The Research Desk.

Analytic non-axisymmetric 3D MHD equilibria without stellarator symmetry

Opal Issan, Jonathan Citrin, Matt Landreman, Brendan Tracey

Google DeepMind, University of California San Diego, University of Maryland, College Park

MemT-1.0: Towards Memory-Native Evolution

Views

Koi: Evidence-Guided, Self-Calibrating Planning for Multi-Cloud, Heterogeneous LLM Inference Fleets

Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience

Nanyang Technological University, Duke University, University of California, San Diego, Tsinghua University

EgoXperience: Learning World-Grounded Egocentric Representations from Human Experience

Nanyang Technological University, Tsinghua University, Peking University

HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs

Affiliation: Tsinghua University, Affiliation: Shanghai Jiao Tong University, Affiliation: University of Warwick, Tsinghua University, Shanghai Jiao Tong University

Text-to-Image Generative Modeling Without Language Pretraining

Affiliation: Stanford University, Affiliation: Stanford University Affiliation: Radical Numerics, Stanford University

StopLoss: Teaching Agents to Predict Failure and Stop Early

Jiarong Liang, Xueguang Ma, Dongfu Jiang, Yuxuan Zhang, Ping Nie

Predicting Privileged Information Effectiveness for On-Policy Self-Distillation

Kaijian Zou, Emiliano Penaloza, MohammadHossein Rezaei, Anisha Gunjal, Mohit Raghavendra

Legal Guidance in the Wild: How Users Seek Legal Help in Real-World LLM Conversations

Amirreza Mohseni, Gerasimos Spanakis

First Passage Attention

Yuxuan Liu