#autoregressive models
6 papers · 1 filter
Mitigating Compounding Error via Video Representation Regularization
Taiye Chen, Qi Zhang, Yisen Wang
The paper studies why autoregressive video generation models accumulate errors over time and introduces a lightweight regularization that stabilizes hidden representations, reducin…
DistillAlign: Coordinating Mode Covering and Mode Seeking in Autoregressive Video Distillation
Jiaxing Li, Kai Zou, Cindy Zhou +7
The paper studies autoregressive video distillation, showing that aligning the student model’s mode coverage with the teacher’s distribution improves generation quality and diversi…
PSG: Pair-Space Generation for Efficient Generative Reranking
Chao Feng, Li Ma, Xiancheng Gao +3
The paper introduces Pair-Space Generation (PSG), which generates ordered item pairs instead of single items to speed up generative reranking in recommender systems while preservin…
Local Interaction Autoregressive Model for High Dimension Time Series Data
Jingyang Li, Yang Chen
The paper introduces a Local Interaction Autoregressive (LIAR) framework for forecasting high‑dimensional matrix and tensor time series by exploiting local dependencies, and provid…
BeatEdit: Symbolic Music Generation as Explicit Editing
Haoyu Gu, Lekai Qian, Haowu Zhou +2
BeatEdit introduces a framework for symbolic music generation that edits existing drafts using a beat-grid-anchored encoding, enabling tasks like error correction, accompaniment re…
Active Exploration via Autoregressive Generation of Missing Data
Tiffany Tianhui Cai, Hongseok Namkoong, Daniel Russo +1
The paper proposes using autoregressive sequence models to quantify uncertainty and guide exploration in online decision-making, showing that Bayesian regret can be bounded by offl…