GCJR recovers minimal repair families with 1.000 FEM and cuts replay by 55.1% on 90 controlled cases.

Bingjie Li, Yumeng Song, Zhongming Yao et al.

Study reveals sustained heterogeneity in LLM-driven traffic, key metric is critical penetration rate at various densities.

Yujun Qi, Yangyang Guan

Chat-Edit-3D++ enables flexible 3D and 4D scene editing via large language models, enhancing editing efficiency and outcomes.

Shuangkang Fang, Yufeng Wang, Yi-Hsuan Tsai et al.

Convergence analysis of PPMD combining geometric, spectral, and kernel methods reveals error propagation and model limitations.

Jiaming Guo, Dunhui Xiao

Detecting and guiding LLM-generated Korean poetry using interpretable features, achieving 83.60 AUC-ROC.

Keunhyeung Park, Seunguk Yu, YoungBin Kim

Study shows LALMs struggle with acoustic-semantic incongruence, improved by CREMA-ASIS dataset.

Yu-Wen Chen, William Ho, Maxim Topaz et al.

Accurate plate reverb parameter estimation using two-stage evolutionary search, achieving NMSE of 10^-14.

Byunghoo Park, Jayeon Yi, Takyoung Kim et al.

First large-scale evaluation of text-guided models like Nano Banana for facial editing, assessing ~1M images.

Rahul Nair, Saurav Pandit, Hannah Kerner

Developed Aero Hand Open with a MuJoCo-based tendon-driven simulation model enabling zero-shot transfer of reinforcement learning policies for dexterous grasping.

Nan Wang, Mohit Yadav, Jonathan Wulff et al.

ChainSplat uses screw theory to jointly learn 3D geometry and dynamics of deformable linear objects from multi-view RGB videos, enabling accurate prediction and high-fidelity rendering.

Seungyeon Kim, Noémie Jaquier

This paper introduces GeBDA, a sequence prediction approach using the Gemma model for end-to-end building damage assessment from bi-temporal satellite images, achieving competitive localization and classification.

Olivier Dietrich, Krishna Sapkota, Konrad Schindler et al.

Provides sharp asymptotic analysis of kernel ridge regression under power-law anisotropic Gaussian data, revealing how input geometry shapes learning curves and generalization.

Lorenzo Rizzi, Arie Wortsman Zurich, Bruno Loureiro

This work uses information theory to establish fundamental limits on extracting human intent solely from textual data.

Emily Cheng, Ryan Cotterell

This paper systematically organizes 2025-2026 optimizers into a multi-dimensional framework, emphasizing matrix-awareness, temporal estimation, and system representation, highlighting no single optimizer dominates.

Ruoran Xu

QUEST combines synthetic query generation, topic extraction, and LLM-based relevance assessment to identify credibility indicators in asylum appeal documents, achieving MAP@100 of 0.572 and 0.589 on datasets.

Maria Vlachou, Anna Murphy Høgenhaug, Mohammad N. S. Jahromi et al.

Logos extends spatiotemporal composability to multi-process systems using append-only transcripts, ensuring fault tolerance and reversibility.

Hanzhang Jia, Liheng Zeng, Hao Cheng et al.

Optimized Relief-based algorithms (e.g., MultiSWRF*, μ-Relief) significantly enhance detection of 2nd-order interactions with 10-35x faster runtime.

Kia Kazemi-Nia, Harsh Bandhey, Philip J. Freda et al.

Utilizes pretrained video diffusion models for joint monocular depth and normal estimation via next-frame prediction, reducing data needs significantly.

Haosen Yang, Jifei Song, Zhensong Zhang et al.

DARTS uses entropy-weighted L1 and position biases to correct autoregressive decoder biases, improving multi-task model merging with only 0.1% parameter overhead.

Aaryan Ajay Sharma, Sai Nishanth Padala, Seganrasan Subramanian

InstructMesh employs latent-space region operations combined with multimodal interaction to enable non-experts to selectively repair generative 3D models for fabrication.

Faraz Faruqi, Ahmed Katary, Demircan Tas et al.