Proposes Graph Engineering to organize tasks, coordinate heterogeneous agents, and manage states for system-level intelligence.

Yuyuan Feng, Zhishang Xiang, Chaobin Yang et al.

COTA enhances runtime intervention in LLM agents by comparing alternatives, improving performance in environments like WebShop.

Yanze Jiang, Mingxuan Li, Yuhao Wang et al.

Introduces free-probability kernels for zero-rollout hyperparameter selection in reservoir computing, reducing simulation costs while maintaining accuracy.

Sara Malacarne, Andrea Ceni, Claudio Gallicchio

Classical time series methods reveal linear models outperform GNNs on benchmarks, questioning dataset discriminative power.

Kenneth Martin, Simon Heilig, Asja Fischer et al.

UpgradeBench offers a decision-centric benchmark for upgrading LLM specialists, with a mean quality regret of 0.37pp.

Ye Chen, Weining Zhang

Proposes ingest-time semantic compilation (ISC) with maintained semantic index, outperforming query-time interpretation in accuracy and cost efficiency.

Kyle Wild, Yusuke Takahashi, Asako Uraki

KoViDoRe employs multi-stage synthetic data generation to benchmark Korean multi-page document retrieval, revealing current models' struggles with complex, structured content.

Yongbin Choi, Yongwoo Song, Mujeen Sung

SAC-Copula uses Gaussian copula to construct smooth correlated Gumbel fields for quality-preserving watermarking in diffusion language models.

Baixin Li, Haiyun He

Test-time scaling for scientific equation discovery, search width is key, improves efficiency and performance.

Haowei Lin, Hubert Lim, Xiangyu Wang et al.

Proposes CAIRO, a framework that structures large-scale metadata for personalized item profiling, boosting LLM reranking performance.

Dojun Hwang, Seunghan Lee, Cheonyoung Park et al.

AESR combines global prompt enhancement and semantic repair to improve identity-preserving video generation.

Jiayi Gao, Changcheng Hua, Jiaqi Tang et al.

This paper systematically evaluates diffusion-based block-parallel decoding in multimodal models, highlighting potential speedups up to 3.6× with current architectures.

Yantao Li, Huanlin Gao, Fang Zhao et al.

GBC introduces class-conditional Gaussian feature bridges, improving long-tailed SSL by enforcing geometric consistency and feature interpolation, boosting tail-class accuracy.

Hongyang He, Xinyuan Song, Yan Zhong et al.

ArtiMo employs large models for zero-shot text-guided articulated mesh animation, integrating URDF constraints and causal reasoning.

Chunyu Zou, Peng Dai, Yi-Hua Huang et al.

TopoSurfel introduces a differentiable mesh extraction and mesh-guided surfel evolution, achieving high-precision surface reconstruction with Chamfer Distance 0.70M and F1-Score 0.64, outperforming baselines.

Chuanjin Fan, Wenjie Chang, Bohao Liao et al.

Structured parameter initialization (SPIQ) enhances gate-based QAOA for join order optimization, improving convergence and solution quality by up to 5×.

Divya Shekar, Ruokun Wu, Dhanvi Bharadwaj et al.

Conditional-Independence-Regularized Autoencoder effectively models full conditional distributions of mixed data, improving categorical recovery by 30%.

Siyuan Tang, Gongjun Xu, Ji Zhu

LiLiCorr improves speculative decoding acceptance length by 9-19% using lightweight likelihood correlation.

Matan Rusanovsky, Yoav Miron, Roy Uziel et al.

Analytical and particle-based methods for uncertainty propagation in random neural networks using Leaky ReLU's piecewise linearity, extended to autonomous dynamical systems.

Janice Adams, Daniele Venturi

ConceptGuard uses dual-use concepts to evaluate context-sensitive unlearning, revealing limitations of current methods in contextual separation.

Sahil Kale, Ian Harris