Proposes WLDS system combining factual and logical calibration for multi-domain emergency scenario deduction with 8.5% accuracy boost.

Zhengqing Hu, Dong Chen, Junkun Yuan et al.

CONTRA employs normalizing flows to produce multi-dimensional prediction regions with guaranteed coverage, outperforming geometric methods.

Zhenhan Fang, Aixin Tan, Jian Huang

ZAYA1-VL-8B integrates vision-specific LoRA adapters and bidirectional attention, boosting multimodal performance efficiently.

Hassan Shapourian, Kasra Hejazi, Olabode M. Sule et al.

This paper demonstrates that standard softmax-attention transformers can implement preconditioned Richardson iteration for Gaussian kernel ridge regression with prediction guarantees.

Mingsong Yan, Dongyang Li, Charles Kulick et al.

STARFlow2 integrates language models and normalizing flows for unified multimodal generation, showing strong performance.

Ying Shen, Tianrong Chen, Yuan Gao et al.

Enhanced monocular depth prediction using distance transform over pre-semantic contours, significantly improving performance in low-texture areas.

Marwane Hariat, Antoine Manzanera, David Filliat

Proposes a multi-agent model combining active inference and symbol evolution to simulate social emotion reality emergence, showing convergence of preferences and dynamic symbol meanings.

Kentaro Nomura, Yushi Tsubamoto, Takato Horii

SOD uses step-wise reweighted OPD to lift a 0.6B agent to 26.13% on AIME 2025.

Qiyong Zhong, Mao Zheng, Mingyang Song et al.

Proposes a collective predictive coding-based model combining POMDP and MHNG for dynamic latent belief synchronization in parent-infant interaction.

Yushi Tsubamoto, Takato Horii

LiteGUI enhances lightweight GUI agents using Guided On-policy Distillation and Multi-solution Dual-level GRPO, achieving SOTA performance.

Yubin Wu, Zicheng Cai, Liping Ning et al.

SkillSelect-Serve recommends QoS-aware budgeted skill services for LLM agents, improving hit rate to 0.9091.

Jingyuan Zheng, Dongjing Wang, Xin Zhang et al.

ForgeVLA learns VLA models from distributed vision-action pairs using federated learning, without language annotations, significantly improving performance.

Yuhao Zhou, Yunpeng Zhu, Yang Zhou et al.

Proposes ROPD, a rubric-based on-policy distillation method, achieving 10× sample efficiency.

Junfeng Fang, Zhepei Hong, Mao Zheng et al.

RC-aux combines multi-horizon prediction and reachability supervision to improve latent world models for planning, achieving up to 15% success rate gains.

Wenyuan Li, Guang Li, Keisuke Maeda et al.

DiffRetriever enhances retrieval efficiency by leveraging DLM's native masked-position prediction.

Shuai Wang, Yu Yin, Shengyao Zhuang et al.

HyperEyes enhances efficiency in multimodal search, improving accuracy by 9.9% with 5.3x fewer tool calls.

Guankai Li, Jiabin Chen, Yi Xu et al.

By linear feature extraction, uncover latent 3D scene topology in vision-language models, improving spatial tasks by up to 12.1%.

Haoming Wang, Wei Gao

Introduced the first systematic fairness benchmark for SNNs, revealing impacts of bias and hardware constraints on model performance.

Hudi He, Fukun Wang, Zhe Wang et al.

RepoZero reformulates repository generation as behavior reproduction, using automated execution-based validation, revealing current LLM limitations in end-to-end code synthesis.

Zhaoxi Zhang, Yiming Xu, Jiahui Liang et al.

LoopQ introduces loop-aware PTQ for recursive Transformers, boosting W4A4 accuracy by 68.8%.

Rui Fang, Hsi-Wen Chen, Ming-Syan Chen