# PaperBridge

> PaperBridge publishes visual, plain-language explanations of important AI research papers in Chinese, English, and Japanese.

## Main language editions

- [中文 AI 论文解读](https://paperbridge.wiki/zh): Chinese visual explainers covering the methods, evidence, limitations, and practical meaning of important AI papers.
- [AI papers explained in English](https://paperbridge.wiki/en): Plain-English research paper explainers with original diagrams, evidence checks, and limitations.
- [日本語の AI 論文解説](https://paperbridge.wiki/ja): Japanese explanations of important AI research papers with diagrams and source-grounded evidence.

## Product and editorial information

- [Editorial method](https://paperbridge.wiki/en/about): Explains how PaperBridge uses primary papers, separates evidence from interpretation, recreates diagrams, and records uncertainty.
- [How to read an AI research paper](https://paperbridge.wiki/en/guides/how-to-read-ai-papers): A practical seven-step guide for non-experts, product managers, and machine-learning practitioners.
- [AI foundations collection](https://paperbridge.wiki/en/collections/ai-foundations): A curated reading path through 20 papers that shaped modern AI.
- [Pricing](https://paperbridge.wiki/en/pricing): Public reading limits, Pro generation credits, supported languages, and additional paper pricing.

## Selected research paper explainers

- [ExtractBench: schema-guided extraction that scores accuracy, completeness, grounding, and cost](https://paperbridge.wiki/paper/arxiv-2607-29677-en-extractbench-a-benchmark-for-schema-guided-enterpris): Enterprise document extraction is not just about getting the right value once. It also has to find every record, point back to the source, and stay affordable when documents get l…
- [SkillOpt: controlled text editing for reusable agent skills](https://paperbridge.wiki/paper/arxiv-2605-23904-en-skillopt-executive-strategy-for-self-evolving-agent-): This paper studies how to improve a frozen agent by editing a separate skill document instead of changing model weights. The core idea is plain: keep the agent fixed, let another…
- [Grasp-MPC: Closed-Loop Visual Grasping via Value-Guided Model Predictive Control](https://paperbridge.wiki/paper/arxiv-2509-06201-en-grasp-mpc-closed-loop-visual-grasping-via-value-guid): Jun Yamada, Adithyavairavan Murali, Ajay Mandlekar, Clemens Eppner, Ingmar Posner, and Balakumar Sundaralingam · ICRA 2026 · arXiv:2509.06201
- [Self-Improving Vision-Language-Action Models with Data Generation via Residual RL](https://paperbridge.wiki/paper/arxiv-2511-00091-en-self-improving-vision-language-action-models-with-da): Wenli Xiao et al. · ICLR 2026 conference paper · Primary paper
- [DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos](https://paperbridge.wiki/paper/arxiv-2602-06949-en-dreamdojo-a-generalist-robot-world-model-from-large-): Shenyuan Gao et al. · ICML 2026 Spotlight · arXiv:2602.06949
- [FLARE: Robot Learning with Implicit World Modeling](https://paperbridge.wiki/paper/arxiv-2505-15659-en-flare-robot-learning-with-implicit-world-modeling): Ruijie Zheng et al. · Conference on Robot Learning (CoRL) 2025 · Paper · Proceedings
- [DreamGen: Unlocking Generalization in Robot Learning through Video World Models](https://paperbridge.wiki/paper/arxiv-2505-12705-en-dreamgen-unlocking-generalization-in-robot-learning-): Joel Jang et al. · Conference on Robot Learning (CoRL) 2025 · Paper · Proceedings
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://paperbridge.wiki/paper/arxiv-2503-14734-en-gr00t-n1-an-open-foundation-model-for-generalist-hum): NVIDIA technical report on arXiv (arXiv:2503.14734, version 2, 27 March 2025). This is not presented as a peer-reviewed conference paper.
- [ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills](https://paperbridge.wiki/paper/arxiv-2502-01143-en-asap-aligning-simulation-and-real-world-physics-for-): Tairan He et al. · arXiv:2502.01143v3 · 2025 · Primary source
- [HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots](https://paperbridge.wiki/paper/arxiv-2410-21229-en-hover-versatile-neural-whole-body-controller-for-hum): Tairan He et al. · arXiv:2410.21229v2 · ICRA 2025 · Primary paper
- [OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning](https://paperbridge.wiki/paper/arxiv-2406-08858-en-omnih2o-universal-and-dexterous-human-to-humanoid-wh): Tairan He, Zhengyi Luo, Xialin He, Wenli Xiao, Chong Zhang, Weinan Zhang, Kris Kitani, Changliu Liu, and Guanya Shi · arXiv:2406.08858 · submitted 13 June 2024 · paper
- [LingBot-World: an open-source video world simulator with long memory and real-time control](https://paperbridge.wiki/paper/arxiv-2601-20540-en-arxiv-2601-20540-plain-english-guide): This paper argues that video generators can be pushed beyond short clips into an interactive world simulator: one that keeps track of what happened earlier, reacts to user actions…
- [ImageNet Classification with Deep Convolutional Neural Networks](https://paperbridge.wiki/paper/en-proceedings-neurips-cc-plain-english-guide-2): Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton · NeurIPS 2012
- [Segment Anything](https://paperbridge.wiki/paper/arxiv-2304-02643-en-arxiv-2304-02643-plain-english-guide): Paper: Alexander Kirillov et al. · 2023 · arXiv:2304.02643 · Full paper
- [Visual Instruction Tuning](https://paperbridge.wiki/paper/arxiv-2304-08485-en-arxiv-2304-08485-plain-english-guide): Authors: Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee Published: NeurIPS 2023 (Oral); arXiv v2, 11 December 2023 Primary source: Paper abstract · Full paper (PDF) Readi…
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://paperbridge.wiki/paper/arxiv-2210-03629-en-arxiv-2210-03629-plain-english-guide): Authors: Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao Published: ICLR 2023 · arXiv v3, 10 March 2023 Primary source: arXiv:2210.03629…

## Key facts

- PaperBridge is a web-based AI research education and paper explanation service.
- Public reports focus on Chinese, English, and Japanese; generation also supports Spanish.
- Reports link to the primary paper and distinguish reported evidence, interpretation, limitations, and uncertainty.
- Reports include original explanatory diagrams and an article or mind-map reading view.
- Users can search with an arXiv page, public paper page, or direct PDF URL.
- The curated AI foundations collection covers 20 influential papers across vision, language, generative models, multimodal systems, and agents.
- Every public report keeps a link to the original paper for verification.
- PaperBridge reports are educational explainers, not peer review or a replacement for the primary paper.

## Canonical site and feeds

- Website: https://paperbridge.wiki/
- RSS: https://paperbridge.wiki/feed.xml
- Complete AI-readable index: https://paperbridge.wiki/llms-full.txt