MILD integrates bidirectional perception and multi-layered alignment with ECPO optimization for human-vehicle collaboration.

Jiyao Wang, Yunbiao Wang, Yubo Jiao et al.

Proposed DATR framework with multi-turn query fusion significantly improves health video retrieval, achieving R@1 of 19.5% on MHVRC.

Chengzheng Wu, Ke Qiu, Baoming Zhang et al.

SBCA framework uses cross-modal BERT and Actor-Critic for multi-asset portfolio optimization, enhancing annual return and Sharpe ratio.

Jinfeng Pan, Jiahao Chen

EPIC replaces textual demonstrations with embeddings, boosting LLMs as text encoders with reduced token overhead, achieving state-of-the-art on MTEB.

Ailiang Lin, Zhuoyun Li, Keyu Mao et al.

DiagramNet employs multi-stage training and a decoupled multi-agent workflow to achieve superior system-level diagram recognition, outperforming GPT-5 and industry benchmarks.

Jincheng Lou, Ruohan Xu, Jiapeng Li et al.

Faithful-Agent framework boosts Trap SR from 13.88% to 80.21% using Guided Advantage Estimator.

Haowen Hu, Pengzhou Cheng, Zheng Wu et al.

VLA-ATTC reduces failure rate by over 50% on LIBERO-LONG using a Relative Action Critic model.

Wenhao Li, Xiu Su, Yichao Cao et al.

AnimationDiff combines overlay, grid, and diff views to optimize 3D animation comparison, achieving 30% efficiency improvement in user studies.

Ludwig Sidenmark, Qian Zhou, George Fitzmaurice et al.

MathArena extends to a dynamic platform, achieving 98% on 2026 USAMO and 74% on research problems, tracking LLM progress in mathematical reasoning.

Jasper Dekoninck, Nikola Jovanović, Tim Gehrunger et al.

GUI-SD introduces on-policy self-distillation with visual privileged context and entropy-guided loss, boosting GUI grounding accuracy to 68.4% and training speed 4× faster.

Yan Zhang, Daiqing Wu, Huawen Shen et al.

Proposes a denoising-centric IR framework emphasizing signal-to-noise ratio optimization for LLM fidelity.

Lu Dai, Liang Sun, Fanpu Cao et al.

RouteProfile uses graph-based profiling to enable cold-start LLM routing, achieving ~6% improvement in training-free scenarios.

Jingjun Xu, Hongji Pu, Tao Feng et al.

Proposes Policy Gradient Penalty (PGP) for constrained exploration, achieving ε-approximate optimality and feasibility via quadratic regularization.

Florian Wolf, Ilyas Fatkhullin, Niao He

MeshReGen framework regenerates 3D objects from 2D images and initial 3D shapes using VecSet mechanism, enhancing detail consistency.

Geon Yeong Park, Roman Shapovalov, Rakesh Ranjan et al.

DriftBench evaluates constraint adherence in multi-turn LLM ideation, revealing complexity inflation and constraint violations.

Garvin Kruthof

This paper introduces a reinforcement learning framework for GUI agents, categorized into offline, online, and hybrid strategies, emphasizing reward engineering and data efficiency.

Junan Hu, Jian Liu, Jingxiang Lai et al.

This survey systematically analyzes robot world models, emphasizing their role as predictive tools for environment evolution, covering architectures, applications, and future challenges.

Bohan Hou, Gen Li, Jindou Jia et al.

Proposes an agentic RL framework for LLMs integrating meta-reasoning, multi-step planning, and external tools, achieving 50% improvement on complex tasks.

Fangming Cui, Ruixiao Zhu, Cheng Fang et al.

Motubrain uses UniDiffuser and Mixture-of-Transformers for robot control, achieving 95.8% success rate.

Motubrain Team, Chendong Xiang, Fan Bao et al.

Compared Google Search, AI Overviews, and Gemini; found AIO generated in 51.5% of real queries, sources differ significantly.

Riley Grossman, Songjiang Liu, Michael K. Chen et al.