LIBERO-Safety supplies a scalable benchmark, data-generation pipeline, and 19,664-demonstration dataset that exposes a generalization-safety tension in current VLA models where diverse training improves collision avoidance but task success stays limited by trajectory quality and semantic understandi
hub Baseline reference
Robobrain 2.0 technical report
Baseline reference. 60% of citing Pith papers use this work as a benchmark or comparison.
31
Pith papers citing it
Baseline
60% of classified citations
abstract
We introduce RoboBrain 2.0, our latest generation of embodied vision-language foundation models, designed to unify perception, reasoning, and planning for complex embodied tasks in physical environments. It comes in two variants: a lightweight 7B model and a full-scale 32B model, featuring a heterogeneous architecture with a vision encoder and a language model. Despite its compact size, RoboBrain 2.0 achieves strong performance across a wide spectrum of embodied reasoning tasks. On both spatial and temporal benchmarks, the 32B variant achieves leading results, surpassing prior open-source and proprietary models. In particular, it supports key real-world embodied AI capabilities, including spatial understanding (e.g., affordance prediction, spatial referring, trajectory forecasting) and temporal decision-making (e.g., closed-loop interaction, multi-agent long-horizon planning, and scene graph updating). This report details the model architecture, data construction, multi-stage training strategies, infrastructure and practical applications. We hope RoboBrain 2.0 advances embodied AI research and serves as a practical step toward building generalist embodied agents. The code, checkpoint and benchmark are available at https://superrobobrain.github.io.
hub tools
citation-role summary
baseline 3
background 2
citation-polarity summary
representative citing papers
Astra couples an RL-trained VLM policy with a view-consistent Bagel-based world simulator to enable agentic imagination during spatial reasoning, yielding benchmark gains on MMSI-Bench and MindCube.
Backward token warping in ViT-based MLLMs enables reliable reasoning from nearby viewpoints by preserving semantic coherence better than pixel-wise warping or fine-tuning baselines.
RoboCOIN is a large multi-embodiment bimanual manipulation dataset with hierarchical annotations and an open processing pipeline that improves model performance across robotic platforms.
A zero-shot MLLM pipeline for last-mile navigation that selects a view, grounds a task affordance, lifts it to 3D, and predicts a collision-aware base pose, beating MoTo by 3.13 points on OVMM.
VLA skills that score 77-100% in isolation stall from chained states in BEHAVIOR-1K; failures trace to next-skill readiness, target grounding, and control execution.
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
Training a 4B vision-language model to emit identity-tracked, visually grounded reasoning anchors improves embodied spatial, multi-view, and pointing task performance over 7B baselines.
Introduces a new diagnostic benchmark and million-scale reasoning corpus showing that training on explicit causal traces improves next-state prediction in embodied planning, with reported gains from data scaling.
A co-evolutionary VLM-VGM loop on 500 unlabeled images raises planner success by 30 points and simulator success by 48 percent while beating fully supervised baselines.
The work creates NIABench and an LLM-plus-scoring-model framework that enables robots to deliver proactive assistance during human multi-step activities while avoiding interruptions and reducing human effort.
LoHo-Manip enables robust long-horizon robot manipulation by using a receding-horizon VLM manager to output progress-aware subtask sequences and 2D visual traces that condition a VLA executor for automatic replanning.
State-of-the-art vision-language-action models catastrophically fail dynamic embodied reasoning due to lexical-kinematic shortcuts, behavioral inertia, and semantic feature collapse caused by architectural bottlenecks, as shown by the new BeTTER benchmark with real-world validation.
AssemLM fuses SO(3)-equivariant point-cloud features into a VLM to predict discrete 6D assembly poses, reaching ~89% success on AssemBench and improved real-robot multi-step assembly.
GAPL learns a compact set of canonical forgery prototypes and applies two-stage LoRA training to build a low-variance feature space that improves generalization across GAN and diffusion generators.
MiMo-Embodied is a single foundation model that achieves state-of-the-art results on 17 embodied AI benchmarks and 12 autonomous driving benchmarks through multi-stage learning, curated data, and CoT/RL fine-tuning that produces positive cross-domain transfer.
InternVLA-M1 uses spatially guided pre-training on 2.3M examples followed by action post-training to deliver up to 17% gains on robot manipulation benchmarks and 20.6% on unseen objects.
OmniView-Space framework with MPSM, tool-guided reasoning, and distillation achieves SOTA on spatial reasoning benchmarks for MLLMs while reducing external geometry dependencies.
FutureNav proposes a 4B-scale VLM that jointly optimizes action prediction, inverse/forward dynamics, and future state generation for VLN and reports SOTA results on multiple benchmarks.
OmniAct framework integrates planning, memory, and verification to enable persistent autonomy in omnimodal embodied agents, showing improved success and stable context in 40 real-world tasks.
SPARC generates reliable spatial annotations for robot demonstrations by leveraging spatio-temporal task structure, outperforming detection baselines on localization accuracy while retaining more samples and enabling competitive model performance without manual annotations.
Introduces embodied trajectory-coupled data and a three-stage training recipe to bridge VLMs to generalizable VLAs without steep degradation of pre-trained representations.
GR3D is a VLM that combines explicit 2D, implicit 2D, and monocular 3D grounding mechanisms to improve performance on spatial understanding benchmarks.
Introduces EQA-Decision dataset with 4M+ QA pairs across four embodied reasoning dimensions and RoboDecision baseline for joint perception-reasoning-decision evaluation.
citing papers explorer
-
LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
LIBERO-Safety supplies a scalable benchmark, data-generation pipeline, and 19,664-demonstration dataset that exposes a generalization-safety tension in current VLA models where diverse training improves collision avoidance but task success stays limited by trajectory quality and semantic understandi
-
Thinking with Imagination: Agentic Visual Spatial Reasoning with World Simulators
Astra couples an RL-trained VLM policy with a view-consistent Bagel-based world simulator to enable agentic imagination during spatial reasoning, yielding benchmark gains on MMSI-Bench and MindCube.
-
Token Warping Helps MLLMs Look from Nearby Viewpoints
Backward token warping in ViT-based MLLMs enables reliable reasoning from nearby viewpoints by preserving semantic coherence better than pixel-wise warping or fine-tuning baselines.
-
RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation
RoboCOIN is a large multi-embodiment bimanual manipulation dataset with hierarchical annotations and an open processing pipeline that improves model performance across robotic platforms.
-
UniLM-Nav: A Unified Framework for Zero-Shot Last-Mile Navigation
A zero-shot MLLM pipeline for last-mile navigation that selects a view, grounds a task affordance, lifts it to 3D, and predicts a collision-aware base pose, beating MoTo by 3.13 points on OVMM.
-
Diagnosing Semantic Handoff Failures in Agent-Orchestrated Vision-Language-Action Skill Composition
VLA skills that score 77-100% in isolation stall from chained states in BEHAVIOR-1K; failures trace to next-skill readiness, target grounding, and control execution.
-
Vesta: A Generalist Embodied Reasoning Model
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
-
RoboPIN: Grounded Embodied Reasoning via Pinned Chain-of-Thought
Training a 4B vision-language model to emit identity-tracked, visually grounded reasoning anchors improves embodied spatial, multi-view, and pointing task performance over 7B baselines.
-
Token Predictors Are Not Planners: Building Physically Grounded Causal Reasoners
Introduces a new diagnostic benchmark and million-scale reasoning corpus showing that training on explicit causal traces improves next-state prediction in embodied planning, with reported gains from data scaling.
-
RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation with Limited Data
A co-evolutionary VLM-VGM loop on 500 unlabeled images raises planner success by 30 points and simulator success by 48 percent while beating fully supervised baselines.
-
Assistance Without Interruption: A Benchmark and LLM-based Framework for Non-Intrusive Human-Robot Assistance
The work creates NIABench and an LLM-plus-scoring-model framework that enables robots to deliver proactive assistance during human multi-step activities while avoiding interruptions and reducing human effort.
-
Long-Horizon Manipulation via Trace-Conditioned VLA Planning
LoHo-Manip enables robust long-horizon robot manipulation by using a receding-horizon VLM manager to output progress-aware subtask sequences and 2D visual traces that condition a VLA executor for automatic replanning.
-
Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Models
State-of-the-art vision-language-action models catastrophically fail dynamic embodied reasoning due to lexical-kinematic shortcuts, behavioral inertia, and semantic feature collapse caused by architectural bottlenecks, as shown by the new BeTTER benchmark with real-world validation.
-
AssemLM: A Spatial Reasoning Multimodal Large Language Model for Robotic Assembly
AssemLM fuses SO(3)-equivariant point-cloud features into a VLM to predict discrete 6D assembly poses, reaching ~89% success on AssemBench and improved real-robot multi-step assembly.
-
Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes
GAPL learns a compact set of canonical forgery prototypes and applies two-stage LoRA training to build a low-variance feature space that improves generalization across GAN and diffusion generators.
-
MiMo-Embodied: X-Embodied Foundation Model Technical Report
MiMo-Embodied is a single foundation model that achieves state-of-the-art results on 17 embodied AI benchmarks and 12 autonomous driving benchmarks through multi-stage learning, curated data, and CoT/RL fine-tuning that produces positive cross-domain transfer.
-
InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy
InternVLA-M1 uses spatially guided pre-training on 2.3M examples followed by action post-training to deliver up to 17% gains on robot manipulation benchmarks and 20.6% on unseen objects.
-
OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mapping
OmniView-Space framework with MPSM, tool-guided reasoning, and distillation achieves SOTA on spatial reasoning benchmarks for MLLMs while reducing external geometry dependencies.
-
FutureNav: Unified World-Action Modeling for Vision-and-Language Navigation
FutureNav proposes a 4B-scale VLM that jointly optimizes action prediction, inverse/forward dynamics, and future state generation for VLN and reports SOTA results on multiple benchmarks.
-
Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy
OmniAct framework integrates planning, memory, and verification to enable persistent autonomy in omnimodal embodied agents, showing improved success and stable context in 40 real-world tasks.
-
SPARC: Reliable Spatial Annotations from Robot Demonstrations at Scale
SPARC generates reliable spatial annotations for robot demonstrations by leveraging spatio-temporal task structure, outperforming detection baselines on localization accuracy while retaining more samples and enabling competitive model performance without manual annotations.
-
Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data
Introduces embodied trajectory-coupled data and a three-stage training recipe to bridge VLMs to generalizable VLAs without steep degradation of pre-trained representations.
-
Grounded 3D-Aware Spatial Vision-Language Modeling
GR3D is a VLM that combines explicit 2D, implicit 2D, and monocular 3D grounding mechanisms to improve performance on spatial understanding benchmarks.
-
Extending Embodied Question Answering from Perception to Decision
Introduces EQA-Decision dataset with 4M+ QA pairs across four embodied reasoning dimensions and RoboDecision baseline for joint perception-reasoning-decision evaluation.
-
Rethinking VLM Representation for VLA Initialization
Experiments indicate original VLM representations are crucial for VLA performance, LoRA outperforms full finetuning, and staged robot-data pretraining yields the strongest initialization.
-
OpenVLThinkerV2: A Generalist Multimodal Reasoning Model for Multi-domain Visual Tasks
OpenVLThinkerV2 applies a new Gaussian GRPO training objective with response and entropy shaping to outperform prior open-source and proprietary models on 18 visual reasoning benchmarks.
-
RoboAgent: Chaining Basic Capabilities for Embodied Task Planning
RoboAgent chains basic vision-language capabilities inside a single VLM via a scheduler and trains it in three stages (behavior cloning, DAgger, RL) to improve embodied task planning.
-
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models
An 8B embodied foundation model trained on 15B tokens with multi-task RL and a Planner-Grounder-Corrector loop claims SOTA on 16/24 embodied VLM benchmarks and strong VLA/real-robot transfer.
-
Kwai Keye-VL-2.0 Technical Report
Kwai Keye-VL-2.0-30B-A3B is a 30B MoE model with 3B active parameters using DSA adaptation and MOPD distillation that reports SOTA results on video understanding and agent benchmarks.
- RoboProcessBench: Benchmarking Process-Aware Understanding in Vision-Language Robotic Manipulation
- DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models