Prépublication
· 2026 arXiv
EN
Long contexts enlarge key--value (KV) caches, increasing memory use and attention costs during generation. Compressing these caches must preserve information for unseen questions. We introduce **Latent Compaction**, which learns a compact document-specific cac...
Yingfa Chen, Xingyu Shen et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
Post-training binds the competence of an LLM agent to the interface it was trained on. Semantics-preserving perturbations of that interface, such as renaming an action or changing the notation, leave the underlying task unchanged yet make the agent worse at it...
Artin Tajdini, Manpreet Kaur et al.
#cs.LG
Prépublication
· 2026 arXiv
EN
We introduce a novel method for verifying Lean formalizations that goes beyond LLM-judged faithfulness. Unlike most verification methods, which rely on LLMs or trained models, our method combines LLM-based semantic review with mechanical numerical and vacuity...
Anna Mei, Michael R. Douglas
#cs.LO
Prépublication
· 2026 arXiv
EN
On-policy distillation (OPD) trains a student model on its own rollouts with token-level supervision from a teacher, assuming that the teacher can reach a correct answer from any prefix the student writes. We call this property the recoverability of the teache...
Kaixiang Zhao, Tianrun Yu et al.
#cs.LG
Prépublication
· 2026 arXiv
EN
Brazilian Portuguese remains underserved by open-source language models, and the few that are safety-aligned rarely document what did not work or validate comparisons with statistical significance. We present Manacá-1B-Instruct-2, the second-generation post-tr...
Bruno Leonardo Santos Menezes, Carlos Leonardo Souza Cardoso et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
Automated research with LLM agents has shown promising results in various domains. Through iterative refinement and scored verifiers, LLM agents are capable of discovering new ideas. There are different ways to scale automated research: let one agent iterate f...
Shangyin Tan, Ben Zhou et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
3D object composition is pivotal to digital content creation and robot simulation, while also serving as a critical testbed for physical intelligence in foundation models. Previous methods either leverage large language models to directly predict object coordi...
Weichen Lyu, Chong Xia et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and multi-agent environments. Social intelligence comprises two aspects: cognitive intelligence, the ability to infe...
Ali Zamani, Soroush Esfahanian et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
Conventional deep convection schemes remain a primary source of uncertainty in global climate models, yet replacing them with data-driven alternatives while maintaining physical consistency and numerical stability is an open challenge. We present a hybrid conf...
Xavier J. Levine, Griffin Mooers et al.
#physics.ao-ph
Prépublication
· 2026 arXiv
EN
Measurements of optical turbulence are essential in both ground-based astronomy and free-space optical communication systems contexts for multiple reasons including supporting and optimizing Adaptive Optics. Similarly, forecasting these parameters is highly va...
Camilo Weinberger, Elena Masciadri et al.
#astro-ph.IM
Prépublication
· 2026 arXiv
EN
High-resolution precipitation datasets are essential for weather forecasting, hydrological modeling, and climate risk assessment, yet observation-based gridded products often lack the spatial detail required for regional applications. Here we present PrecipCor...
Bipin Kumar, Rajib Chattopadhyay et al.
#physics.ao-ph
Prépublication
· 2026 arXiv
EN
Inferring climate fields at higher spatio-temporal resolutions from coarse representations is essential for accessing socioeconomic and environmental impacts. The downscaling of modelled climate fields faces a trade-off between data storage constraints, simula...
Yu Huang, Shangshang Yang et al.
#physics.ao-ph