LLM-driven evolutionary search enables autonomous wireless algorithm design, outperforming traditional methods in OTFS equalization and pilotless OFDM receivers.

Fayçal Aït Aoudia, Jakob Hoydis, Sebastian Cammerer et al.

LaT uses a pretrained large language model as an external trainer, enhancing multi-task VRP solutions by stage-wise validation analysis, with minimal overhead.

Yang Wang, Ya-Hui Jia, Wei-Neng Chen et al.

ShotPlan introduces learnable planning tokens with FRoPE for frame-level shot control, outperforming existing cinematic video methods.

Su Guo, Guangce Liu, Haosen Yang et al.

ConsiSpace leverages geometric consistency to enhance video spatial reasoning, improving stability and efficiency with a 12.6-point gain.

Ting Huang, Zhenyu Zhang, Wenyuan Huang et al.

The study proposes a one-step lowest-variance selection method using a Gaussian random-field model to analyze total correlation and square root collision threshold.

Linjun Li

GeoWorldAD embeds explicit 3D geometry and future scene prediction, achieving state-of-the-art autonomous driving trajectory planning with PDMS of 91.0 on NAVSIM v1.

Songyan Zhang, Jinyuan Tian, Hanbing Li et al.

Proposed GeAR framework combines geometry grounding and appearance restitution for single painting 3D reconstruction, outperforming baselines on HeriArch dataset.

Qinyu Zhang, Xinda Liu, Yunchen Li et al.

RAMP enhances ad recommendation accuracy via masking and alignment pathways, especially when personalized features are unavailable.

Dairui Liu, Zhongyi Lu, Roger Zhe Li et al.

Multilingual sentence embeddings replace translation in LiRA, recovering responses excluded by translation failures with <1% reliability difference.

Ummugul Bezirhan, Ji Yoon Jung, Matthias von Davier

Proposed Matryoshka Hypencoder with multi-scale Q-Net, reducing parameters by 7× while maintaining performance.

Majd Alkawaas, Sean MacAvaney

TimeLens2 leverages multimodal LLMs with Wasserstein-based rewards for multi-interval video grounding, outperforming models up to 397B parameters.

Yuhan Zhu, Changlian Ma, Xiangyu Zeng et al.

EditCLEVR: A benchmark for evaluating object-centric representations under semantic edits.

Anuraag Gadehothur Karnam, Tarunesh Sathish

CoDID achieves disentangled learning under hidden correlations via iterative mode discovery and meta-optimization, improving accuracy by 7.8%.

Rong Hu, Ling Chen

This study systematically evaluates trajectory data filtering for LoRA fine-tuning of code agents, revealing data quantity outweighs quality at small scales.

Yunze Han

Intervention-centered auto research with independent axis search and held-out transfer validation improves decision attribution and generalization in materials ML, confirming 89.3% of choices across 10 benchmarks.

Jingjie Ning, Xiaochuan Li, Shanshan Zhong et al.

ParaCAD employs a surface-centric autoregressive model to directly generate native parametric B-Rep surfaces conditioned on point clouds, improving geometric accuracy.

Dafei Qin, Rui Xu, Zeyu Shen et al.

WHALE model integrates Wukong and HSTU architectures to enhance performance in large-scale recommendations.

Renqin Cai, Dawei Sun, Yuanjun Yao et al.

Proves weak NP-completeness of Prim-Dijkstra VLSI routing; introduces HP-RCRST with (2,2) balance guarantee, outperforming existing methods.

Keren Zhu

Proposed PSDPO method improves physical plausibility by 2x on VideoPhy-2 while maintaining semantic consistency on VBench.

Siwei Meng, Yawei Luo, Shu Zhang et al.

Proposed CCS framework disrupts both understanding and generation branches in UMMs, reducing facial identity preservation from high to zero, with superior robustness.

Weiwei Tan, Junxian Li, Rui Wang et al.