ESFP benchmark measures epistemic stance flexibility in large language models under different prompts.

Binwen Liu, Yilin Ren

EcoSpec optimizes MoE model inference by considering expert activation costs, achieving a 1.62x decoding speedup.

Jincheng Xie, Runheng Liu, Heyan Huang et al.

Proposes ASOC framework integrating engineering principles for trustworthy autonomous AI agents.

Amin Beheshti, Rong N. Chang, Boualem Benatallah et al.

LookME introduces hierarchical two-level retrieval and sparse injection to enhance multimodal embeddings in vision-language models, outperforming text-only PLE methods with significant efficiency gains.

Zeyu Xu, Xingzhong Hou, Pengkai Guo et al.

CAtFM enhances flow matching with contrastive learning for style-content disentanglement, improving performance on datasets like ImageNet.

Yusong Li, Pingchuan Ma, Ming Gui et al.

Proposes MESH, a modular retrieval framework that mitigates heterogeneity scaling bias, boosting sparse content scaling by 14× in Pinterest experiments.

Jiaxing Qu, Yilin Chen, Junpeng Hou et al.

SinAE: A single-architecture flow-matching autoencoder significantly reduces reconstruction errors across atomic systems.

Yuxuan Ren, Fan Yang, Jianhua Yao et al.

SeamGen uses flow matching and Mesh Transformer to generate artist-aligned UV seams, outperforming traditional methods with 20% lower deviation.

Hao Xu, Yuqing Zhang, Yiqian Wu et al.

XScientist introduces a git-like protocol for long-term autonomous research, exporting inspectable research artifacts with DAG structure.

Jixiang Luo

SlimPer reformulates personalized ranking as iterative knowledge base refinement, decoupling depth from input length, improving efficiency and interpretability.

Siqi Wang, Xianjie Chen, Shaofeng Deng et al.

SARSI couples a self-model with evidence-gated recursive improvement; this paper is a design proposal, not an empirical result.

Chengshuai Yang

RegHead generates non-humanoid head blendshapes via feed-forward registration, faster than optimization methods.

Jiahao Luo, Hao Zhang, Jianqi Chen et al.

This paper introduces SOAP and Muon optimizers, with algorithmic improvements enabling stable, efficient large-scale LLM pretraining at billion-parameter scales.

Mikail Khona, Aditya Vavre, Boxiang Wang et al.

SymbOmni employs symbolic concept learning for continuous model evolution, boosting generation quality and efficiency.

Jinxiu Liu, Jianru Li, Tanqing Kuang et al.

ABot-3DWorld 0 introduces a unified spatial primitive for multimodal 3D scene generation, surpassing state-of-the-art in fidelity and versatility.

Mingchao Sun, Luyang Tang, Yu Liu et al.

Xiaomi-Robotics-U0 is a 38-billion-parameter multimodal autoregressive model enabling multi-robot, multi-view scene generation and fine-grained embodied transfer.

Xinghang Li, Jun Guo, Qiwei Li et al.

WarpMPC uses ADMM with unrolled LDL⊤ factorization for large-batch GPU MPC, achieving 8,000–250,000 SQP iterations/sec.

Henrik Hose, Se Hwan Jeon, Charles Khazoom et al.

WALA learns executable latent actions from action-labeled demonstrations and action-free videos, achieving 75.2% success on RoboCasa.

Jiahao Liu, Zhongpu Xia, Shuai Tian et al.

RefineEvo employs planning-guided heuristic evolution and bidirectional experience pools to enhance combinatorial optimization.

Yang Wu, Junran Pan, Yifan Zhang et al.

StrideDiffusion accelerates time-series generation with spectral-aware sampling, reducing evaluations to 14-66.

Du Yin, Estrid He, Julián Jerónimo Bañuelos et al.