The list as it stood that day: the papers researchers were upvoting on Hugging Face's daily papers. Papers stay on arXiv.
In this work, we present Route Anything, a unified framework for studying model and policy routing across general-purpose AI systems.
Pengfei Zhou, Yixing Ma, Hexin Wang and 4 moreWe propose Pandora, a psychotherapy client simulation framework that models how psychological experience becomes situated, individual expression.
Mengxi Xiao, Jingjing Chen, Ziyan Kuang and 12 moreWe propose LoCo, a local preference learning method for reasoning compression. LoCo detects candidate overthinking states from the model's own traces, freezes the corresponding reasoning prefixes, regenerates shorter verified continuations from the same states, and trains on the…
Muyao Yuan, Liyan Xu, Weizhan Zhang and 4 moreWe introduce DreamX, which repurposes a pretrained video diffusion model as a universal latent action learner.
Jisoo Kim, Chaehyun Kim, Junsu Kim and 6 moreAlthough recent approaches leveraging Multimodal Large Language Models (MLLMs) have shown significant progress, we argue that the MLLM's compositional understanding remains underutilized.
WonJun Moon, Seongchan Kim, Dahun Kim and 1 moreWe propose ToolSpace, a tool-centric framework that reverses this order. ToolSpace mines data, state, and retry dependencies from adjacent calls in execution traces of real MCP services and open agent corpora, using deterministic rules rather than LLM judgment, and clusters the…
Guanting Dong, Zhichao Sun, Xinyu Yang and 22 moreUnified Multimodal Models (UMMs) integrate visual understanding, generation, and editing within a single unified framework.
Shengyu Wang, Tengyue Nong, Junming Liu and 10 moreTo bridge this gap, we present VidUEU-Agent, an automated pipeline that synthesizes instruction data from raw videos for multimodal understanding, editing, and unified tasks.
Tengju Ru, Weitong Lian, Zecong Tang and 12 moreLLM-based multi-agent systems (MAS) rely on collaboration among specialized agents to solve complex tasks.
Qinuo Cao, Xin He, Dianxuan Fu and 7 moreWe propose WebHarbor, a suite of 153 executable and dockerized website environments that faithfully reconstruct the visual interfaces, content, and functional workflows of real-world web services under deterministic control.
Zhaoyang Wang, Qianhui Wu, Boyu Gou and 54 moreWe address this with GameVerse, a minute-scale gameplay video dataset purpose-built for long-horizon interactive world modeling.
Kang He, Wenshuo Peng, Chuanhao Li and 7 moreHere we introduce MoLRSI, an AI agent that combines systematic optimization of workflows and model architectures with persistent cross-task memory to recursively improve its search policy for molecular discovery.
Yikun Zhang, Xiwei Cheng, Mingze Yuan and 15 moreWe ask whether a coding agent can take over the teleoperator's job. In our DataFarm, the agent receives a data request and one calibration episode of the robot in the target scene and rebuilds the task in simulation: it writes the scene, an expert controller, and a success judge.
Kewei Zhang, Juncheng Ma, Guodong Lv and 11 moreWe introduce NEO-VLA, a structurally encoder-free vision-language-action framework. Images are tokenized by lightweight patch convolutional layers and evolve jointly with language throughout the backbone, allowing a flow-matching action expert to draw from…
Weichen Fan, Haiwen Diao, Penghao Wu and 14 moreWe introduce FrontierPhysics, an end-to-end benchmark with 56 tasks made by 42 physicists based on their own research, covering nine domains across theoretical, computational, and experimental physics.
Bingran You, Wenjun Ke, Ria Rosenauer and 69 moreWe study cross-game streaming commentary, in which a model trained on one game commentates games it has never seen, and find that current commentators transfer their timing but not their knowledge of the game world.
Zixian Gao, Penghao Luo, Xiangcheng Cao and 5 moreIn this paper, we argue that neither of them is sufficient. Instead, every token needs its own teacher, as the teacher preference can vary token by token within the same response.
Yifu Luo, Chuang Ma, Zeyu Chen and 3 moreAs advanced text-to-image (T2I) models have reached a practically usable level of multilingual text generation, we argue that this capability can be transferred to IIMT and present FlowTranslator.
Zhixian Li, Xingsong Ye, Yongkun Du and 6 moreWe introduce Solv3R, a looped transformer that learns to continue solving multi-view geometry from its current state.
Jeong-gi Kwak, Andrea Tagliasacchi, Sho Kagami and 2 moreWe introduce H2RB (Human-To-Robot Bridge), a human-robot co-training framework for adapting pretrained VLA policies from unpaired human and robot demonstrations.
Junhyeong Park, Wook Jung, Taeyoung Kim and 2 more