RiVER leverages score-based optimization with instance ranking and winner-heavy rewards, improving LLMs on algorithmic tasks by 8.9%-9.4%.

Yingyu Lin, Qiyue Gao, Nikki Lijing Kuang et al.

RouterVLA leverages smoke tests as supervision via outcome-disjoint cross-fitting, boosting heterogeneous VLA policy selection success from 46.86% to 61.49%.

Xingyu Ren, Chugang Yi, Ge Ma et al.

Error-conditioned neural solver (ENS) uses PDE residuals as direct inputs, achieving 10× accuracy improvement and avoiding costly optimizers.

Haina Jiang, Liam Wang, Peng-Chen Chen et al.

Verifier-driven framework enables parallel LLM agents with 29.9% RMSE reduction in BEA 2026, ensuring reproducibility and auditability.

Vassili Philippov, Pavel Katunin, Dmitry Andreev et al.

SAM2Matting leverages pre-trained VOS trackers (e.g., SAM2, SAM3) with region proposal and multi-scale alpha refinement, achieving state-of-the-art zero-shot video and image matting performance.

Ruiqi Shen, Guangquan Jie, Chang Liu et al.

PEEU framework enables small multimodal models to achieve 30.6% success in web navigation by autonomous exploration and hindsight experience, outperforming larger models.

Tianyi Men, Zhuoran Jin, Pengfei Cao et al.

Proposes DexAC-WM, a structured high-DoF action representation, significantly improving visual-temporal prediction accuracy by 15% in FID and 20% in FVD on egocentric datasets.

Zizhao Yuan, Zhengtu Liang, Taowen Wang et al.

Introduces two activation regularizers for Top-k sparse autoencoders, enhancing interpretability and robustness without sacrificing reconstruction quality.

Nathanaël Jacquier, Maria Vakalopoulou, Mahdi S. Hosseini

This study pioneers the application of Large Language Models (LLMs) for securities eligibility screening at the German Central Bank, achieving up to 91% document-level precision through a generative information extraction pipeline.

Serhii Hamotskyi, Akash Kumar Gautam, Christian Hänig

This paper establishes the algorithmic equivalence between Blackwell approachability and Gradient Equilibrium (GEQ), unifying online decision frameworks.

Brian W. Lee, Nika Haghtalab, Michael I. Jordan et al.

Ribbon: Influence-function-based scalable approximation for Bayesian uncertainty, improving calibration and efficiency.

Graham Gibson, John Tipton, Kellin Rumsey et al.

BINEVAL decomposes evaluation into binary questions, enabling interpretable, multi-dimensional assessment with superior correlation to human judgments.

Sangwoo Cho, Kushal Chawla, Pengshan Cai et al.

Introduces syntactic belief update model using Rényi divergence, outperforming lexical surprisal in predicting human reading difficulty on garden path sentences.

Alan Zhou, Miloš Stanojević, John T. Hale

Proposes iterative self-improving codebooks leveraging model understanding to enhance image generation safety, reducing harmful content by over 85%.

Yunqi Xue, Zhijiang Li, Philip Torr et al.

JERP jointly updates experiential rule pool and policy, boosting decision accuracy in multi-step tasks.

Shicheng Ye, Chao Yu

SubdivAR enhances mesh subdivision accuracy via autoregressive prediction, reducing Hausdorff Distance by 18.8%.

Huipeng Guo, Zikai Song, Hang Long et al.

MinGram uses BPE seed, Hard EM on minimal path, and single pruning to improve compression and morphology.

Sander Land

Introduces decision-alignment framework for uncertainty quantification (UQ), critiques traditional metrics, and proposes prior-weighted utility (PWU) metrics, validated through experiments showing superior decision utility alignment.

Annika Schneider, Tommy Rochussen, Joshua Stiller et al.

Unison evaluates multimodal models' synergy via internal consistency, guided understanding and generation, and mutual enhancement, using 169 samples and a fine-tuned Unison-Judge.

Jinyu Liu, Xincheng Shuai, Henghui Ding et al.

SharpMoE leverages predicted clean latent features for saliency-aware routing, boosting diffusion MoE image generation by ~20% in FID.

Haoyou Deng, Keyu Yan, Chaojie Mao et al.