Explore · 每周精选
发现最受关注的研究论文,追踪研究趋势,订阅感兴趣的期刊与关键词。
50 篇论文 · 按点赞排序
Dingjie Song, Zhiling Yan, Yuxuan Zhang +1 authors
LEV is a looped decision model that increases reasoning depth by iterating its layers without generating tokens, improving accuracy on hard questions while allowing adaptive computation via a stopper.
MOSI. AI
MOSS-Image is an open diffusion transformer that combines high-fidelity synthesis with intent-aware prompt enhancement to generate text-rich, layout-coherent images, achieving state-of-the-art open-weight results on long-form text and layout benchmarks.
Zhaoyang Wang, Qianhui Wu, Haorui Wang +9 authors
WildAWM synthesizes controllable computer-use environments and tasks to train general agents via interaction, improving out-of-distribution performance across web, CLI, and GUI benchmarks.
Peisong Wang, Feixue Liu, Yize Meng +16 authors
OR-License evaluates whether LLM agents can solve real-world operations research problems from original client materials, finding that while some achieve high objective quality, none match expert-level feasibility and deployment-adjusted quality.
Yihui Yang, Yuanhong Yu, Jiawei Wang +7 authors
ScaleNav learns goal-directed visual dynamics from internet egocentric videos to generate robot actions for language-conditioned local navigation, achieving high success rates in simulation and real-world deployment.
Zhongxin Chen, Songhao Wu, Heng Cui +3 authors
SpecMRL improves compressed embeddings by reorganizing features via a learned spectral basis that prioritizes informative directions, outperforming standard MRL at reduced dimensions.
Kuan Zhang, Yukun Chen, Zhihao Yang +8 authors
Continued pre-training across data scale, diversity, and model size improves embodied learning in virtual worlds but reveals limits in distant-world generalization, motivating richer training beyond imitation.
Yubo Huang, Yuxuan Zhang, Jiaming Liu +2 authors
Speculative Sampling enables interactive video generation with large bidirectional windows by caching predictions and delivering them incrementally, reducing latency without sacrificing quality.
Meng Luo, Zechen Bai, Shanqing Xu +9 authors
Recursive self-improvement (RSI) describes a system that improves not only its behavior but also the process by which it improves. Recent digital agents test this idea by rewriting their own code and checking each version with executable tests. Bringing these mechanisms into the physical world requires learning from partial observations through costly trials, while accounting for changing operating conditions and potentially irreversible consequences. We position RSI within robot learning and trace pathways from improving robot capabilities to improving the processes that generate them. We examine how physical experience can strengthen these improvement mechanisms and when the resulting gains can support further self-improvement of the mechanisms themselves. Gains carry forward only when feedback addresses remaining learning needs, the receiving component can use the change, its value survives changes in tasks or hardware, and continued practice remains safe and affordable. Existing studies, concentrated in manipulation, support behavioral improvement, benefits to later learning, and procedure improvement. Some systems also change the improver's knowledge or tools, but the physical studies reviewed here have yet to isolate gains attributable to recursion. We synthesize these findings into evaluation principles and research directions for machines that, trial by trial, safely improve the way they improve.
Xialie Zhuang, Mingcong Lei, Zunhai Su +7 authors
DecisionPivot is a diagnostic benchmark that evaluates decision models across controlled variations, revealing gaps between aggregate accuracy, contrast-group performance, calibration, and reasoning effort.
Wenjie Lou, Tianshuo Peng, Shiyang Feng +96 authors
Sage-A1.5 is a 35B mixture-of-experts agentic model that improves long-horizon problem solving via adaptive task delegation, verification, and open scientific exploration.
Qiushi Sun, Jingyang Gong, Yang Liu +14 authors
PrismaForge uses a rubric-based pipeline to synthesize and filter high-quality multimodal code data, enabling open models that rival proprietary systems on visual coding tasks.
Sicheng He, Yi Pan, Qiuting Yu +10 authors
VLA-Rolling improves vision-language-action model efficiency by reusing discarded action tails with adaptive denoising budgets, reducing inference cost and boosting success rates.
Xuanyu Liu, Hongyan Feng, Miao Pan +5 authors
Generating complete, coherent, and usable 3D indoor scenes from natural language requires strong design intelligence: the ability to design high-quality scenes and critique their quality. However, existing approaches either rely on general-purpose models lacking domain-specific design experience or learn from static scenes. In this work, we introduce Scene-Designer, an agentic framework that learns design intelligence through complementary design and critique experiences. We construct two datasets: Scene-OTA-35K, which provides Observation--Thought--Action supervision for learning scene design and efficient error correction, and Scene-OTC-15K, which provides Observation--Thought--Critic supervision for learning scene quality critique across physical validity, functional correctness, and layout coherence. Based on these experiences, we propose Scene-GRPO, a Critic-guided GRPO framework that improves design decisions and reduces unnecessary trial-and-error corrections. Extensive experiments demonstrate that Scene-Designer achieves superior performance in physical validity, functional correctness, and layout coherence, while the learned Scene-Critic significantly outperforms general-purpose models in scene critique. Moreover, test-time Critic guidance further improves scene generation, showing that the Critic can provide useful guidance beyond training. Additional analysis suggests a trend of mutual enhancement between scene design and critique abilities.
Pei Fu, Yike Liu, Yue Sun +12 authors
Xiaomi-Lens-0 is a 4B multimodal model that predicts masks from internal visual representations to improve fine-grained region reasoning and referring comprehension.
Jun Zhao, Leiming Fu, Yanbo Wen +9 authors
LiveMACEBench evaluates persistent LLM agents in live financial markets using decision traces to reveal gaps between outcomes and underlying capabilities.
Chenyang Yan, Yingying Zhang, Kanzhi Cheng +21 authors
OpenMobile-2 is an open mobile agent trained with diverse simulated apps, large-scale supervised and reinforcement learning data, and hybrid GUI-tool use, achieving strong benchmark results and real-device transfer.
Rubin Wei, Siyuan Sun, Hao Sun +5 authors
Deep Fusion Memory integrates predicted parametric memory into intermediate hidden states via gated cross-attention, reducing retrieval overhead and improving knowledge-intensive tasks without online retrieval.
Junha Song, Geonmo Gu, Jeongeun Park +5 authors
LaLa-World is a language-conditioned latent world model that jointly learns from diverse simulation, robot, and video data to enable efficient model-predictive planning and policy initialization.
Heming Zou, Qi Wang, Yixiu Mao +7 authors
TreeOPSD enables long-horizon reinforcement learning for language model agents without critics by using tree-structured self-distillation with bounded updates derived from terminal rewards.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号