vulnerabilities despite alignment efforts.

Urbex reframes city-scale 3D grounding as an active search process using a vision-language agent with reinforcement learning, improving accuracy and speed without precomputed proposals.

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.

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.

VideoJev converts video observations into structured decisions by scoring temporal intervals and text options, enabling efficient reusable inference for visual agents.

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.

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.

StartLux-Decision introduces efficient and adaptive decision models using single-pass prediction and confidence-guided reasoning, trained via a three-stage pipeline with RLCD² reinforcement learning, and evaluates them on Decision Index and the DecisionPivot benchmark.

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.

PaddleOCR-VL-1.6 enhances document parsing performance through targeted data optimization and progressive post-training techniques, achieving state-of-the-art results on OmniDocBench v1.6.

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.

VisionHOPE introduces a self-modifying visual backbone where memory and learning co-evolve through coupled memories and stability-controlled updates, achieving strong results across vision tasks.

3DCodeVerse is a research ecosystem that provides datasets, automated verification harnesses, and empirical guidance to enable scalable, editable, and verifiable 3D content generation via executable code.

Transformers library provides state-of-the-art Transformer architectures and pretrained models for natural language processing tasks with a unified API and emphasis on extensibility and robust deployment.

This survey unifies research on agentic robots by tracing the evolution from robot policies to integrated systems using vision-language-action models, world models, and structured functional domains.

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.

Prism accelerates native joint video-audio generation at 2K by using dynamic sparse attention guided by local visual variance and audio-visual cross-attention to preserve semantic coherence and cross-modal interactions.

OpenGCUA introduces an open generalist computer-use agent that integrates visual and programmatic interfaces, harness-to-model decision transfer, and code execution to improve task performance and efficiency across models.

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.

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.

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

UniForge3D unifies 3D reconstruction and generation in a shared anchor-guided space to build complete assets from sparse images while preserving observed geometry and selecting relevant views for detail recovery.

Researchers benchmarked ten self-distillation methods against GRPO across multiple agent tasks and model sizes, finding limited consistent gains, hidden training instability, and no reliable cost-to-performance benefit.

SkillOpt introduces a systematic text-space optimizer for agent skills that trains skills as external agent state with stable updates and zero deployment inference overhead, achieving superior performance across multiple benchmarks and execution environments.

Jeva is a decision-native architecture that jointly enforces null-awareness, permutation invariance, option scalability, and cross-request caching for structured decision-making.

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

A new sparse voxel representation called O-Voxel enables high-quality 3D generative modeling with efficient inference and robust topology handling.

FlashVSR+ improves video super-resolution efficiency and preference alignment via hybrid attention and reward-weighted distribution matching distillation.

EvoOntology introduces a self-evolving ontology layer with schema, content, and tool components that enables data agents to query heterogeneous sources at runtime and improves performance across benchmarks.

Multi-session agentic reinforcement learning is introduced to handle trajectories across multiple sessions, with Mosaic optimizing session-level contributions and allocation via hierarchical advantage estimation.

DiffusionART improves reinforcement learning fine-tuning of diffusion models by removing hidden consistency forces and enabling stable, efficient training without EMA tuning or differentiable rewards.

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.

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.

Person-grounded AI skills are automatically distilled from heterogeneous traces into inspectable, correctable packages that capture both capabilities and behavioral patterns.

Multics introduces hierarchical group-level communication and dynamic grouping to scale multi-player world models efficiently while reducing interference and maintaining consistency.