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
Hanyang Wang, Yimo Cai, Weiliang Chen +14 authors
Code-as-World represents physical environments as executable code to enable quantitative reasoning and scalable supervision for vision-language models.
Jiale Kang, Ziyin Yue, Zheng Zhan +2 authors
Normalized Low-Rank Adaptation stabilizes LoRA training by normalizing down-projection matrices, accelerating convergence and improving performance without extra parameters or inference cost.
Xionghao Wu, Yijun Yang, Shiyang Zhou +17 authors
ZimaBlue learns generalizable world action models from large-scale egocentric video via a three-stage curriculum and a slow-fast architecture, substantially improving zero-shot robotic manipulation.
Danze Chen, Zeqing Wang, Ziyue Lin +2 authors
We present H3-World, an efficient framework that turns the 33B MiniMax-H3 video generator into an interactive world model. Our key finding is that, as large video generators become more capable, language is emerging as a natural interface for control. MiniMax-H3, for example, already supports zero-shot control of character behavior and camera motion through natural-language instructions. Building on this, H3-World turns this coarse language interface into precise, temporally grounded world control, without introducing dedicated action modules. Specifically, we represent each action as a structured combination of character and camera instructions, and align them with the corresponding temporal video latents. To make the control temporally precise, we further introduce temporal attention routing, which restricts each instruction to its intended time interval and reduces control leakage across actions. Importantly, H3-World directly reuses the semantic representations learned during large-scale video pretraining and requires only lightweight adaptation. With only 8,000 gameplay samples, 10,000 LoRA optimization steps, and 0.199% trainable parameters, H3-World achieves effective character and camera control while preserving strong generation quality. It also generalizes to unseen scenarios. These results show that the control capabilities emerging in large video generators can be efficiently transformed into interactive world control.
Chin-Yang Lin, Yang-Che Sun, Cheng Sun +5 authors
Scal3R improves long-video online 3D reconstruction by querying multi-reference relative poses with lightweight tokens and pose-graph optimization, reducing drift without retraining the backbone.
Gyouk Chu, Myeongho Jeon, Eunho Yang
J-Zero enables self-improving language models across verifiable and unverifiable domains through adversarial co-evolution of a task generator, solver, and judge using predefined preference pairs.
Yuhan Wang, Zhengxi Lu, Yuchen Yan +6 authors
PaperGym converts scientific papers into training environments by separating research questions from evaluation rubrics, enabling reinforcement learning that improves research planning across multiple model sizes.
Esakkivel Esakkiraja, Denis Akhiyarov, Vikas Yadav +4 authors
StarHarness evolves fixed-weight agent harnesses via stratified task pools and hidden selection to improve enterprise tool-use performance and cross-model transfer.
Kairong Luo, Jiarui Cui, Yaorui Yin +8 authors
A cost-efficient open-source pretraining recipe trains 2B-parameter models on consumer GPUs for under $7K, yielding performance near larger baselines while deriving cost scaling laws and studying data curricula.
Jiajun Shi, Siyuan Tao, Yuhao Wu +18 authors
S³Gym evaluates whether large language model agents can self-test, self-judge, and self-improve through interaction experience across text-based games, revealing that effective self-improvement depends on task structure and experience representation.
Jueun Kim, Sungho Park, Wook-Shin Han
Hi-Q is an evidence-conditioned framework that dynamically refines multi-hop queries into hierarchical trees guided by corpus support signals, improving retrieval and answer accuracy without fixed graphs.
Olga Tsymboi, Dmitrii Stoianov, Ramil Latypov +11 authors
A smaller self-hosted LLM trained with separate GRPO experts merged via SLERP outperforms a much larger baseline on instruction following, function-calling, and internal tasks while serving half of platform traffic at lower cost.
Haechan Kim, Yoonho Lee, Gisang Lee +2 authors
WHALE alternates model weight updates and harness search to jointly optimize agent performance across reasoning tasks.
Jiarong Han, Jincheng Xiong, Yuzhou Liu +6 authors
ABot-Recon achieves stable long-horizon streaming 3D reconstruction by using only local temporal context and frame-independent predictions composed sequentially, reducing drift via a lightweight temporal refiner and composition-aware pose loss.
Chaohao Yuan, Ruifeng Yuan, Zhuoxu Huang +4 authors
Pixel Linguist II improves visual text encoding through variable resolution training, natural image-text grounding, layout-aware rendering, and multilingual curricula, achieving state-of-the-art results and strong compression robustness.
Shaoan Wang, Aocheng Luo, Fei Huang +17 authors
LightNav-0 is a compact generalist navigation model that leverages a pretrained vision-language model’s spatial reasoning via unified pointing tokens and action tokenization to achieve state-of-the-art embodied navigation across diverse tasks and robots.
Shangqing Tu, Daniel Zhang-Li, Yucheng Wang +21 authors
CogEvol is a family of models that generate structured learning artifacts in a single pass using supervised fine-tuning and reinforcement learning with vision-language rewards, achieving high quality with far fewer parameters and lower cost.
Yifan Zhang, Steve Ta, Jasper Zhang +8 authors
Recurrent fast-weight memories and selective state-space models are analyzed as online learning rules under autoregressive semantics, yielding normalized update families with stable renormalization that improve length extrapolation.
Hongyu Qu, Guangming Yao, Ling Xing +7 authors
LatentStream introduces a progressive latent working memory framework that internalizes streaming visual evidence into compact evolving tokens for continuous reasoning.
Zhiqin Yang, Jingwen Fu, Yuhan Liu +16 authors
This work proposes a structured ladder for scaling large reasoning models beyond human supervision by tracing autonomous rewards and self-generated experience, while identifying risks and evaluation dimensions.
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