Intrinsic Distribution Matching
Following prior work, we estimate this difference with internal velocity branches, avoiding a separate fake-score network.
Deyuan Liu, Yansen Han, Kaiwen Zheng and 7 more239Upvotes
AI papers on 31 Oct 2026
The list as it stood that day: the papers researchers were upvoting on Hugging Face's daily papers. Papers stay on arXiv.
Updated 31 Oct 2026, 06:23 UTCSource: Hugging Face daily papers
01 · Oct 28 02 · Oct 27 A-RSI: Adaptive Task Refresh for Zero-Data Multimodal Self-Improvement
We address these two questions separately: training dynamics determine when to acquire new experience, while the answers to new tasks remain independently checkable.
Yu Liu, Youxing Li, Yuedong Tan and 17 more158Upvotes03 · Oct 20 Your Benchmark Is an Empirical Measure Over Difficulty
In this work, we develop this measure-based view of benchmark evaluation. This perspective allows us to identify and formalize two central failure modes: measure mismatch, where the benchmark difficulty distribution differs from the intended difficulty distribution, and coverage…
Yifan Sun, Naicheng Yu, Jiawen Gong and 2 more143Upvotes04 · Oct 28 Youtu-Parsing-Omni: One Encoder, One Schema, All Modalities
We present Youtu-Parsing-Omni, an omni-modal parser that converts documents, images, audio, and video into structured records grounded in space and time, providing a perception layer for agents that read, locate, and cite evidence across modalities.
Yunhang Shen, Zhangquan Chen, Xiaotian Li and 27 more122Upvotes05 · Oct 26 Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs
Since visual tokens make up 89-97% of this input on the benchmarks we study, most of the decoding computation is spent repeatedly processing the same image.
Donghyeon Cho, Joowon Kim, Seungho Shin and 1 more93Upvotes06 · Oct 28 Proving and Improving the Platonic Representation Hypothesis with JEPAs
We turn the hypothesis into a theorem. First, we prove a Platonic Representation Theorem: Joint-Embedding Predictive Architectures (JEPAs) trained separately on any number of modalities learn the same representation up to one rotation whenever their positive pairs are two…
Randall Balestriero, Yann LeCun54Upvotes07 · Oct 28 Read, Allocate, Route: Role-Coupled Tokenization for Pixel-Space Generative Transformers
We observe that finer detail does not necessarily require more global communication: low-frequency structure benefits from global exchange, while high-frequency detail can be modeled locally.
Jiongfan Yang, Weichen Fan, Haiwen Diao and 2 more36Upvotes08 · Oct 28 L3G-World: Real-Time Infinite World Generation with Latent 3D Gaussian Memory
In this paper, we introduce a Latent 3D Gaussian Memory (L3G) that combines the efficiency of feed-forward reconstruction with a compact, rapidly renderable scene representation.
Jinhyeok Choi, JoungBin Lee, Jongmin Lee and 5 more35Upvotes09 · Oct 26 WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training
We propose WorkFactory, a scalable framework for constructing coherent and executable professional tasks.
Ziyang Zeng, Tingyang Chen, Jia Deng and 10 more30Upvotes10 · Oct 22 A Broader Look into MLLM's Retrieval Heads
Locating relevant evidence in long inputs is a core capability of Multimodal Large Language Models.
Jia Wang, Yuqing Li, Jiangnan Li and 2 more28Upvotes11 · Oct 21 VibeGame: Prompt-to-Game Development with AI-Native Engine and Self-Evolving Adversarial Agent Team
We introduce VibeGame, a framework for creating complete 2D game projects from natural-language prompts and optional reference images.
Wenbo Hu, Ken Li, Jiazhe Wei and 13 more20Upvotes12 · Oct 25 Co-Design Bench: Multimodal Collaborative Coding Evaluation for Agentic Models
We introduce Co-Design Bench, a reference-free benchmark of 273 tasks curated from 37,463 deduplicated queries across 26,314 real user coding sessions.
Xilin Wei, Shengyuan Ding, Xinyu Fang and 13 more19Upvotes13 · Oct 28 HairDeck: Animation-Ready Stylized Hair Cards from a Single Image
Observing that a card hairstyle is an unordered set of artist-defined primitives, we propose HairDeck, which generates it as one: each card is a single token, and a set diffusion transformer denoises all cards together.
Yuze He, Yixiang Wei, Wang Zhao and 3 more18Upvotes14 · Oct 27 WeVisCut: The Whole Editing Room in One Agent From Narration to Long-Form Video
We address this gap with WeVisCut, a long-form script-to-video system that connects tool execution with policy learning.
Chenyu Mu, Wenming Weng, Qinrui Wang and 8 more17Upvotes15 · Oct 14 RealCLI: Toward a Unified Evaluation of Human-Agent Collaboration for CLI Agents
In this paper, we present REALCLI, a paired benchmark that unifies the evaluation of task execution and human-agent collaboration quality on realistic CLI tasks.
Yunqi Liu, Peiyu Wang, Bingli Wang and 21 more17Upvotes16 · Oct 15 PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas
We introduce PersonaCUABench, which holds 74 tasks across 11 domains fixed and varies only who asks.
Haoming Li, Defu Cao, Xu Hu and 23 more15Upvotes17 · Oct 27 IntReal: A Real-World Dataset and Benchmark for Multimodal Interleaved Generation
To this end, we present IntReal, comprising two complementary components: IntReal-100K, a real-world dataset of multimodal interleaved samples spanning four major categories, and the IntReal Evaluation Suite, which evaluates model performance across seven complementary…
Shaohan Yu, Jie Qin, Wende Tan and 4 more14Upvotes18 · Oct 28 PhysInsert: Physically Plausible Object Insertion with Geometric Priors and Bidirectional Object-Scene Interactions
We present PhysInsert, a framework for physically plausible reference-based object insertion that jointly adapts objects and their surrounding scenes.
Yue Gong, Le Zhang, Rui Chen and 5 more13Upvotes19 · Oct 27 Multi-Token Prediction Done Right
We identify a flaw in standard MTP training, in which a residual component of the MTP gradient interferes with both next-token and multi-token learning, increasingly so with scale.
Jayden Teoh11Upvotes20 · Oct 26 WebAnything: From the Wild Web to Scalable Agent Environments
We present WebAnything, a fully automated framework that reconstructs source-grounded, reusable Web environments from real URLs without per-site manual implementation or repair after initial setup and authorization.
Tianci Xie, Zikun Cai, Chaoyun Zhang and 4 more11Upvotes