PulseAugur
EN
LIVE 11:30:39
中文(ZH) 清华徐梦迪:预训练做再好都不够,机器人需要在部署中学习 | CoRL 2026

Tsinghua University researcher advocates for robots learning during deployment

At CoRL 2026, Tsinghua University Assistant Professor Xu Mengdi presented her research on building adaptable generalist robots for human-centered environments. Her work emphasizes the necessity for robots to learn during deployment, rather than solely relying on pre-training. Key projects include CoVer-VLA, which uses a validator to refine robot actions based on user instructions, and Dex-X, which learns dexterous manipulation from human demonstrations by leveraging simulation to bridge the gap in tactile information. AI

IMPACT Highlights the need for robots to learn adaptively in real-world scenarios, moving beyond static pre-training.

RANK_REASON Research presentation at an academic conference. [lever_c_demoted from research: ic=1 ai=1.0]

Read on 雷峰网 (Leiphone) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Tsinghua University researcher advocates for robots learning during deployment

How we ranked this

Signal score
21 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Research presentation at an academic conference. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. 雷峰网 (Leiphone) TIER_1 中文(ZH) ·

    Tsinghua University's Xu Mengdi: No matter how good pre-training is, robots need to learn during deployment | CoRL 2026

    <section style="text-align: center; margin: 0px 16px; line-height: 1.75em; display: block;"><img class="rich_pages wxw-img" src="https://static.leiphone.com/uploads/new/images/20261118/6afd614fc16c5.jpg?imageMogr2/quality/90" style="width: 100%; display: inline-block; text-align:…