| 2026-11-04 | 2611.03634 | WebHarbor: Docking Real Websites for Evolving GUI Agent Environments | Zhaoyang Wang et al. | arXiv | 24 | |
| 2026-11-04 | 2611.02079 | ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost | Xingzhi Qian et al. | arXiv | 18 | |
| 2026-11-04 | 2610.40747 | Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses | Xiaoyi Zhang et al. | arXiv | 17 | |
| 2026-10-27 | 2610.34886 | ThinkBit: Search Codewords, Recalibrate Scales, Select by Reasoning | Rhea Sanjay Sukthanker et al. | arXiv | 2 | |
| 2026-10-21 | 2610.26542 | Breaking Diffusion: Parallel Reasoning with Draft Refinement | Jaesang Yu et al. | arXiv | 22 | |
| 2026-10-19 | 2610.23194 | OfficeTown: Generative Organizations for Evolving Agent Intelligence | Hongchen Wei et al. | arXiv | 23 | |
| 2026-10-15 | 2610.18072 | LensErase: A Foundation Model for Erase Anything | Yunuo Chen et al. | arXiv | 46 | |
| 2026-10-14 | 2610.18860 | AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards | Xin Gao et al. | arXiv | 43 | |
| 2026-10-14 | 2610.18472 | Simplifying and Improving On-Policy Distillation for Few-Step Diffusion | Jiawei Zhang et al. | arXiv | 27 | |
| 2026-10-14 | 2610.14163 | AutoEnvScaling: Automating the Data Flywheel with Terminal Agents | Simon Yu et al. | arXiv | 11 | |
| 2026-10-09 | 2610.12230 | From Prompting to Composing: A Spatial Canvas Interface for Poster Generation | Yitong Wang et al. | arXiv | 13 | |
| 2026-10-07 | 2610.05912 | MiniCorp: The Last Mile of the AI Agent Firm | Jingying Zeng et al. | arXiv | 36 | |
| 2026-10-06 | 2610.06361 | Capability-Driven Self-Evolution of Agent Memory | Yaoqi Chen et al. | arXiv | 15 | |
| 2026-10-06 | 2610.02076 | LLM-as-Jev: LLMs Are Already Jev-Style Decision Models -- When and How to Fine-Tune Them | Yinheng Li et al. | arXiv | 14 | |
| 2026-10-02 | 2610.00906 | ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization | Sungho Park et al. | arXiv | 59 | |
| 2026-10-01 | 2609.35646 | Rubric Rewards from Item Response Theory | Milad Yazdani et al. | arXiv | 11 | |
| 2026-09-30 | 2609.37226 | Follow the Entities: A Corpus Map for Agentic Search | Soyeong Jeong et al. | arXiv | 68 | |
| 2026-09-23 | 2609.25804 | The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks | Wenbo Pan et al. | arXiv | 53 | |
| 2026-09-18 | 2609.20511 | When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation | Yuxiao Yang et al. | arXiv | 59 | |
| 2026-09-18 | 2609.19671 | When2Think: Learning When and How Much to Reason | Jaejun Shim et al. | arXiv | 31 | |
| 2026-09-03 | 2609.02812 | VibeVoice-ASR-Streaming Technical Report | Yujie Tu et al. | arXiv | 16 | |
| 2026-08-31 | 2608.28444 | Sliding-window beats linear attention | Alexia Jolicoeur-Martineau et al. | arXiv | 18 | |
| 2026-08-26 | 2608.23041 | AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Traces | Sungho Park et al. | arXiv | 66 | |
| 2026-08-25 | 2608.19741 | One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflows | Zhuochun Li et al. | arXiv | 8 | |
| 2026-08-19 | 2608.17528 | Agent Lightning v1.0: Towards Harnessed Agentic RL | Zhiyuan He et al. | arXiv | 35 | |
| 2026-08-11 | 2608.08097 | OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching | Can Xiao et al. | arXiv | 23 | |
| 2026-08-04 | 2608.00155 | AgentStream: How Well Do Self-Evolving LLM Agents Perform Under Streaming Tasks? | Dong Yan et al. | arXiv | 19 | |
| 2026-07-29 | 2607.24904 | Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation Model | Senqiao Yang et al. | arXiv | 38 | |
| 2026-07-24 | 2607.21557 | OpenForgeRL: Train Harness-native Agents in Any Environment | Xiao Yu et al. | arXiv | 11 | |
| 2026-07-22 | 2607.19064 | Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing | Xinjie Zhang et al. | arXiv | 76 | |
| 2026-07-07 | 2607.04439 | ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes | Qihao Zhao et al. | arXiv | 59 | |
| 2026-07-07 | 2607.04438 | ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog | Lingao Xiao et al. | arXiv | 56 | |
| 2026-07-07 | 2606.31145 | SeKV: Resolution-Adaptive KV Cache with Hierarchical Semantic Memory for Long-Context LLM Inference | Amirhossein Abaskohi et al. | arXiv | 6 | |
| 2026-07-02 | 2606.31179 | HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents | Qianchu Liu et al. | arXiv | 6 | |
| 2026-07-02 | 2606.28430 | Building to the Test: Coding Agents Deliver What You Check, Not What You Requested | Yanuo Ma et al. | arXiv | 5 | |
| 2026-06-18 | 2606.13802 | A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets | Tejas Agrawal et al. | arXiv | 1 | |
| 2026-06-16 | 2606.14066 | FastContext: Training Efficient Repository Explorer for Coding Agents | Shaoqiu Zhang et al. | arXiv | 77 | |
| 2026-06-12 | 2606.09426 | WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces | Wanli Li et al. | arXiv | 46 | |
| 2026-06-09 | 2606.05597 | AsyncWebRL: Efficient Multi-Step RL for Visual Web Agents | Hao Bai et al. | arXiv | 5 | |
| 2026-06-02 | 2606.02031 | OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents | Rui Yang et al. | arXiv | 20 | |
| 2026-05-25 | 2605.21573 | Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models | Dong Chen et al. | arXiv | 104 | |
| 2026-05-13 | 2605.12501 | Covering Human Action Space for Computer Use: Data Synthesis and Benchmark | Miaosen Zhang et al. | arXiv | 15 | |
| 2026-05-01 | 2604.28181 | Synthetic Computers at Scale for Long-Horizon Productivity Simulation | Tao Ge et al. | arXiv | 20 | |