Paper 2509.01106
Robix: A Unified Model for Robot Interaction, Reasoning and Planning
- Published
- Sep 2025
- Research lab
- Independent
- Citations
- 21
- GitHub
- Not linked
01 In brief
Summary
Robix is a unified vision-language model that integrates robot reasoning, task planning, and natural language interaction within a single architecture, serving as the high-level cognitive layer in a hierarchical robot system.
It generates atomic commands for a low-level controller and verbal responses for humans, enabling complex instruction following, long-horizon planning, and natural interaction.
Robix introduces capabilities like proactive dialogue, real-time interruption handling, and context-aware commonsense reasoning.
It uses chain-of-thought reasoning and a three-stage training strategy: continued pretraining on embodied reasoning data (3D spatial understanding, visual grounding, task-centric reasoning), supervised finetuning on synthesized human-robot interaction trajectories, and reinforcement learning with GRPO to improve reasoning-action consistency.
Experiments show Robix outperforms baselines like GPT-4o and Gemini 2.5 Pro on interactive task benchmarks and real-world scenarios (table bussing, grocery shopping, dietary filtering), with Robix-32B exceeding Gemini-2.5-Pro by 3.0 and 11.8 percentage points on OOD settings and by 1.6-4.3 points on task progress in online evaluations.
02 From the paper
Abstract
We introduce Robix, a unified model that integrates robot reasoning, task planning, and natural language interaction within a single vision-language architecture. Acting as the high-level cognitive layer in a hierarchical robot system, Robix dynamically generates atomic commands for the low-level controller and verbal responses for human interaction, enabling robots to follow complex instructions, plan long-horizon tasks, and interact naturally with human within an end-to-end framework. Robix further introduces novel capabilities such as proactive dialogue, real-time interruption handling, and context-aware commonsense reasoning during task execution. At its core, Robix leverages chain-of-thought reasoning and adopts a three-stage training strategy: (1) continued pretraining to enhance foundational embodied reasoning abilities including 3D spatial understanding, visual grounding, and task-centric reasoning; (2) supervised finetuning to model human-robot interaction and task planning as a unified reasoning-action sequence; and (3) reinforcement learning to improve reasoning-action consistency and long-horizon task coherence. Extensive experiments demonstrate that Robix outperforms both open-source and commercial baselines (e.g., GPT-4o and Gemini 2.5 Pro) in interactive task execution, demonstrating strong generalization across diverse instruction types (e.g., open-ended, multi-stage, constrained, invalid, and interrupted) and various user-involved tasks such as table bussing, grocery shopping, and dietary filtering.