Qwen · 37 records

dateidtitleauthorsvenue↑
2026-11-052611.02846SpecOPD: Branch-Parallel On-Policy Distillation for Speculative DecodingTianyu Liu et al.arXiv14
2026-10-282610.35439Omni-InteractScope: Benchmarking Proactive and Agentic Interaction in Continuous Omnimodal StreamsYinsong Yan et al.arXiv15
2026-10-272610.33472Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt EnhancementZekai Zhang et al.arXiv33
2026-10-262610.35138Co-Design Bench: Multimodal Collaborative Coding Evaluation for Agentic ModelsXilin Wei et al.arXiv20
2026-10-232610.31158Beyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research ReportsJiahao Meng et al.arXiv26
2026-10-202610.23986BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World ReferencesXinyu Fang et al.arXiv34
2026-10-142610.14766Last Secure Code Benchmark: Is the Code You Vibe Secure?Tianyi Wu et al.arXiv21
2026-10-122610.07987VisionWeave: Weaving Elastic Visual Representations as a Native Capability of MLLMsYuan Feng et al.arXiv61
2026-10-072610.07767TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language ModelsXin Wang et al.arXiv55
2026-10-062609.39490OmniReasoning: Pushing the Limits of Audio-Visual Joint ReasoningJunming Lin et al.arXiv26
2026-09-292609.33848QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon AgentsWeiqi Wang et al.arXiv49
2026-09-242609.27321Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved MechanismsXinjie Shen et al.arXiv21
2026-09-212609.22000RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use AgentsShuai Bai et al.arXiv76
2026-09-212609.21465OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual DialogueHaolin He et al.arXiv69
2026-09-042609.04148Terminal-Universe: Turning Agent Trajectories into Scalable Terminal EnvironmentsJie Wu et al.arXiv80
2026-09-022608.30730E-Commerce Bench: Evaluating LLM Agents on Long-Horizon Autonomous Business OperationWei Fan et al.arXiv17
2026-09-012608.30320On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training StabilityZihan Qiu et al.arXiv61
2026-08-312609.00111Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous DrivingXin Zhou et al.arXiv131
2026-08-052607.28661Are the Financial Reasoning from LLMs Credible? A Real World Test over Long-Horizon StatementsXinke Tong et al.arXiv19
2026-07-202607.11699Qwen-Music Technical ReportJin Xu et al.arXiv33
2026-06-292606.27608Qwen-Image-2.0-RL Technical ReportYixian Xu et al.arXiv53
2026-06-292606.17846Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation ModelsHaoqi Yuan et al.arXiv34
2026-06-292606.18112Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation SystemJiazhao Zhang et al.arXiv30
2026-06-262606.26907Qwen-Image-Agent: Bridging the Context Gap in Real-World Image GenerationZekai Zhang et al.arXiv53
2026-06-242606.24597Qwen-AgentWorld: Language World Models for General AgentsYuxin Zuo et al.arXiv165
2026-06-242606.26300The Verification Horizon: No Silver Bullet for Coding Agent RewardsBinghai Wang et al.arXiv55
2026-06-232606.21906Deeper is Not Always Better: Mitigating the Alignment Tax via Confident Layer DecodingXuanming Zhang et al.arXiv27
2026-06-182606.19341Native Active Perception as Reasoning for Omni-Modal UnderstandingZhenghao Xing et al.arXiv20
2026-06-172606.18249Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to UnificationWujian Peng et al.arXiv14
2026-06-162606.17030Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video GenerationJie Zhang et al.arXiv48
2026-06-112606.12370Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection SamplingYucheng Li et al.arXiv18
2026-06-112606.12373Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning GeneralizationHao Xiang et al.arXiv7
2026-06-042606.03746Qwen-Image-Flash: Beyond Objective DesignTianhe Wu et al.arXiv37
2026-05-282605.30280Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot EmbodimentsQiuyue Wang et al.arXiv144
2026-05-262605.25624CUA-Gym: Scaling Verifiable Training Environments and Tasks for Computer-Use AgentsBowen Wang et al.arXiv30
2026-05-142605.13565Qwen-Image-VAE-2.0 Technical ReportZekai Zhang et al.arXiv61
2026-05-122605.10730Qwen-Image-2.0 Technical ReportBing Zhao et al.arXiv119