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。
686 篇论文 · 按点赞排序
Zheng-Hui Huang, Zhixiang Wang, Jiaming Tan +6 authors
A large-scale dynamic dataset derived from AAA games is introduced to improve generative inverse and forward rendering, featuring high-resolution synchronized RGB and G-buffer data alongside a novel VLM-based evaluation method that correlates well with human judgment.
Junyao Gao, Sibo Liu, Jiaxing Li +6 authors
MegaStyle presents a scalable data curation pipeline for creating high-quality, style-consistent datasets using large generative models and proposes style-supervised contrastive learning for effective style representation extraction.
Haozhe Wang, Cong Wei, Weiming Ren +3 authors
Training reward models to generate multi-dimensional critiques improves visual generation through both enhanced reinforcement learning rewards and test-time refinement loops, achieving state-of-the-art performance with reduced training data requirements.
Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao +10 authors
AgentOPSD uses recursive Bayesian belief updates over turn-level log-probability gaps to assign credit in long-horizon agentic reinforcement learning without extra critics or rollouts.
Maksim Savkin, Mikhail Goncharov, Alexander Gambashidze +7 authors
Compact task-specialized language models demonstrate superior performance in multi-hop reasoning and faithfulness compared to larger general-purpose models through a novel training pipeline and structured reasoning traces.
Wanghan Xu, Shuo Li, Tianlin Ye +47 authors
ResearchClawBench evaluates autonomous scientific research capabilities across 40 tasks from 10 domains using expert-curated criteria and reveals current limitations in re-discovery accuracy among AI agents and LLMs.
Zhen Fang, Wenxuan Huang, Yu Zeng +8 authors
Flow-OPD addresses limitations in Flow Matching text-to-image models through a two-stage alignment approach combining on-policy distillation and manifold anchor regularization, achieving significant improvements in generation quality and alignment metrics.
Linhao Yu, Tianmeng Yang, Siyu Ding +8 authors
KnowRL is a knowledge-guided reinforcement learning framework that improves reasoning in language models by optimizing compact, interaction-aware guidance subsets through constrained subset search and addressing pruning interaction paradoxes.
Ruiyan Gong, Yingnan Guo, Junjun Hu +37 authors
ABot-N1 improves visual language navigation by separating reasoning from control through a slow-fast architecture that uses explicit chain-of-thought reasoning and pixel goals to guide continuous waypoint generation, achieving strong results across diverse embodied tasks.
Zhuofeng Li, Dongfu Jiang, Xueguang Ma +7 authors
OpenResearcher presents a reproducible pipeline for training deep research agents using offline search environments and synthesized trajectories, achieving improved accuracy on benchmark tasks.
Shaowen Wang, Ge Zhang, Kairong Luo +6 authors
Looping middle layers in sparse Mixture-of-Experts Transformers improves training efficiency and downstream performance while matching per-token FLOPs, parameters, and cache budgets.
Weinan Dai, Hanlin Wu, Qiying Yu +13 authors
CUDA Agent, a large-scale agentic reinforcement learning system, achieves state-of-the-art performance in CUDA kernel optimization by combining scalable data synthesis, skill-augmented development environment, and reinforcement learning techniques.
Xinyu Tang, Gangqiang Cao, Yurou Liu +13 authors
Scaling zero-reinforcement-learning to 1 trillion parameters improves reasoning quality and reveals emergent cognitive behaviors, supported by optimized training and structured evaluation.
Zelai Xu, Zhexuan Xu, Ruize Zhang +7 authors
Multi-agent systems using reinforcement learning enable parallel information seeking with scalable orchestration, achieving performance comparable to larger single agents.
Jiashu Zhu, Yanhao Zheng, Ruitian Tian +7 authors
A compact 7B native joint audio-video generator uses cross-modal attention, progressive joint training, reinforcement learning with multimodal feedback, and an autoregressive 2K refinement pipeline to produce synchronized high-resolution outputs.
Shuo Yang, Xiaoze Fan, Melissa Pan +8 authors
FreeToken is an edge-native Mixture-of-Experts serving system that dynamically maps computation and model state onto heterogeneous local hardware to run large open-weight models on personal machines.
Songhao Wu, Zhongxin Chen, Yuxuan Liu +3 authors
Text embeddings from large language models are enhanced by EmbedFilter, a linear transformation that reduces the influence of high-frequency tokens and improves semantic representations while enabling dimensionality reduction.
Xiangru Jian, Shravan Nayak, Kevin Qinghong Lin +5 authors
CUA-Suite introduces a large-scale ecosystem of expert video demonstrations and annotations for computer-use agents, providing continuous screen recordings and detailed reasoning annotations to advance desktop automation capabilities.
Jeremías Figueiredo Paschmann, Juan Kaplan, Francisco Nattero +3 authors
Pairwise ranking prompting is reformulated as active learning from noisy comparisons, with improved rankers that enhance ranking quality under call constraints and address position bias through a randomized oracle.
Patrice Bechard, Orlando Marquez Ayala, Emily Chen +5 authors
Simple terminal-based coding agents using programmatic interfaces and foundation models can effectively perform enterprise tasks comparable to or better than complex tool-augmented agents.
Yuling Shi, Chaoxiang Xie, Zhensu Sun +7 authors
Multimodal large language models can effectively understand source code when represented as compressed images, achieving significant token reduction while maintaining or improving performance on code comprehension tasks.
Jiangnan Li, Yuqing Li, Mo Yu +2 authors
RARG integrates relevance scores into direct corpus search to prioritize documents and excerpts, accelerating accurate evidence retrieval for complex questions.
Yunhao Yang, Yuexin Bian, Yunjie Tian +6 authors
Co-RL enables unsupervised reasoning via cooperative multi-agent reinforcement learning with peer-derived rewards, improving performance across text and vision tasks without ground-truth labels.
Xiangru Jian, Hao Xu, Wei Pang +13 authors
FORGE introduces a high-quality multimodal manufacturing dataset with fine-grained domain semantics to evaluate MLLMs on real-world tasks, revealing that domain-specific knowledge rather than visual grounding limits performance, and demonstrating that supervised fine-tuning on structured annotations significantly improves accuracy.
Yanke Zhou, Yiduo Li, Hanlin Tang +6 authors
RTPurbo leverages intrinsic sparsity in full-attention LLMs to achieve efficient long-context inference with minimal training overhead, enabling significant speedups while maintaining near-lossless accuracy.
DreamX Team, Rui Chen, Xiangxiang Chu +7 authors
DreamX-Phi 1.0 is an action-conditioned video world model for robotic manipulation that uses geometric attention encoding, depth estimation, object masks with a frozen teacher, and distillation to generate faithful future observations.
Liliana Hotsko, Yinxi Li, Yuntian Deng +1 authors
Code2LoRA is a hypernetwork framework that generates repository-specific LoRA adapters for code language models, supporting both static and evolving codebases with efficient parameter-efficient fine-tuning.
Jinghui Lu, Jiayi Guan, Zhijian Huang +47 authors
OneVL presents a unified vision-language-action framework that improves latent chain-of-thought reasoning for autonomous driving by integrating language and visual world model supervision for faster, more accurate trajectory prediction.
Jian Zhang, Zhiyuan Wang, Zhangqi Wang +7 authors
MAXS is a meta-adaptive reasoning framework for LLM agents that improves multi-tool reasoning through lookahead strategies and trajectory convergence mechanisms, balancing global effectiveness and computational efficiency.
Min Zhao, Hongzhou Zhu, Kaiwen Zheng +6 authors
A novel causal consistency distillation method enables efficient frame-wise video generation with reduced latency and improved quality compared to existing chunk-wise approaches.
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