AlphaGRPO enhances UMMs' multimodal generation via Decompositional Verifiable Reward, significantly improving benchmarks like GenEval.

Runhui Huang, Jie Wu, Rui Yang et al.

LongMemEval-V2 achieves 72.5% accuracy with AgentRunbook-C, evaluating long-term memory in agents.

Di Wu, Zixiang Ji, Asmi Kawatkar et al.

Pion optimizer preserves spectrum via orthogonal equivalence transformation, enhancing LLM training stability.

Kexuan Shi, Hanxuan Li, Zeju Qiu et al.

Task-Adaptive Embedding Refinement via Test-time LLM Guidance improves zero-shot search and classification by up to 25%.

Ariel Gera, Shir Ashury-Tahan, Gal Bloch et al.

Proposes a sparse-to-dense reward principle combining GRPO and OPD to enhance language model post-training.

Yuanda Xu, Hejian Sang, Zhengze Zhou et al.

ToolCUA optimizes GUI-Tool path selection via staged training, achieving 46.85% accuracy.

Xuhao Hu, Xi Zhang, Haiyang Xu et al.

OmniNFT enhances audio-video generation quality and synchronization through a modality-aware online diffusion RL framework.

Guohui Zhang, XiaoXiao Ma, Jie Huang et al.

MEME evaluates multi-entity and evolving memory tasks, exposing dependency reasoning failures in current systems.

Seokwon Jung, Alexander Rubinstein, Arnas Uselis et al.

The paper introduces a parameter-free online K-Means router leveraging geometric coupling for effective expert assignment, reducing load imbalance with only a slight perplexity increase.

Sagi Ahrac, Noya Hochwald, Mor Geva

The study proposes a framework to diagnose reward hacking in rubric-based RL, finding that even strong verification does not eliminate reward hacking.

Anas Mahmoud, MohammadHossein Rezaei, Zihao Wang et al.

KV-Fold: A training-free protocol for long-context inference achieving 100% exact-match retrieval.

Alireza Nadali, Patrick Cooper, Ashutosh Trivedi et al.

Attractor Models enhance language modeling and reasoning via fixed-point solving, improving training efficiency by 46.6% and accuracy by 19.7%.

Jacob Fein-Ashley, Paria Rashidinejad

DR-Gym environment optimizes electric utility demand response using reinforcement learning, enhancing grid flexibility and energy affordability.

Jose E. Aguilar Escamilla, Lingdong Zhou, Xiangqi Zhu et al.

Multi-stream LLMs unlock language models with parallel streams of thoughts, inputs, and outputs, enhancing efficiency and security.

Guinan Su, Yanwu Yang, Xueyan Li et al.

TextSeal uses dual-key generation and entropy-weighted scoring to watermark LLMs, enhancing detection strength without distortion.

Tom Sander, Hongyan Chang, Tomáš Souček et al.

Proposed AI/ML-based 6G mobility solution using real datasets to optimize handover and beam management.

Mannam Veera Narayana, Rohit Singh, Deepa M. R et al.

Using a Computational Social Science framework, audit LLM-generated political discourse across nine crisis events, finding it more negative and structurally consistent.

Gunjan, Sidahmed Benabderrahmane, Talal Rahwan

FuTCR framework improves new-class panoptic quality by up to 28% in continual panoptic segmentation while enhancing base-class performance.

Nicholas Ikechukwu, Keanu Nichols, Deepti Ghadiyaram et al.

Proposes a synchronized multi-view 3D Gaussian Splatting framework for efficient dynamic scene reconstruction without temporal coupling, validated on a new benchmark.

Yunxiao Zhang, Suryansh Kumar

Proposed a model-based bootstrap method for finite controlled Markov chains, improving confidence interval coverage.

Ziwei Su, Imon Banerjee, Diego Klabjan