Study optimizes information aggregation rates, proposing M²/D error rate, applicable to linear regression and logistic classification.

MohammadHossein Bateni, Zahra Hadizadeh, MohammadTaghi Hajiaghayi et al.

mHC optimizes residual streams via selective routing and near-identity mixing, improving C4 perplexity by 1.9%.

Pengxiang Zhao, Xing Li, Xianzhi Yu et al.

Using β-VAE to extract features from ECG for myocardial scar differentiation, achieving ROC of 0.643.

Shayan Sharifi, Riccardo Treu, Ilaria Gandin et al.

CV-CCI combines observational and randomized data for reliable counterfactual KPI inference in wireless networks.

Abdessamed Qchohi, Jessica Moysen Cortes, Matteo Zecchin

Using susceptibility techniques to analyze Turing machines' algorithmic structures, revealing symmetry and path separation features.

Billy Snikkers, Rumi Salazar, Daniel Murfet et al.

Open-source platform combines miniature Ackermann vehicle with behavior cloning, achieving 6.1cm average cross-track error in real-world lane following.

Gustavo Claudio Karl Couto, Eric Aislan Antonelo, Gabriel George Zipperer

PreferenceEKF employs subspace extended Kalman filtering for efficient active reward learning, improving sample efficiency and scalability.

Yutai Zhou, Erdem Bıyık

RecurTrace enhances reasoning accuracy with loop-time memory, achieving 56.9% on MathQA.

Yuxiang Wang, Kunyu Feng, Yingda Shen et al.

Graph Machine (GM) maintains O(n) state with sparse, dynamic routing, replacing 75% of dense Transformer layers, achieving near state-of-the-art performance on 15.7B tokens.

Lintai Hou

Teacher gating in on-policy distillation (TGOPD) verifies prompt-level teacher reliability, boosting performance and GPU utilization by 8-fold.

Zhiwei Zhang, Zechen Sun, Fei Zhao et al.

End-to-end pipeline combining large-scale problem curation, synthetic reasoning, SFT, RL, and GenCorrect enabled AI to surpass human top scores in IOI 2026.

Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi et al.

Introduces UE5M3 block scaling with periodic tensor scaling, enabling stable FP4 pretraining of a 8B model with 190B tokens, outperforming NVFP4 in loss metrics.

Robert Hu, Carlo Luschi, Paul Balanca

H3DNAS compresses 3D point cloud models on ONNX, reducing parameters by 65.5% and speeding up inference by 1.99x.

Anchit Mulye, Rhythm Baghel, Sujay Kumar Ingle et al.

DynG-Diff framework enhances time series forecasting accuracy and robustness through dynamic guidance diffusion model.

Zhente Zhang, Zhengwei Ni, Wei Fan

SPD uses the Hungarian algorithm for single-pass decoding, achieving a 64x speed-up.

Emil Laftchiev, Prachi Agrawal, Moe Kayali et al.

Using causal mixed-precision intervention, the study finds damage is widespread; global fine-grained quantization outperforms local layer repair by 21-52 points.

Jundong Hu, Shekar Ramachandran

LRNBA architecture employs linear neural bases for parameter-efficient network compression, achieving comparable or superior performance with fewer parameters.

Binshuai Wang

NashDreamer employs centralized world models to efficiently converge to Nash equilibrium in two-player zero-sum imperfect-information games, improving early training sample efficiency by ~30%.

Tomáš Holeček, Viliam Lisý

LatentPress encodes long contexts into continuous soft tokens with 4-16× compression, enabling direct reading without text reconstruction, boosting efficiency.

Zhengze Zhou, Hejian Sang

FCCA method excels in low-resource LLM adaptation, outperforming others by 2.3 points on Qwen2.5-3B.

Wentao Ye, Zhanming Shen, Zhiqing Xiao et al.