# LessMimic: Long-Horizon Humanoid Interaction with Unified Distance Field Representations

arXiv: https://arxiv.org/abs/2602.21723 · HTML: https://mlsanity.com/paper/2602.21723/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Feb 2026.

## Built on by (most cited first, 5 of 5)

- [Heracles: Bridging Precise Tracking and Generative Synthesis for General Humanoid Control](https://mlsanity.com/paper/2603.27756/) (2026)
- [SUGAR: A Scalable Human-Video-Driven Generalizable Humanoid Loco-Manipulation Learning Framework](https://mlsanity.com/paper/2605.20373/) (2026)
- [LYRIC: Language-Driven Physics-Based Character Control for Contact-Rich Whole-Body Object Interaction](https://mlsanity.com/paper/2609.19688/) (2026)
- [Counterfactual Video Generation Enables Scalable Humanoid Loco-Manipulation](https://mlsanity.com/paper/2609.38172/) (2026)
- [Being-M0.7: A Latent World-Action Model for Humanoid Robots](https://mlsanity.com/paper/2610.11283/) (2026)

## Abstract (arXiv)

Humanoid robots that autonomously interact with physical environments over extended horizons represent a central goal of embodied intelligence. Existing approaches rely on reference motions or task-specific rewards, tightly coupling policies to particular object geometries and precluding multi-skill generalization within a single framework. A unified interaction representation enabling reference-free inference, geometric generalization, and long-horizon skill composition within one policy remains an open challenge. Here we show that Distance Field (DF) provides such a representation: LessMimic conditions a single whole-body policy on DF-derived geometric cues--surface distances, gradients, and velocity decompositions--removing the need for motion references, with interaction latents encoded via a Variational Auto-Encoder (VAE) and post-trained using Adversarial Interaction Priors (AIP) under Reinforcement Learning (RL). Through DAgger-style distillation that aligns DF latents with egocentric depth features, LessMimic further transfers seamlessly to vision-only deployment without motion capture (MoCap) infrastructure. A single LessMimic policy achieves 80--100% success across object scales from 0.4x to 1.6x on PickUp and SitStand where baselines degrade sharply, attains 62.1% success on 5 task instances trajectories, and remains viable up to 40 sequentially composed tasks. By grounding interaction in local geometry rather than demonstrations, LessMimic offers a scalable path toward humanoid robots that generalize, compose skills, and recover from failures in unstructured environments.