Author

Heng Ji

h-index 3 View on Semantic Scholar

We have 2 of 7 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

Generalizable VLA Finetuning via Representation Anchoring and Language-Action Alignment

Anchor-Align is proposed, which augments BC with two objectives: Vision-Language Anchoring distills layer-wise representations from a frozen VLM copy to prevent this drift, and Language-Action Alignment converts each action target into a discrete motion-direction label and jointly trains language and action prediction...

Dwip Dalal, Shivansh Patel, Chahit Jain et al. · arXiv.org · 2 citations

Jul 2026

Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation

This work introduces Explorative Modeling, a new paradigm that instead factors the training loop, exploring K candidate matches between model generations and data, and training on the best, so predictions commit to modes rather than blurring them, and establishes XMs as both a new pretraining axis for existing generati...

Alexi Gladstone, Heng Ji, Yilun Du · arXiv.org · 3 citations · ⚡1