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
Jiayin Chen, Yicheng Xu, Muting Wang
Mi-Ripple reduces digital ripple artifacts in edited images by separating lattice artifacts from texture and applying targeted spectral filtering and reference cleaning.
Suhyeong Park, Junha Jung, Jaewoo Kang
CVRR enforces recurrent hidden-state computation as the required image-conditioned pathway for visual reasoning, preserving model competence while distinguishing latent information from actual predictive use.
Yixin Wan, Tianle Zheng, Kai-Wei Chang
VDiff-Bench evaluates multimodal language models on fine-grained image difference identification, revealing major weaknesses in detecting subtle low-level visual changes.
Hongxiang Zhao, Mutian Xu, Zeyu Jin +3 authors
A two-stage framework predicts sparse gripper keyframes and continuous actions to transfer complex hand demonstrations to robotic grippers while reducing drift via kinematic optimization.
Zili Wang, Zhaopeng Qiu, Yuekai Zhang +2 authors
A system for large-scale online draft co-training accelerates speculative decoding in RL post-training by extending context-parallel attention and adding cross-stage feature transport.
Yinming Huang, Shuyuan Tu, Xi Yan +7 authors
The abstract proposes a four-level framework classifying agentic visual generation by the controller's direct decision scope over generation operations, from fixed support to experience-adaptive control.
Mohammad Mahdi Abootorabi, Armin Saghafian, Ali Bazshoushtari +5 authors
Activation steering vectors in large language models encode theory-aligned human value geometry when derived via distribution-driven methods, with geometric fidelity scaling with model size but declining after instruction tuning.
Ngoc Bui, Tinglin Huang, Rex Ying
Kalman Delta Networks reformulate linear attention as a linear-Gaussian state-space model with Kalman-filter updates to track memory uncertainty, yielding efficient scan-compatible approximations that improve language modeling performance.
Christopher Stith, Hossein Rahmani, Jesse C. Cresswell
Causal foundation models apply pretrained neural networks to estimate causal effects on new datasets via in-context learning without fine-tuning.
Elisei Rykov, Timur Ionov, Nikolay Ivanov +5 authors
Enoki is an open information extraction framework that unifies claim-level verification and span-level hallucination localization through shared relational facts, reducing resource use while improving detection accuracy.
Zhenxuan Fan, Bo Zhang, Yutong Lin +9 authors
RoboSPA is a large-scale robotic manipulation benchmark that evaluates vision-language-action models on fine-grained spatial reasoning and long-horizon procedural planning across progressively harder task variants.
Haojun Zhang, Yi Zou, Min Chen +7 authors
X-AuT progressively prunes audio-encoder layers in speech large language models and restores accuracy via behavioral probes, representation alignment, cross-scale distillation, and LoRA adaptation.
Muyao Niu, Jixuan He, Ruihan Yu +9 authors
Adapting a single-object 3D generative prior to multi-view observations enables scalable compositional mesh reconstruction of densely cluttered scenes with severe occlusion.
Hongbang Yuan, Zhuoran Jin, Yixin Cao
Feedback-Enriched Environments adapt task settings to provide observation-level guidance, improving reinforcement learning stability and exploration for long-horizon agent tasks.
Sizhe Zhao, Haozhe Xie, Weiyu Zhao +5 authors
MaP-WAM improves non-Markovian robotic manipulation by separating memory-grounded planning from plan-conditioned execution, using compact episodic segment records and progress-calibrated action chunks to maintain fixed inference latency.
Nhat-Tan Bui, Varshini Elangovan, Arun Reddy Anugu +7 authors
CoVeR is a training-free spatial token selector that preserves 3D reasoning performance by enforcing exact budgets and full scene coverage across multi-view visual tokens.
Anqi Li, Yuxin Chen, Zhaobo Li +4 authors
TANGO is a vision-language framework that predicts whole-body joint actions for humanoid robots to navigate cluttered indoor environments using only simulated training data.
Ayoub Kirouane, Georgios Giaples, Christos Petrocheilos
Adding Greek to a robot vision-language-action model via machine-translated instructions reveals measurement pitfalls and shows bilingual training improves performance over monolingual baselines, though it remains far below English levels.
Mostafa Elhoushi, Alex Pretko, Nolan Dey +6 authors
Layer dropout improves large language model training efficiency and enables faster inference via early exit and speculative decoding without sacrificing accuracy.
Kun Li, Zexuan Qiu, Tianhua Zhang +2 authors
PARSER decouples parallel chunk reading from iterative reasoning via scatter-gather subagents, improving long-context multi-hop accuracy and reducing latency.
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