Proposes proof minimization algorithms reducing unsat proof size by 37%-82%, boosting verification efficiency.

Omri Isac, Idan Refaeli, Haoze Wu et al.

IGNS integrates Hamiltonian dynamics into graph neural simulators, greatly improving long-range interaction accuracy and stability.

Tai Hoang, Alessandro Trenta, Alessio Gravina et al.

DiffuGR uses diffusion language models to generate DocIDs, enhancing retrieval accuracy and efficiency.

Xinpeng Zhao, Zhaochun Ren, Yukun Zhao et al.

Stuart-Landau Oscillatory Graph Neural Network (SLGNN) dynamically adjusts amplitudes to address oversmoothing in GNNs.

Kaicheng Zhang, David N. Reynolds, Piero Deidda et al.

PGA-based formulas for k-simplex volume, centroid, and inertia offer a coordinate-free, efficient approach, validated on complex mesh data.

Steven De Keninck, Martin Roelfs, Leo Dorst et al.

SparseAutoencoder-based lightweight reward model achieves high accuracy with less than 1% trainable parameters.

Dengcan Liu, Jiahao Li, Zheren Fu et al.

ViPRA leverages video prediction and latent action learning to enable robot high-frequency smooth control, achieving 16% performance gain.

Sandeep Routray, Hengkai Pan, Unnat Jain et al.

AIA Forecaster combines agentic search, reconciliation, and calibration to achieve expert-level forecasting performance.

Rohan Alur, Bradly C. Stadie, Daniel Kang et al.

EquiTac leverages SO(2) symmetry to enhance sample efficiency in tactile policy learning, significantly reducing training samples.

Yizhe Zhu, Zhang Ye, Boce Hu et al.

Proposed Supervised Rational Attention (SRA), aligning model attention with human rationales, boosting interpretability and fairness.

Brage Eilertsen, Røskva Bjørgfinsdóttir, Francielle Vargas et al.

Llama-Embed-Nemotron-8B employs contrastive learning and multi-model data synthesis, achieving SOTA performance on multilingual and cross-lingual tasks, based on 16.1M query-document pairs, with open-source weights.

Yauhen Babakhin, Radek Osmulski, Ronay Ak et al.

FIBO leverages structured long captions and DimFusion to significantly improve controllability and expressiveness in text-to-image generation.

Eyal Gutflaish, Eliran Kachlon, Hezi Zisman et al.

Study analyzes Brazilian YouTube climate discourse using psycholinguistic methods, revealing generative AI manipulation risks; dataset includes 226K videos and 2.7M comments.

Wenchao Dong, Marcelo S. Locatelli, Virgilio Almeida et al.

Introduces Klear-Qwen3-AgentForge, combining supervised fine-tuning and multi-turn RL on Qwen3-8B, achieving state-of-the-art multi-task agentic performance.

Qi Wang, Hongzhi Zhang, Jia Fu et al.

OckBench jointly evaluates accuracy and token efficiency, revealing large disparities in reasoning density among models.

Zheng Du, Hao Kang, Song Han et al.

Introduces average cost stability, combining outstanding tasks and irrecoverable deadline failures, for assessing system reliability under time constraints.

Roee M. Francos, Daniel Garces, Orhan Eren Akgün et al.

BACO framework uses token-level model collaboration with routing strategies to optimize diversity and quality, achieving 21.3% joint improvement.

Yichen Wang, Chenghao Yang, Tenghao Huang et al.

QueStER uses lightweight LLMs to generate keyword queries, combining traditional BM25 retrieval with learned query reformulation, achieving +4.0 nDCG@10 over BM25.

Arthur Satouf, Yuxuan Zong, Habiboulaye Amadou-Boubacar et al.

DeepEyesV2 employs a two-stage training and dynamic tool invocation to enhance multimodal reasoning, achieving 63.7% accuracy on RealX-Bench.

Jack Hong, Chenxiao Zhao, ChengLin Zhu et al.

X-Diffusion employs diffusion models to learn cross-embodiment robot policies from human demonstrations, achieving 16% success rate improvement.

Maximus A. Pace, Prithwish Dan, Chuanruo Ning et al.