Read AI papers faster
with ChatPaper

Bing Yan, Gregory Wolfe, Stefano Martiniani, Kyunghyun Cho · New York University

Chemistry literature synthesis often requires assembling specific findings scattered across many publications, yet existing literature-search systems primarily return ranked…

Hanzhang Zhou, Panrong Tong, Xu Zhang, Quyu Kong et al. · TongyiLab

GUI agents have the potential to become a general purpose executor over existing digital devices. To advance them toward real-world use, we envision agents that operate reliably…

Zeyu Zhang, Ziliang Guo, Yihang Sun, Xichong Zhang et al. · MemTensor

Recent advances in AI agents have increasingly internalized native capabilities into their underlying foundation models, giving rise to multimodal foundation models and large…

Junlin Yang, Che Jiang, Yu Fu, Tianwei Luo et al. · Frontis AI

Recursive self-improvement (RSI) requires AI systems that improve the process of building AI (i.e., AI4AI); machine learning engineering (MLE) offers a concrete, executable…

Shuyao Shang, Yuqi Wang, Ruopeng Gao, Xu Chen et al. · Chinese Academic of Science Institute of Automation

We introduce PhiZero, a physical world model built around physical language, a compact discrete representation of world-state transitions. Existing physical world models typically…

Haodong Li, Tianfei Ren, Xiaoxiao Ma, Chunmei Qing et al.

Text-to-video models have achieved remarkable visual quality, yet they still struggle to generate physically consistent dynamics because the temporal evolution of a scene must be…

Rubin Wei, Jiaqi Cao, Jiarui Wang, Junming Zhang et al.

Decoder-only language models entangle long-term memory and reasoning in a single parameter set, making it difficult to scale memory capacity independently. Memory Decoder…

Qixun Wang, Yang Shi, Letian Cheng, Zhuoran Zhang et al. · Kling Team

The fundamental goal of agentic visual reasoning is to improve the success rate of multimodal large language models (MLLMs) on complex tasks, rather than merely equipping them…

Pengyu Wang, Benfeng Xu, Shaohan Wang, Xin Zeng et al. · muset.ai

Retrieval-augmented generation (RAG) spans lexical and dense retrieval, graph-based indexing, and agentic search, but these paradigms are usually evaluated on different benchmarks…

Yuran Wang, Zekun Wang, Bohan Zeng, Ruixu Zhang et al. · Peking University

Large language model training in open-ended domains lacks verifiable rewards, making task preferences difficult to formalize as effective supervision. Contexts can convey such…

Jiajia Lin, Mingxuan Du, Tuowen Zhou, Benfeng Xu et al. · muset.ai

Text-to-image and personalized editing models now synthesize high-fidelity single-subject images with ease. Yet placing multiple named people into shared contact actions such as…

Yukang Cao, Haozhe Xie, Beichen Wen, Runmao Yao et al.

Embodied intelligence faces a fundamental data bottleneck. Models must capture how first-person perception, whole-body motion, dexterous manipulation, object state, sound, and…