Preliminary results indicate that a world-action model (WAM) post-trained from OmniDreams achieves strong performance on the Physical AI Autonomous Vehicles NuRec dataset, surpassing the VLA-based Alpamayo 1.5 research policy model while using only 1/5 the total parameters.
This work presents DREAM (Developing Recommender Engine with Agentic Methods), an autonomous optimization control architecture that adds a perception-aware, orchestrable, and auditable policy layer atop existing pipelines without replacing them, supporting agentic meta-control as a viable paradigm for industrial recomm...
Bin Zhang, Bo-Wen Zheng, Chao Yi et al.· 0 citations
Experiments on synthetic and real-world benchmarks show that FG$^2$-GDN and its variant improve associative recall and long-context understanding over GDN and KDA, with comparable computational efficiency.
Ping-Wei Sun, Yuxuan Hu, Jian-Chao Tan et al.· arXiv.org· 1 citation· ⚡1
RA-CLIPScore provides more robust and interpretable evaluations than prior methods, particularly under distribution misalignment or partially irrelevant textual attributes, and aligns more closely with human perception of visual diversity than existing semantic metrics.
Yifan Lu, Taras Kucherenko, H. Kjellström et al.· 0 citations
RecGPT-V3 is presented, a stateful, hybrid-modal recommender that reasons over natural language for open-world knowledge and Semantic IDs (SIDs) for concrete item grounding and achieves consistent gains in large-scale online A/B tests.
Gaussian Mixture Modeling for Event-Aware Visual Allocation is proposed, which leverages Gaussian Mixture Models to model event-level structure from discrete frame-wise observations and achieves comparable performance to baseline selection methods while utilizing only approximately half of the visual token budget.
Yifan Lu, Ziqi Zhang, C. Yuan et al.· arXiv.org· 0 citations
Experiments on the SLAKE and MIMIC-CXR datasets demonstrate that CAST consistently outperforms strong baselines and surpasses decoding strategies reliant on ground truth, and indicates that compact, automatically selected regions provide highly effective contrastive guidance without expert annotations.
Yifan Lu, A. Dukre, Abhijit Das et al.· 0 citations
A synthetic player population is constructed whose traits are ground truth by construction, and an opportunity-aware decision-moment representation is introduced that disentangles preference from the chance to express it; ablating it selectively degrades opportunity-dependent traits.
