Advances in Agentic Computer Graphics
Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback.
Hengyu Liu, Jihyeon Je, Ruiqi Wang and 11 more122Upvotes
AI papers on 4 Nov 2026
The list as it stood that day: the papers researchers were upvoting on Hugging Face's daily papers. Papers stay on arXiv.
Updated 4 Nov 2026, 06:23 UTCSource: Hugging Face daily papers
01 · Oct 31 02 · Nov 2 Behind Video Generation: Observations and Insights on Multimodal Understanding from Kling
This report summarizes the observations and insights on multimodal understanding gained from successive generations of Kling.
Kling Team, Dong Chen, Jialu Chen and 50 more35Upvotes03 · Nov 1 IdeaTaste: Do LLMs Share the Research Taste of Human Researchers?
We introduce IdeaTaste, a benchmark that scores an LLM by how closely the research proposals it writes from a literature snapshot frozen at 2025 match the work the same communities published in 2026.
Shanshan Zhong, Chen Xu, Jingjie Ning and 3 more34Upvotes04 · Oct 31 SWE-Trail: History-Guided Trajectory Collection for Coding Agent Training
Coding agents solve repository issues through long sequences of exploration, editing, and validation.
Ziming Li, Lu Yu, Jiatan Huang and 3 more34Upvotes05 · Nov 1 PhotoGuider: Composition-Aware Cropping and 360^{circ} View Framing with Explainable Photographic Assessments
We present PhotoGuider, a measure-then-frame method that answers both questions with compact learned scorers over one shared set of composition-aware features.
Yuan-Kang Lee, Kuan-Lin Chen, Chia-Che Chang and 1 more32Upvotes06 · Nov 1 MoviEdit: Long-form Multi-Shot Video Editing Made Easy
We introduce MoviEdit, an end-to-end framework for scalable long-form multi-shot video editing that jointly addresses the challenges of data, modeling, and evaluation.
Suhyeon Lee, Zhengang Li, Soo Ye Kim and 6 more28Upvotes07 · Oct 31 Order-Only On-Policy Distillation
On-policy distillation (OPD) trains a student on its own generations by matching the teacher's token probabilities.
Guangzeng Han, Yu Wu, Xiaolei Huang28Upvotes08 · Nov 1 Axion 1.0: Open Agent Foundation Models, Simplified
We introduce Axion 1.0, a fully open agent foundation model that combines strong and broad agentic capabilities with a remarkably simple training recipe.
Yuqi Jia, Yunjie Tian, Jiayu Zheng and 4 more26Upvotes09 · Oct 30 Words Caught in the Heart: Simulating Person-Specific Expression for Self-Opaque Clients in Psychotherapy
We propose Pandora, a psychotherapy client simulation framework that models how psychological experience becomes situated, individual expression.
Mengxi Xiao, Jingjing Chen, Ziyan Kuang and 12 more24Upvotes10 · Nov 1 RoboColosseum: Evaluating Physical Intelligence at Scale In-The-Wild
We introduce RoboColosseum, a distributed evaluation framework that combines unified policy serving and standardized protocols across a network of locally operated robots in diverse environments.
Tan Sang Nguyen, Yi Ru Wang, Shailesh Shailesh and 24 more23Upvotes11 · Nov 1 When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models?
We present an evaluation framework that separates what frozen representations contain, how they organize appearance and physical differences, and what changes a model's own prediction error.
Anton Hristov, Aleksandar Slavov, Vasil Vasilev23Upvotes12 · Nov 2 YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction
Unified Multimodal Models (UMMs) integrate visual understanding, generation, and editing within a single unified framework.
Shengyu Wang, Tengyue Nong, Junming Liu and 10 more22Upvotes13 · Nov 2 History Forcing: Distilling History-Dependent Continuations for Long-Horizon Video World Models
To address this gap, we introduce History Forcing, a unified on-policy distillation framework that strengthens historical supervision for few-step autoregressive video generators.
Hangyi Kuang, Chengzhi Wang, Yefei He and 8 more14Upvotes14 · Nov 1 SpecPilot: Autonomous Algorithm Discovery and Systems Optimization for Speculative Decoding
We identify three challenges. First, with one request, each step accepts few tokens but reads every target weight.
Kunjun Li, Wenhao Chai, Yihao Liang and 7 more14Upvotes15 · Nov 1 Recursa: Toward Recursively Self-Improving Foundation Model Systems
We introduce Recursa, a foundation model system that adapts to new domains by learning from its own experience without human experts or a stronger teacher model.
Erkang Zhu, Yin Lin, Xiaorong Shi and 2 more9Upvotes16 · Nov 1 LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents
We present LeanGuard, a runtime that checks every proposed tool call against a policy written in Lean4 before the call reaches the environment.
Sreehari Sankar, Debargha Ganguly, Vikash Singh and 17 more9Upvotes17 · Oct 31 ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost
We introduce ParetoSaddler, which addresses this mismatch by separating intermediate candidate promotion from final deployment selection.
Xingzhi Qian, Sungho Park, Rongyuan Tan and 13 more9Upvotes18 · Nov 2 PrimeSim: A Generalist World Model Simulator from Large-Scale Cross-Embodiment Demonstrations
We present PrimeSim, an action-conditioned, multiview world model simulator that addresses these challenges through progressive midtraining and posttraining.
Travis Davies, Wentao Ma, Lihao Guo and 10 more8Upvotes19 · Oct 31 Can AI Reviewers See Your Efforts to Improve Your Paper? A Large-Scale Study of Paper Revision Trajectories
We investigate this question using PaperTrajectory-2026, the 2026 release of PaperTrajectory, a dataset of 6,959 paper trajectories across five AI conferences, with automatic annotations of changes along 14 dimensions and separate judgments of overall quality.
Yaochen Wang, Chenguang Wang, Ming Li and 8 more8Upvotes20 · Oct 29 MoLRSI: A Recursive Self-Improving AI agent for Molecular Discovery
Here we introduce MoLRSI, an AI agent that combines systematic optimization of workflows and model architectures with persistent cross-task memory to recursively improve its search policy for molecular discovery.
Yikun Zhang, Xiwei Cheng, Mingze Yuan and 15 more7Upvotes