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。
Yihao Wang, Yuheng Ji, Mingyu Cao +54 authors
Orca establishes a unified world latent space through next-state-prediction modeling using multimodal data and demonstrates superior performance in downstream tasks compared to specialized baselines.
50 篇论文 · 按点赞排序
Wentao Zhang, Liliana Hotsko, Woojeong Kim +3 authors
Fuzzy-function programming compiles natural-language specifications into compact neural artifacts using a 4B compiler and 0.6B interpreter, achieving efficient, local execution with reduced memory usage and faster inference.
Han Luo, Bingbing Wen, Lucy Lu Wang
Agentic abstention involves determining when an AI agent should cease interaction under uncertainty, requiring sequential decision-making across multiple environments and task types.
Wenhao Zeng, Yuling Shi, Xiaodong Gu +10 authors
A Dockerless environment-free agentic patch verifier improves code patch evaluation accuracy and enables effective post-training without execution-based verification costs.
Xinlei Yu, Gen Li, Qingyi Si +13 authors
DOPD addresses privilege illusion in on-policy distillation by dynamically routing token-level supervision between teacher and student policies based on advantage gaps and probabilities, improving capability transfer in large and vision-language models.
Lei Bai, Zongsheng Cao, Yang Chen +47 authors
Agents-A1, a 35B Mixture-of-Experts Agentic Model, achieves trillion-parameter-level performance through long-horizon trajectory scaling and heterogeneous agent ability scaling via a three-stage training approach involving supervised fine-tuning, domain-level teacher models, and multi-teacher distillation.
Xinyu Wang, Chongbo Zhao, Fangneng Zhan +1 authors
A novel streaming video editing framework enables causal, frame-by-frame editing with stable long-horizon preservation and real-time responsiveness through a three-stage distillation pipeline and AR-oriented mask cache.
Nikita Kachaev, Andrey Moskalenko, Matvey Skripkin +10 authors
Act2Answer protocol evaluates embodied vision-language-action models by having agents answer questions through physical actions, revealing knowledge retention and generalization patterns across different semantic categories.
Hao Zhang, Yiming Hu, Yong Wang +3 authors
Speculative decoding with adaptive block size selection improves inference efficiency by predicting optimal block sizes from prefilling representations, achieving significant speedup with minimal overhead.
Xiangchen Cheng, Yunwei Jiang, Jianwen Sun +7 authors
A bounded contract approach for long-horizon LLM agents uses typed retrieval to assemble fresh prompts, enabling isolated analysis of memory components and demonstrating improved performance in complex decision-making tasks.
Fahd Seddik, Fatemeh Fard
An axiomatic evaluation framework reveals systematic failures in latent thought representations of LLMs across multiple reasoning tasks, demonstrating that current representations fail to satisfy fundamental functional axioms consistently across different model architectures.
Yixian Xu, Kaiyuan Gao, Yuxiang Chen +25 authors
A reinforcement learning and on-policy distillation approach enhances the visual quality and instruction-following capabilities of a diffusion model for image generation and editing tasks.
Mijin Yoo, In Cho, Subin Jeon +3 authors
A feed-forward framework decomposes 3D scenes into instance-structured token groups from multi-view images, enabling direct object-level reconstruction, segmentation, and manipulation without 3D annotations.
Peiwen Zhang, Yufan Deng, Shangkun Sun +11 authors
PhysisForcing enhances embodied video generation by enforcing physical consistency through pixel-level trajectory alignment and semantic-level relational alignment losses in a DiT-based framework.
Disen Lan, Jianbin Zheng, Yuxi Ren +5 authors
FlashMorph is an efficient layer selection method that formulates hybrid layer selection as a budget-constrained optimization problem, using morphable models and linearization regularization to improve long-context efficiency in Transformers.
Zhilin Wang, Han Song, Runzhe Zhan +13 authors
Autonomous agents evaluate policy improvement through iterative editing within fixed budgets, revealing that successful policy evolution requires both task-specific mechanisms and feedback-constrained refinement.
Shoufa Chen, Luyuan Wang, Xuan Yang +7 authors
TUA-Bench presents a comprehensive benchmark for evaluating general-purpose terminal-use agents across diverse digital activities and specialized workflows, revealing significant performance gaps among current frontier agents.
Xuanfan Ni, Liyan Xu, Chenyang Lyu +6 authors
ReFreeKV addresses the limitations of threshold-dependent KV cache pruning by introducing a threshold-free approach that adaptively allocates compression budgets while maintaining full-cache performance across diverse datasets and model sizes.
Zhiwei Li, Yong Hu
SkillHone enables continuous evolution of agent skills by maintaining persistent decision histories and incorporating practice feedback for improved performance across research and tool-mediated analysis tasks.
Lennart Purucker, Andrej Tschalzev, Nick Erickson +7 authors
Tabular foundation models show varying performance across different data conditions, with traditional methods still outperforming newer approaches on complex, large-scale datasets.
Young-Jun Lee, Seungone Kim, Minki Kang +5 authors
Evolutionary fine-tuning enables large language models to develop cross-task problem-solving capabilities by learning from search trajectories, demonstrating improved performance on mathematical conjectures and optimization tasks.
Bingxuan Li, Yining Hong, Cheng Qian +6 authors
Current visual world models demonstrate limited generalization beyond common physical interactions, struggling with rare and irregular scenarios despite achieving realism on standard benchmarks.
Yijie Jin, Jiajun Xu, Yuxuan Liu +8 authors
Multi-Block Diffusion Language Models extend single-block diffusion to concurrent block decoding with improved training strategies and optimized decoding algorithms.
Yana Wei, Hongbo Peng, Yanlin Lai +14 authors
PerceptionRubrics presents a rubric-based evaluation framework that identifies gaps between benchmark scores and real-world performance through atomic auditing and gated scoring mechanisms.
Xingyu Zheng, Xianglong Liu, Yifu Ding +4 authors
MrFlow accelerates text-to-image diffusion by combining low-resolution generation with pixel-space super-resolution and noise injection, achieving up to 25x speedup without training or runtime modifications.
Sijin Chen, Kaixuan Jiang, Haixin Shi +6 authors
Human manipulation skills are transferred to robots more effectively by using a bridging action representation based on relative wrist translation in the initial head-camera frame, combined with a vision-language-action model that handles embodiment differences through interleaved action tokens and attention masking.
Zhaoyan Sun, Shan Zhong, Daizhou Wen +10 authors
A comprehensive benchmark named AgenticDataBench is introduced to evaluate data agents across diverse domains with fine-grained task annotations and skill-based coverage metrics.
Youhe Jiang, Haoxu Wang, Haotong Bao +5 authors
TurboServe is a specialized serving system for streaming video generation that addresses session state management and dynamic resource allocation challenges through integrated scheduling, autoscaling, and migration mechanisms.
Hanlin Wang, Hao Ouyang, Qiuyu Wang +10 authors
WorldDirector enables controllable video generation with persistent object memory by decoupling semantic motion planning from visual rendering through LLM coordination of 3D trajectories and camera movements.
Haoqi Yuan, Zhixuan Liang, Anzhe Chen +20 authors
A Vision-Language-Action foundation model for robotic manipulation achieves generalization through unified alignment across representation, motion, and behavior dimensions, enabling large-scale training on diverse data sources.
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