Tag
提出 LLMMAD，一种基于大语言模型的自回归多模态异常检测框架，用于微服务系统，通过模态专属重建头与预测驱动目标解决跨模态融合和身份捷径问题，并在三个真实数据集上取得最优 F1 提升。
This paper formulates indoor radio map estimation as a collaborative in-context learning problem and proposes Poly-Ripple, a framework where multiple clients use large autoregressive vision models without model updates, employing ripple-ordered autoregressive modeling and importance-aware aggregation to outperform ICL baselines.
ECHO integrates multi-horizon prediction directly into the backbone via self-recursion and Dynamic Depth Sampling, improving pretraining efficiency and enabling the model to serve as its own inference drafter without auxiliary MTP modules.
本文提出 CQG-Single 与 CQG-AR 两种连续查询范式，消除基于 Semantic ID 的生成式推荐中有限宽度 beam search 带来的可达性约束，并在多个数据集上超越 TIGER 风格基线。
Introduces RosettaSim, a framework that uses structured autoregressive modeling for long-term traffic simulation, achieving state-of-the-art performance on the Waymo Open Sim Agent Challenge. Also proposes Retrieval-based Traffic Evaluation (RTE) for better long-horizon fidelity assessment.
DREAM trains dense retrieval embeddings by using autoregressive language model attention to supervise query-document similarity, eliminating the need for labeled data. It consistently outperforms baselines on BEIR and RTEB benchmarks across model scales.
Discrete autoregressive MRI reconstruction using privileged information distillation achieves superior performance under extreme undersampling by leveraging visual autoregressive modeling techniques.