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。
Shengqiong Wu, Meng Luo, Bobo Li +5 authors
This survey unifies video understanding and generation through world modeling, organizing progress around inferring, synthesizing, sharing, and interacting with latent temporal states to emphasize persistent dynamics over clip-level perception.
42 篇论文 · 按点赞排序
Jujia Zhao, Jiakai Tang, Jing Yu +6 authors
UserArena introduces a real-world interactive benchmark with long-horizon shopping trajectories and an evaluation framework, alongside On-Policy Trajectory Distillation to improve simulator fidelity for agentic recommendations.
Zecong Tang, Zian Guan, Yuncheng Guo +3 authors
HiSTEVE is a hierarchical vision-language-action agent that combines long-horizon planning with native keyboard-and-mouse control to perform complex open-ended tasks in Minecraft.
Zimu Wang, Xiaobo Wang, Siyu Ye +11 authors
RSIHub is an open-source framework that standardizes agent-evolution research by providing shared infrastructure for building, comparing, and extending methods with reusable components and consistent evaluation.
Minghao Zhu, Bohan Hou, Kun Xiang +20 authors
RynnUnify is a 16B unified embodied model that integrates visual planning, physical reasoning, and robot control, achieving state-of-the-art results on long-horizon embodied tasks.
Zijian Liu, Chaoyou Fu, Yan Yang +2 authors
Omni-Woodpecker is a training-free framework that reduces cross-modal hallucinations in omni LLMs by identifying modality conflicts and rebalancing hidden representations to suppress misleading priors.
Ammar Ali, Stamatis Lefkimmiatis
TWIG represents neural network weights as shared discrete decision paths, unifying compression methods and enabling efficient fixed-weight circuits with optimized discrete choices.
Zhefan Rao, Meng Chu, Xuan Zhang +80 authors
Dual Exam evaluates multimodal agents and world models separately across diverse domains, revealing uneven performance and no coupled success in pilot tests.
Zixian Gao, Lixin Xiu, Penghao Luo +5 authors
DirCast is a streaming video language model that simultaneously directs live esports broadcasts by selecting cameras and generates real-time commentary using only past context.
Yegor Anichkov, Valery Vinogradov, Andrey Furov +2 authors
TenetBench evaluates whether language models adhere to professional ethical codes by pairing conflicting requests with benign equivalents, revealing frequent unflagged violations across most tested models.
Leon Sick, Lukas Hoyer, Yannick Strümpler +3 authors
Foundation models can self-improve across text and vision tasks by generating synthetic questions and reasoning traces guided only by high-level topics, without external supervision.
Siyang He, Qiqi Wang, Fan Liu +5 authors
DraMaze improves long-form microdrama script generation by using evolutionary tree search guided by trope memory and backward narrative dependency trees to enhance plot diversity and coherence.
Shuyang Cui, Zhi Zhong, Keisuke Toyama +10 authors
Flow matching is investigated for four-source full-band stereo music separation, evaluating prediction targets, flow paths, modeling strategies, and consistency constraints with both signal-level and perceptual metrics.
Liam Dugan, Callum McDougall, Matthew Denton +3 authors
A pipeline using sparse autoencoders extracts interpretable linguistic features from neural detectors, yielding more accurate and explainable AI-generated text classification.
Yuhan Sun, Jiaxin Zhao, Yongdong Luo +6 authors
MultiAgent-OPD jointly trains a hierarchical multi-agent team via on-policy distillation and reinforcement learning, using a frozen teacher for local process supervision to improve end-to-end performance across multiple datasets.
Shirley Anugrah Hayati, Patricia Anugrah Setiani, Ruizi Wang +4 authors
Kolab is a dataset of professional screen recordings that reveals frontier AI agents overestimate collaboration quality and fail to predict expert adoption of contributions in long-horizon workflows.
Jiyuan Tan, Vasilis Syrgkanis, Jose Blanchet
Even a carefully proofread paper can leave small gaps in its proofs or its stated assumptions. We introduce PaperFormalizer, a Lean-based agentic pipeline that helps authors find and close such gaps in their own papers and turn each paper into a Lean-verified version. Starting from the LaTeX source, PaperFormalizer transcribes the paper into a dependency graph of statements, audits each statement in natural language for faithfulness, ambiguity and correctness, and then formalizes and proves the statements in Lean against Mathlib, optional domain libraries and the author's personal toolkit. Only the author knows what a statement is meant to say and how a formalization gap in a result is best fixed. The pipeline therefore stops and asks the author whenever a statement admits more than one reading, presents every repair, in natural language and TeX symbols, as a proposal that the author can accept, decline or redirect, and leaves to the author every decision that could potentially change what the paper claims. The output is an annotated and revised version of the paper, a report of the changes, and Lean code in which each declaration is linked to the statement it formalizes. The author's toolkit grows as more papers are formalized, and authors can share their verified papers so that researchers who build on them can reuse their Lean code. By making verified papers practical to produce, PaperFormalizer aims to make research that rests on mathematical proofs more robust. PaperFormalizer runs inside a coding agent (Claude Code or Codex) and is available at https://github.com/Jiyuan-Tan/PaperFormalizer. Papers with Lean code shared by users are available at https://paperformalizer.org.
Juekai Lin, Honglin Lin, Yuqian Yuan +7 authors
MMVistaReason improves open multimodal reasoning through structured data construction, difficulty-aware distillation, and specialized reinforcement learning integration.
Jiang Wu, Yutong Song, Hongbo Wang +2 authors
DuoLM accelerates personalized recommendation by combining a compact encoder and ranker with optimized inference, achieving high throughput with minimal quality loss.
Ziang Zhang, Wenyou Yi, Zehan Wang +4 authors
Easy-3Dfy separates object generation from scene layout estimation and uses geometric alignment to combine high-quality 3D objects with spatial understanding for improved scene reconstruction.
Zhongwei Xie, Xingsu Li, Jingqiao Yu +8 authors
JockeyBench evaluates repeated event forecasting via Japanese horse racing, finding that reusable predictors improve efficiency but all configurations yield negative average betting returns.
Aaron Zorel, Guy Van den Broeck, Aditya Grover +1 authors
This paper explores trading inference speed for output quality in speculative decoding by jointly adjusting diffusion effort and acceptance leniency to extend the speed-quality Pareto frontier.
Siyuan Ma, Yu Qi, Cheng Ye +8 authors
Researchers introduce a benchmark and training dataset to evaluate multimodal language models as judges of physically valid, instruction-aligned embodied video generation, yielding a fine-tuned 9B judge with high human agreement.
Yen-Ku Liu, Hongjie Chen, Franck Dernoncourt +1 authors
DISC is a learnable framework that calibrates and combines time-series dataset similarity metrics to better predict downstream forecasting transfer performance.
Panagiotis Koromilas, Eleanna Vali, Diogenis Antonopoulos +7 authors
LoCA is a two-stage pipeline that trains co-aligned multimodal image encoders and uses them as high-rank targets to train location and metadata encoders via contrastive learning, outperforming existing methods across diverse geospatial tasks.
Yufeng Liu, Bohao Fan, Juntu Zhao +11 authors
Echo integrates stop and failure signals into a vision-language-action model to guide robot execution and recovery without fine-grained progress labels.
Xin Zhang, Haochen Wang, Beibei Lin +4 authors
Generalized CycleGRPO unifies region understanding and localization via reinforcement learning with self-verified cycle consistency across images and videos, improving grounding and temporal distinctiveness without external judges.
Xihan Ma, Bo Zhang, Boxiong Li +3 authors
HoloBoxing learns humanoid boxing by combining a frozen whole-body controller with a compact planner that predicts body-point targets, using imitation initialization and reinforcement learning to refine interaction behavior with minimal additional training.
Lukas Niehaus, Robin Rawiel, Ulf Krumnack
KANLab is an interactive browser tool for teaching Kolmogorov-Arnold Networks by visualizing B-spline activation curves, gradient descent, and pruning in a compact 2D setting.
Hongbo Sun
A path-following framework on a fixed-point fiber bundle achieves global linear convergence for variational inequalities with continuous mappings over polytopes without monotonicity assumptions.
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