Paper Archive

Ilay Kamai, Hugues Van Assel, Aviv Regev, Hagai B. Perets, Randall Balestriero 6/9/2026 arxiv

Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each succeeds, when each fails, and when cross-modal training helps at all -- a gap that leaves practitioners, especially in ...

Tong Xie, Yuanhao Ban, Yunqi Hong, Sohyun An, Yihang Chen, Cho-Jui Hsieh 6/9/2026 arxiv

Supervised fine-tuning (SFT) typically maximizes the likelihood of every token in a demonstrated trajectory. However, an observed token can be non-unique, noisy, or misaligned with the model prior. Strictly fitting toward this one-hot target may be suboptimal, especially when the pretrained model en...

Junke Wang, Xiao Wang, Jiacheng Pan, Xuefeng Hu, Feng Li, Jingxiang Sun, Chaorui Deng, Zilong Chen, Yunpeng Chen, Kaibin Tian, Matthew Gwilliam, Hao Chen, Danhui Guan, Kun Xu, Weilin Huang, Zuxuan Wu, Haoqi Fan, Yu-Gang Jiang, Zhenheng Yang 6/9/2026 arxiv

This paper introduces ARM, a discrete representation-based AutoRegressive Model that unifies image understanding, generation, and editing within a next-token prediction framework. ARM is built on three efforts: first, we train a discrete semantic visual tokenizer that maps images into compact token ...

Gangwei Xu, Qihang Zhang, Jiaming Zhou, Xing Zhu, Yujun Shen, Xin Yang, Yinghao Xu 6/9/2026 arxiv

Autoregressive video generation has emerged as a powerful paradigm for World Action Models (WAMs). However, existing approaches suffer from slow training convergence and limited converged accuracy, particularly at high frame rates, as the training supervision is confined to the current chunk without...

Hangfeng Liang, Yutao Hu, Yanhan Hu, Xiaohan Wu, Wenqi Shao, Ying Fu 6/9/2026 arxiv

Low-light video enhancement (LLVE) remains a challenging task due to severe information degradation under low-illumination conditions. Recent multimodal approaches have significantly improved enhancement performance by incorporating auxiliary modalities, such as event streams and infrared images. Ho...

Yujie Zang, Yuhang Zheng, Xian Nie, Yupeng Zheng, Shuai Tian, Songen Gu, Chen Gao, Zining Wang, Shuicheng Yan, Wenchao Ding 6/9/2026 arxiv

Contact-rich manipulation requires robots to continuously perceive and regulate evolving physical interactions under dynamic contact transitions or complex surface geometries. Recent imitation learning methods improve contact-aware control by incorporating tactile or force feedback, but they rarely ...

Weixian Xu, Shilong Liu, Mengdi Wang 6/9/2026 arxiv

In this paper, we propose EEVEE, the first multi-dataset test-time prompt learning framework for LLM agents, enabling test-time prompt learning under real-world task streams. Existing methods are largely designed for single-dataset settings, while real-world applications require models to handle het...

Paul Hyunbin Cho, Jinhyuk Jang, SeokYoung Lee, Joungbin Lee, Siyoon Jin, Heeseong Shin, Jung Yi, Yunjin Park, Chulmin Park, Seungryong Kim 6/9/2026 arxiv

Diffusion-based lip synchronization models achieve strong visual quality and audio-visual alignment, but full-sequence bidirectional attention and many denoising steps make them impractical for real-time inference. We present Lip Forcing, to our knowledge the first autoregressive diffusion method fo...

Kevin Qinghong Lin, Batu EI, Yuhong Shi, Pan Lu, Philip Torr, James Zou 6/9/2026 arxiv

Data tells stories that shape society; the data journalist's job is to turn raw information into stories non-experts can trust. A high-quality news feature takes a newsroom team weeks: hunting for context, running statistics, choosing an angle, and designing visuals. Recent agents handle individual ...

Semih Kara, Oğuzhan Ersoy 6/9/2026 arxiv

Conditioning a language model on additional context, such as feedback on a previous attempt, typically improves its response. Self-distillation trains the model to retain this improvement when the context is not present. The method works by matching the model's output distribution under two settings...