I-FLOP is a fast causal structure learning algorithm from interventional data, leveraging target-filtered BIC scores with theoretical consistency guarantees.

Liuting Chen, Alex Markham

Proposes OCE-based risk-averse decision-making with prediction sets for high-probability risk control, validated in wireless beamforming.

Amirmohammad Farzaneh, Osvaldo Simeone

Proposes GGEW, a Gibbs-based adaptive weighting framework, improving forecast combination by modeling model errors probabilistically.

Prasen R. Nuthanakaluva, Nava K. Gaddam

HubMixer introduces learnable latent hubs for efficient feature interaction, reducing parameters by 50% while outperforming SOTA models.

Jie Zhou, Zixian Gong, Wenhao Li et al.

PhyMamba framework achieves robust battery health prognostics with 31.8% mean error reduction.

Sara Sameer, Yunyi Zhao, Wei Zhang et al.

RA-OPD improves student model performance by filtering misaligned trajectories, excelling in math and code benchmarks.

Siyuan Gan, Yuhan Li, Xiran Wang et al.

AMUR employs an information-theoretic approach for selective modality-interest alignment, achieving 5-8% NDCG improvements in multimodal recommendation.

Wenze Ma, Chenyu Sun, Yanmin Zhu et al.

StreamEMS enhances video streaming understanding in vision-language models with a self-evolving memory scheme, excelling on OVO-Bench.

Yuxin Liu, Peiqin Zhuang, Yali Wang

SpikeOPD stabilizes on-policy distillation for autoregressive spiking language models, improving accuracy by 2.9 points.

Enqiao Lu, Xingrui Yu, Yiwei Fu et al.

Introduces Fast Weight Attention with normalized first-order updates, enhancing long-sequence modeling in continual learning scenarios.

Yifan Zhang, Steve Ta, Jasper Zhang et al.

Using parameter Jacobian-based semigroup perturbation bounds, this work provides new insights into neural network output stability.

Halyun Jeong, Palle E. T. Jorgensen, Hyun-Kyoung Kwon et al.

Diffusion distillation compresses multi-step models into a single-step student, enhancing weather forecast efficiency.

Yiming Yang, Valentin Brekke, James Briant et al.

LitCurate uses large language models within a configurable, stage-wise workflow to build a structured EOS database with 1334 entries from 205 papers.

Abin Shakya, Wilson Samuels, Dominica Wilson et al.

Introduces EMR-HyperNEAT, a tensorized batch approach for multi-resolution substrate discovery, achieving 12-34× GPU speedup at depth 6.

Romain Claret, Michael O'Neill, Paul Cotofrei et al.

VGTok optimizes high-resolution object detection using regional separability, achieving 48.38 AP on VisDrone dataset.

Khayrul Islam

Introduces MCR-Bench, a multi-round defect state-aware benchmark for real-world code review, covering five languages, evaluating LLMs’ defect detection and lifecycle tracking.

Dewu Zheng, Yanlin Wang, Xiwen Wang et al.

Proposes weak-model guided RLVR with external prefixes to enhance exploration, significantly expanding reasoning coverage and diversity.

Xingyu Shen, Huishuai Zhang, Peng Li et al.

This paper compares Merge, Mix RL, and MOPD in multi-domain RLVR, showing performance gaps up to 8.6 points and providing practical guidance for different scenarios.

Siye Wu, Kai Yang, Yuchen Cai et al.

This paper introduces MILO, leveraging Large Reconstruction Models (LRMs) to reconstruct detailed 3D human-object interactions from a single image, outperforming state-of-the-art methods.

Agniv Chatterjee, Georgios Pavlakos

LeVJEPA uses collapse-free invariance loss and SIGReg regularization, reducing pretraining compute by up to 20x while maintaining or surpassing state-of-the-art performance.

Lukas Kuhn, Lucas Maes, Giuseppe Serra et al.