Prépublication
· 2026 arXiv
EN
Existing agent benchmarks typically evaluate agents on well-specified tasks defined by issue reports, specifications, or task tickets. Real-world customer delivery is different: forward-deployed engineers (FDEs) must clarify ambiguous requirements, work in liv...
Dongxu Yang, Hongda Yan et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Large language models (LLMs) are expanding from assisting scientific research and engineering to autonomously developing and improving AI systems, advancing AI for AI (AI4AI) as a pathway toward recursive self-improvement (RSI). Workflow-driven machine learnin...
Shangheng Du, Jinxin Shi et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometric priors. However, the view dependence of these features entangles scene dynamics with camera-induced variati...
Hongchang Liu, Jian Liu et al.
#cs.CV
Prépublication
· 2026 arXiv
EN
We introduce 3DCodeVerse, a research ecosystem that enables coding agents to create editable and verifiable 3D content at scale. 3DCodeVerse comprises large-scale, high-quality datasets of executable 3D code, a coding harness for spatial reasoning and automate...
Yipeng Gao, Ziyao Zeng et al.
#cs.CV
Prépublication
· 2026 arXiv
EN
How can robots autonomously collect the contact-rich experience needed to learn dexterous manipulation? We present DSP: Dexterity from supervised play, a framework in which a robot learns a play policy from minutes of human play through behavior cloning, then...
Botao He, Yiming Dou et al.
#cs.RO
Prépublication
· 2026 arXiv
EN
Large language models (LLMs) can often tell when they are being tested. This ability, known as evaluation awareness, can lead models to act more cautiously during safety evaluations than in real use. Existing work has studied evaluation awareness almost entire...
Trizal Garg, Sibayan Mitra et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Deep search agents solve knowledge-intensive tasks through long-horizon, multi-turn interactions, where assigning credit to individual search decisions is fundamentally difficult. Outcome-based reinforcement learning provides only a single terminal reward shar...
Siqi Chen, Zhengxi Lu et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Continuous diffusion language models (CDLMs) offer an alternative approach to parallel language generation; however, most existing studies directly use transformer architectures from either causal large language models (LLMs) or Diffusion Transformers (DiTs) f...
Weijian Luo, Yujian Chen et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Image editing models produce impressive results for individual edits, but their quality often deteriorates over multiple editing turns. Each editing turn introduces small distribution shifts that accumulate over successive edits, creating a gap between the inp...
Sara Dorfman, Nir Goren et al.
#cs.CV
Prépublication
· 2026 arXiv
EN
In partially-observed dynamical systems, distinct hidden mechanisms can produce similar observable summary statistics while explaining individual events differently. We introduce causal fingerprints: realization-dependent patterns that track how information ab...
Zhang Jiang, Marios Andreou et al.
#math.DS
Prépublication
· 2026 arXiv
EN
Weather prediction is undergoing an important shift, with data-driven AI models outperforming leading numerical weather prediction systems on average at a fraction of the computational cost. Yet, whether this success extends to extreme events remains contested...
Nathan Chalumeau, Xavier Renard et al.
#cs.LG
Prépublication
· 2026 arXiv
EN
Multi-aperture reception converts spatial diversity into a coherent-combining gain, but only if the relative phase and polarization of the collected field replicas are aligned before they are summed. In a satellite-to-ground downlink the atmosphere perturbs am...
Chen Cheng, Tong Luo et al.
#physics.optics