Spectral Forcing (SF) introduces a parameter-free, time-dependent 2D-DCT low-pass filter to improve pixel-space diffusion, boosting ImageNet FID by 14.5%.

Weichen Fan, Haiwen Diao, Penghao Wu et al.

MemoryVAM integrates episodic memory into video world models using Recap-Cue, boosting long-horizon robot manipulation success from 5% to 42.5%.

Yuxin Jiang, Chang Yu, Yunuo Chen et al.

Proposes Ling-2.6 and Ring-2.6 models using hybrid linear attention and transfer pretraining, boosting long-context efficiency and reasoning.

Ang Li, Ben Liu, Bin Han et al.

Under fixed inference budgets, vanilla actors often outperform or match augmented methods in web agent tasks, with multi-run variance highlighting evaluation challenges.

Sina Hajimiri, Masih Aminbeidokhti, Jose Dolz et al.

NEXUS uses neural energy fields for physically consistent contact-rich 3D object dynamics, improving long-horizon accuracy.

Qizhen Ying, Guangming Wang, Yangchen Pan et al.

AdaNAGED method achieves efficient gradient-free fine-tuning of large language models, reducing memory overhead.

Dmitriy Bystrov, Daniil Medyakov, Dmitry Bylinkin et al.

MVEB benchmarks 23 tasks across 33 models, revealing diverse strengths and limitations in multi-modal video embeddings.

Adnan El Assadi, Roman Solomatin, Isaac Chung et al.

AC-GPT introduces a simple modification enabling arbitrary conditional evaluation within a single pass, preserving left-to-right prediction.

Yinhan Lu, Eric Elmoznino, Léo Gagnon et al.

This study identifies a small set of attention heads—gaze heads—in VLMs that causally track the current description region, enabling effective inference-time control via attention masks.

Rohit Gandikota, David Bau

Introduces OmniVideo-100K, a large-scale dataset with structured scripts and evidence chains, boosting audio-visual reasoning by up to 20.59%.

Xinyue Cai, Chaoyou Fu, Yi-Fan Zhang et al.

Introducing RATS (Register Attention Transformers), which self-supervisedly discovers part-level structures with N learnable registers, achieving +12 mIoU on five segmentation benchmarks.

Timing Yang, Predrag Neskovic, Jansen Seheult et al.

Instruct-Particulate employs large-scale heterogeneous datasets and instruction-guided neural networks to efficiently predict 3D articulated structures, significantly improving generalization.

Ruining Li, Yuxin Yao, Matt Zhou et al.

WorkflowView leverages LLMs to abstract low-level action sequences into high-level activities with F1=0.90, demonstrating cross-domain generalization.

Gaurav Verma, Scott Counts

This study reveals that self-verification in vision-language models can cause regression when verifier quality is task-specific, supported by variance theorem analysis.

Jianzhe Lin

Proposes a method combining lexical token analysis and multi-word similarity to quantify cultural localization in AI-generated stories, revealing only 9-17% of vocabulary accounts for cultural differences.

Shaily Bhatt, Supriti Vijay, Jeremiah Milbauer et al.

Proposes MoE-RM-SRL, integrating reward machines, safe distance, and sparse gating experts, achieving safe and efficient highway autonomous driving.

Chufei Yan, Zhihao Cui, Yiyan Lv et al.

VISTA method improves GUI grounding accuracy via multi-view comparison, significantly enhancing Qwen3-VL model performance.

Xinyu Qiu, Yunzhu Zhang, Heng Jia et al.

StreamMemBench evaluates memory's ability to support future tasks from egocentric streams, focusing on evidence retention, initial use, feedback integration, and reuse, with detailed metrics.

Guanming Liu, Yuqi Ren, Hansu Gu et al.

Proposes EM-NeSy, framing neurosymbolic learning as EM, enabling flexible, non-differentiable symbolic inference with neural updates.

Annegret Seibt, Luc De Raedt, Giuseppe Marra

Proposes Elastic Queries Reinforcement Learning (EQRL) for self-aware resource scheduling in VLA models, reducing inference costs while maintaining success.

Ge Wang, Xinyu Tan, Xiang Li et al.