University of California at San Diego · 23 records

dateidtitleauthorsvenue↑
2026-10-232610.30468SplatLab: Autonomous Scientific Discovery for 3DGS ResearchSeemandhar Jain et al.arXiv11
2026-10-192610.21783Learning to Think and Explore in 3D WorldYijiang Li et al.arXiv71
2026-10-132610.09116SPLATIFY: Reproduce, Discover, Innovate! From Papers and Ideas to Trainable 3DGS CodeSeemandhar Jain et al.arXiv11
2026-10-122610.12453Mental-Models for Multi-Agent SystemsHanan Gani et al.arXiv7
2026-10-052609.34826WM-VLM: Probing Internal World Models for Interleaved Visual-Textual ReasoningYuheng Zha et al.arXiv7
2026-09-302609.38146LIFT: Layout-In-Future Video Generation under Large Viewpoint Change via On-Policy Self-DistillationShengxiang Ji et al.arXiv7
2026-09-032504.15476An Empirical Study on Zero-Data Bootstrapping for Conversational Recommender SystemsRohan Surana et al.arXiv1
2026-09-012608.26671RECAP-Forcing: Retaining Content Appearances for Long Video GenerationHaiyang Xu et al.arXiv4
2026-08-202608.17253Co-RL: Unsupervised Reasoning Emerges from Diverse Cohort in Multi-agent RLYunhao Yang et al.arXiv51
2026-08-192608.14036Demystifying Agent Skills: Why They Work-Until They Don'tZhiyuan Jiang et al.arXiv56
2026-08-182608.16072Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy OptimizationYixuan Wang et al.arXiv30
2026-08-172608.14144Self-Supervised Visual On-Policy DistillationYijiang Li et al.arXiv54
2026-08-112608.06296On-Policy Self-Distillation without Any SupervisionYijiang Li et al.arXiv125
2026-06-262606.27326Hallucination in World Models is Predictable and PreventableNicklas Hansen et al.arXiv9
2026-06-242606.24595MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State RecoveryEnze Ma et al.arXiv3
2026-06-232606.16246Demystifying Training-Time Augmentation for Data-Constrained Language Model PretrainingMichael K. Chen et al.arXiv4
2026-06-232606.19750Manifold Bandits: Bayesian Curriculum Learning over the Latent Geometry of Large Language ModelsDarrien McKenzie et al.arXiv4
2026-06-162606.16202EgoPhys: Learning Generalizable Physics Models of Deformable Objects from Egocentric VideoHyunjin Kim et al.arXiv2
2026-06-092606.08481PIPE-Cypher: Automatic Enterprise Benchmark Generation for Text-to-Cypher SystemsSuraj Ranganath et al.arXiv1
2026-06-042606.00477Do Text Edits Generalize to Visual Generation? Benchmarking Cross-Modal Knowledge Editing in UMMsXin Gao et al.arXiv1
2026-06-042606.03965Agentic Chain-of-Thought Steering for Efficient and Controllable LLM ReasoningYu Xia et al.arXiv
2026-05-152605.14169BOOKMARKS: Efficient Active Storyline Memory for Role-playingLetian Peng et al.arXiv6
2026-05-122605.08520FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage OrchestrationZhengding Hu et al.arXiv6