APT accelerates Diffusion Transformers via attention probability-guided pruning and quantization, achieving up to 8.16× speedup and 14.98× energy efficiency improvement.

Sungyeob Yoo, Seeyeon Kim, Joonyong Park et al.

ProViP enhances VLM inference efficiency via head-aware pruning, retaining 95.9% performance with 1.62x speedup.

Chaofang Ma, Lin Jiang, Carol Jingyi Li et al.

V-Link restores visual cues in Action DiT, improving GR00T N1.6 by 31.2% on LIBERO-Plus and 18.8% on RoboTwin 2.0.

Yehao Lu, Jiarui Yang, Yuning Su et al.

Proposes Bayesian Elastic-Net ERGM combining adaptive regularization with latent variables, improving stability and sparsity in high-dimensional network models.

Dan Han, Vicki Modisette, Ting Li et al.

Proposes MO-IKE, a multi-objective RL method, boosting knowledge editing reliability to 92% and retention to 63.4%, balancing generality and specificity.

Xuzhong Wang, Maiqi Jiang, Tejal Nair et al.

Proposes a control-theoretic approach for resource-aware consensus in multi-agent systems, ensuring consensus without resource exhaustion.

James Flagg, Esteban A. Hernandez-Vargas

Expressed entropy production rate (EPR) as quadratic form of antisymmetric matrices, revealing a universal square-root law in nonreciprocal networks.

Thiparat Chotibut, Ewa Gudowska-Nowak, Maciej A. Nowak

LAWA encodes future intentions as compact latent actions, reducing inference latency by 42.9%, achieving 80.8% success on RoboCasa with superior generalization.

Xiang Li, Yupeng Zheng, Songen Gu et al.

Unified framework models smart glasses as closed-loop systems with eight hardware capability axes, connecting perception, state, and action.

Jiangning Zhang, Haojun Chen, Yong Liu

SPO++ aligns advantage normalization with action-token measure using event-time memory, boosting asynchronous RL efficiency with 19-point reward improvement.

Kai Ruan, Jinghao Lin, Qianshan Wei et al.

Proposes Bellman calibration using monotone transformations to reduce occupancy ratio bias in offline RL.

Lars van der Laan, Nathan Kallus

Introduces a reinforcement learning-based decision confidence model for social credibility, revealing non-monotonic effects on group consensus and error propagation.

Gabriel Bontemps, Abhishek Banerjee

FedV-KGQA enables multi-hop reasoning over vertically partitioned knowledge graphs with near-centralized performance, using local graph enrichment and entity anchoring.

Md Saikat Islam Khan Bappy, Oshani Seneviratne

LAION-BVD is a 10M-hour open video dataset with 1.3B URLs, supporting large-scale multimodal pretraining with content-aware scene detection.

Andreas Hochlehnert, Marianna Nezhurina, Mehdi Cherti et al.

CES-PK framework uses lightweight symbolic constraints with three-valued semantics to verify LLM-generated answers over incomplete knowledge graphs, improving precision.

Emanuel Kitzelmann

Proposes a geometric framework analyzing robustness of neighborhood fairness audits, validated by experiments on benchmark datasets.

Binita Maity

ELR (effective learning rate) governs loss dynamics; matching ELR schedules collapses loss trajectories across configurations.

Zihan Liu, Ruiheng Zheng, Shaobo Zhang et al.

Introduces FrontierChallenge, a cross-domain benchmark for scientific workflow completion, with a top model success rate of only 20.6%.

Liangcai Su, Zhaopeng Feng, Zhuo Chen et al.

StarHarness employs stratified search to evolve environment-specific harnesses, boosting performance by 20-35 points without changing model weights.

Esakkivel Esakkiraja, Denis Akhiyarov, Vikas Yadav et al.

CAFE framework enhances search agents via co-evolving feedback, excelling on seven benchmarks.

Boyang Liu, Senjie Jin, Peixin Wang et al.