|
pi.website
|
Physical Intelligence
|
Анализировать url
|
|
scholar.google.com
|
<span>Google Scholar</span><svg aria-hidden="true" viewbox="0 0 16 16"><path d="M3 13 13 3m0 0H5m8 0v8"></path></svg>
|
Анализировать url
|
|
github.com
|
<span>GitHub</span><svg aria-hidden="true" viewbox="0 0 16 16"><path d="M3 13 13 3m0 0H5m8 0v8"></path></svg>
|
Анализировать url
|
|
x.com
|
<span>X</span><svg aria-hidden="true" viewbox="0 0 16 16"><path d="M3 13 13 3m0 0H5m8 0v8"></path></svg>
|
Анализировать url
|
|
linkedin.com
|
<span>LinkedIn</span><svg aria-hidden="true" viewbox="0 0 16 16"><path d="M3 13 13 3m0 0H5m8 0v8"></path></svg>
|
Анализировать url
|
|
github.com
|
<div class="highlight-card__visual"><img class="highlight-card__image highlight-card__image--contain" src="/images/projects/vila-u-hd.jpg" alt="The VILA model family connects image, video, and language understanding." loading="lazy" decoding="async"></div><span class="highlight-card__meta">01 · Multimodal perception</span><h3>Understanding & generation</h3><span class="highlight-card__works">VILA² · NVILA · VILA-U · Seed1.5-VL · WorldModelBench</span>
|
Анализировать url
|
|
github.com
|
<div class="highlight-card__visual"><img class="highlight-card__image highlight-card__image--contain" src="/images/projects/ahn-memory.png" alt="Artificial Hippocampus Networks transform a growing context into compact long-term memory." loading="lazy" decoding="async"></div><span class="highlight-card__meta">02 · Memory</span><h3>Long-context processing</h3><span class="highlight-card__works">AHN · LongVILA</span>
|
Анализировать url
|
|
lumine-ai.org
|
<div class="highlight-card__visual"><video class="highlight-card__image highlight-card__image--cover" src="/images/projects/lumine-loop.webm" poster="/images/projects/lumine-poster.jpg" muted="" loop="" playsinline="" preload="none" data-loop-video="" aria-label="Lumine acting in the open world of Genshin Impact."></video></div><span class="highlight-card__meta">03 · Real-time interaction (virtual)</span><h3>Virtual-world agents</h3><span class="highlight-card__works">Lumine</span>
|
Анализировать url
|
|
pi.website
|
<div class="highlight-card__visual"><video class="highlight-card__image highlight-card__image--cover" src="/images/projects/pi07-loop.webm" poster="/images/projects/pi07-poster.jpg" muted="" loop="" playsinline="" preload="none" data-loop-video="" aria-label="A Physical Intelligence robot autonomously folding a shirt with a generated visual subgoal."></video></div><span class="highlight-card__meta">03 · Real-time interaction (physical)</span><h3>Physical-world agents</h3><span class="highlight-card__works">π₀.₇ · π*₀.₆ · EgoVLA</span>
|
Анализировать url
|
|
arxiv.org
|
<div class="highlight-card__visual"><img class="highlight-card__image highlight-card__image--contain" src="/images/projects/experience-distillation-eq4.png" alt="Equation 4, the Experience Policy Distillation loss for learning from agent experience." loading="lazy" decoding="async"></div><span class="highlight-card__meta">04 · Learn from experience</span><h3>Experience distillation</h3><span class="highlight-card__works">Sample-efficient agent learning</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2025</span><span class="publication-list__title">Seed1.5-VL Technical Report</span><span class="publication-list__venue">Technical report</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2025</span><span class="publication-list__title">WorldModelBench: Judging Video Generation Models As World Models<sup>*</sup></span><span class="publication-list__venue">NeurIPS Datasets & Benchmarks</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
openaccess.thecvf.com
|
<span class="publication-list__year">2025</span><span class="publication-list__title">NVILA: Efficient Frontier Visual Language Models</span><span class="publication-list__venue">CVPR</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2025</span><span class="publication-list__title">VILA-U: A Unified Foundation Model Integrating Visual Understanding and Generation</span><span class="publication-list__venue">ICLR</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2024</span><span class="publication-list__title">VILA²: VILA Augmented VILA<sup>*</sup></span><span class="publication-list__venue">ICCVW Oral</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2023</span><span class="publication-list__title">Distilling Large Vision-Language Model with Out-of-Distribution Generalizability<sup>*</sup></span><span class="publication-list__venue">ICCV</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2026</span><span class="publication-list__title">Artificial Hippocampus Networks for Efficient Long-Context Modeling<sup>*</sup></span><span class="publication-list__venue">ICML</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
openreview.net
|
<span class="publication-list__year">2025</span><span class="publication-list__title">LongVILA: Scaling Long-Context Visual Language Models for Long Videos</span><span class="publication-list__venue">ICLR</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2026</span><span class="publication-list__title">π₀.₇: A Steerable Generalist Robotic Foundation Model with Emergent Capabilities</span><span class="publication-list__venue">Technical report</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2026</span><span class="publication-list__title">EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos</span><span class="publication-list__venue">IROS</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2025</span><span class="publication-list__title">π*₀.₆: A VLA That Learns From Experience</span><span class="publication-list__venue">Technical report</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2025</span><span class="publication-list__title">Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds<sup>*</sup></span><span class="publication-list__venue">Technical report</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2026</span><span class="publication-list__title">Sample-Efficient Learning from Agent Experience</span><span class="publication-list__venue">Technical report</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
arxiv.org
|
<span class="publication-list__year">2023</span><span class="publication-list__title">Deductive Verification of Chain-of-Thought Reasoning<sup>*</sup></span><span class="publication-list__venue">NeurIPS</span><span class="publication-list__arrow" aria-hidden="true">↗</span>
|
Анализировать url
|
|
scholar.google.com
|
<span>Complete list on Google Scholar</span><svg aria-hidden="true" viewbox="0 0 16 16"><path d="M3 13 13 3m0 0H5m8 0v8"></path></svg>
|
Анализировать url
|
|
scholar.google.com
|
Google Scholar
|
Анализировать url
|
|
github.com
|
GitHub
|
Анализировать url
|
|
x.com
|
X
|
Анализировать url
|
|
linkedin.com
|
LinkedIn
|
Анализировать url
|