Teahose.
SIGN IN
NEW HERE — WHAT TEAHOSE DOES
We read the entire AI & tech firehose — so you don't have to.
PODPodcastsAll-In, No Priors, Acquired…
NEWNewslettersStratechery, Newcomer…
PAPPapersPhysical AI research
PHProduct Huntdaily launches
VCInvestor ScoutSequoia, a16z, Benchmark…
CLAUDE DISTILLS →
7 reads, 30 sec each — free, 6 AM ET.
+ a live graph of the companies, people & themes underneath.
HOME/PEOPLE/MENGYA LIU
// PERSON

Mengya Liu

ROLE LEAD AUTHOR / RESEARCHERMENTIONS 6LAST SEEN JUNE 5, 2026
// BIO

Lead author of the LARA framework for Vision-Language-Action models.

// RECENT MENTIONS
// SIGNALS
6 SIGNALS
01
product·arXiv Physical AI·JUNE 5, 2026

LARA: Latent Action Representation Alignment for Vision-Language-Action Models

Source
02
mention·arXiv Physical AI·JUNE 5, 2026

LARA's core bet is that you can squeeze more performance out of existing unlabeled video (human or robot) by jointly training a Latent Action Model (LAM) and a diffusion-based VLA policy, rather than treating them as separate sequential steps.

Source
03
mention·arXiv Physical AI·JUNE 5, 2026

For GR00T-N1.6, this yields improvements of +1.3% on SIMPLER-ENV and +5.56% on real-world G1 humanoid tasks (Table 1, Table 2).

Source
04
mention·arXiv Physical AI·JUNE 5, 2026

LARA is applied to π0.5 as a post-training module, achieving +0.9% improvement on LIBERO average.

Source
05
mention·arXiv Physical AI·JUNE 5, 2026

LARA achieves ~30% average improvement when adapting from OXE-pretrained models to entirely new embodiments (Unitree G1 humanoid, GR1-Sim) not seen during pretraining.

Source
06
mention·arXiv Physical AI·JUNE 5, 2026

LARA (full) outperforms the best LAM pseudo-label baseline (Moto-GPT) by +16.8% on SIMPLER-ENV (Table 1) while using only OXE-constrained data.

Source

AI-extracted from podcast / newsletter / paper summaries. May contain errors.