Introduces multi-axis Max@K reinforcement learning to enhance target mode coverage in text-to-image models, significantly improving diversity and fairness.

Ku Onoda, Paavo Parmas, Hiroki Furuta et al.

TanGO enables training-free 3D editing via tangent-space guidance, significantly reducing structural artifacts.

Siwoo Lim, Sunjae Yoon, Gwanhyeong Koo et al.

D-cut improves high-concurrency inference speed by adaptive pruning, achieving up to 3× acceleration.

Tianyu Liu, Yuhao Shen, Rui Cen et al.

Proposes MCPEvol-Bench, using 11 mutation operators to simulate MCP server evolution, evaluating 12 LLMs' adaptability in dynamic tool environments.

Huanxi Liu, Kun Hu, Jiaqi Liao et al.

MathCoPilot integrates human-in-the-loop proof blueprinting, multi-skill orchestration, and automated knowledge base construction, achieving 20/63 success on FormalMATH with natural language prior to formalization.

Junjie Zhang, Jiayu Liu, Wenbin Liu et al.

HyMobileAgent achieves efficient GUI interaction via data-environment co-scaling, 82.6% success on AndroidWorld.

Hy Vision Team, Huawen Shen, Zhengyang Tang et al.

AdaTurn improves low-budget active visual reasoning by explicit budget conditioning and boundary-aware reinforcement learning, boosting accuracy by over 10%.

Susan Liang, Chao Huang, Filippos Bellos et al.

Introduces a continuous-time RL framework to optimize discrete diffusion models, enhancing performance in reasoning and coding tasks.

Zikun Zhang, Jiayuan Sheng, David D. Yao et al.

Uni-AdaVD achieves concept erasure in visual generation via orthogonal value decomposition, showing strong performance across various models.

Qifan Zhou, Yuan Wang, Yanbin Hao et al.

G2SR leverages multi-view geometry with lightweight neural detection to achieve fast (69-89 Hz), low-memory (203MB) Gaussian surface reconstruction, outperforming end-to-end methods.

Dasong Gao, Vivienne Sze, Sertac Karaman

Depth-recurrent Transformer achieves per-token fixed-point convergence, reducing average depth by 38%.

Joe Logan

Proposes pairwise validator as reward substitute in self-evolving agents, matching or surpassing full reward performance with reduced labeling cost.

Minghao Liu, Yu Wang, Jiayun Wang et al.

MEMORA employs a formation-consolidation-retrieval lifecycle with typed stores, improving long-term robot planning by up to 20.5 points in memory assessment and 16.6% in planning scores.

Zihao Yu, Xiu Yuan, Chongjie Zhang

Formalized Shor's algorithm in Lean, providing detailed resource estimates for RSA-2048 and P-256 quantum attacks, enabling machine-verified cryptanalysis.

Lei Zhang, Yusheng Zhao, Hongshun Yao et al.

TRACE assigns turn-level rewards via credit estimation using frozen reference models and TD changes, boosting long-horizon agent performance without extra critic training.

Leitian Tao, Baolin Peng, Wenlin Yao et al.

Trained small language models using progressive SFT and RL to enhance multi-agent retrieval relevance, achieving NDCG@10 of 0.918.

Gayathri V Kondapalli, Alexander Ng, Hirsh Pithadia et al.

HarnessBank achieves agent-harness self-evolution with semantic gene-bank search and gated verification, improving performance by 5.1%-15.4%.

Xiaotian Luo, Dizhan Xue, Fengxingyu Wang et al.

GMoT module explicitly extracts sparse motion cues, boosting micro-gesture video reasoning accuracy to 67.32% Top-1, surpassing baselines.

Taorui Wang, Wei Xia, Hui Ma et al.

MemCon models memory control as an MDP, using lightweight online UCB policy, achieving 15.2% success boost and 20% token savings without extra LLM calls.

Eric Hanchen Jiang, Zhi Zhang, Yuchen Wu et al.

UniPhys employs a unified physical grounding framework with a 40K dataset, achieving state-of-the-art articulation and physical property estimation.

Xian Li, Rong Wei, Lujie Yang et al.