Meta · 24 records

dateidtitleauthorsvenue↑
2026-11-202611.25387TuringSpace: Grounding Spatial Intelligence in Action with Interactive CAPTCHAsXuwei Ding et al.arXiv22
2026-11-192611.24161AVTR: Adaptive Multimodal Reasoning with Visual ImaginationXiaoqian Shen et al.arXiv8
2026-11-132611.17334Unbundling Multimodal Vision EncodingEllis Brown et al.arXiv14
2026-11-102611.09501e-means: outperforming k-means (in one-tenth the epochs)Swann Bessa et al.arXiv8
2026-11-032610.42128DecisionRank: On-Policy Distillation for Decision-Model RerankersYanran Wu et al.arXiv15
2026-10-252610.33966InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM AssistantsQinsi Wang et al.arXiv56
2026-10-212610.24628Muse Glimmer 30B: An On-Device Agentic ModelArash Ebrahim Abtahizadeh et al.arXiv18
2026-10-212610.28388Self-Credit: Credit Assignment for Long-Horizon Agentic RL via Intra-group RescalingFangcong Yin et al.arXiv13
2026-10-212610.28554OverSeer: Multi-Agent World Models Beyond What Agents SeeRunjia Li et al.arXiv5
2026-10-202610.16078Heterogeneous Agentic CoevolutionWenqian Ye et al.arXiv7
2026-10-082610.08902Agent Plasticity: Measuring Self-Improvement Through ExperienceHarman Singh et al.arXiv6
2026-10-082610.10515RoboJEPA: Scaling Robotic Latent World ModelsArtem Zholus et al.arXiv6
2026-10-072610.02781OPD Before RL: Warm-Starting Rubric-Based RL with On-Policy DistillationXinpeng Wang et al.arXiv14
2026-10-072610.07588Personal-Agent Mediated Recommendation with Cross-Platform User HistoryYu Xia et al.arXiv10
2026-10-052610.02508World Action Modeling with Progressive Visual PlanningFei Zhang et al.arXiv87
2026-10-052610.03391Native Action-Prior Learning from Videos for World Action ModelsZhaochong An et al.arXiv66
2026-10-022610.01509Sharpening Tax in Post-TrainingChangdae Oh et al.arXiv89
2026-10-012609.40195MemLife: Curating and Reasoning over Long-Term Egocentric Video MemoriesGuangzhi Xiong et al.arXiv2
2026-09-302609.37725Context Language ModelsRulin Shao et al.arXiv38
2026-09-302609.38157EmoRES-TTS: Residual-Enhanced Vector Steering for Emotional Speech GenerationKuan-Po Huang et al.arXiv15
2026-09-292609.32540In-Flight KV Cache with Clean Anchors for Faster Autoregressive Video DiffusionYikai Wang et al.arXiv30
2026-09-162609.16591FLAT: Resampling Image and Text into 1D Flexible-Length Aligned Transmodal Tokens for Retrieval and GenerationGuangyu Sun et al.arXiv16
2026-09-102609.05405WearableQA: A Benchmark for Health Reasoning over Real-World Wearable DataJi Soo Lee et al.arXiv24
2026-09-022609.01532Knowledge Distillation During Mid-Training Favors Reasoning over Factual RecallJacqueline He et al.arXiv8