A unified threat model for LLM-enabled robots reveals three cross-boundary attack chains from user input to unsafe physical actuation due to missing validations and unmediated crossings.
super hub Canonical reference
PaLM-E: An Embodied Multimodal Language Model
Canonical reference. 98% of citing Pith papers cite this work as background.
205
Pith papers citing it
Background
98% of classified citations
abstract
Large language models excel at a wide range of complex tasks. However, enabling general inference in the real world, e.g., for robotics problems, raises the challenge of grounding. We propose embodied language models to directly incorporate real-world continuous sensor modalities into language models and thereby establish the link between words and percepts. Input to our embodied language model are multi-modal sentences that interleave visual, continuous state estimation, and textual input encodings. We train these encodings end-to-end, in conjunction with a pre-trained large language model, for multiple embodied tasks including sequential robotic manipulation planning, visual question answering, and captioning. Our evaluations show that PaLM-E, a single large embodied multimodal model, can address a variety of embodied reasoning tasks, from a variety of observation modalities, on multiple embodiments, and further, exhibits positive transfer: the model benefits from diverse joint training across internet-scale language, vision, and visual-language domains. Our largest model, PaLM-E-562B with 562B parameters, in addition to being trained on robotics tasks, is a visual-language generalist with state-of-the-art performance on OK-VQA, and retains generalist language capabilities with increasing scale.
hub tools
citation-role summary
background 55
citation-polarity summary
claims ledger
- abstract Large language models excel at a wide range of complex tasks. However, enabling general inference in the real world, e.g., for robotics problems, raises the challenge of grounding. We propose embodied language models to directly incorporate real-world continuous sensor modalities into language models and thereby establish the link between words and percepts. Input to our embodied language model are multi-modal sentences that interleave visual, continuous state estimation, and textual input encodings. We train these encodings end-to-end, in conjunction with a pre-trained large language model, f
authors
co-cited works
roles
background 55representative citing papers
MME-RealWorld is the largest manually annotated high-resolution benchmark for MLLMs, where even the best models achieve less than 60% accuracy on challenging real-world tasks.
Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.
EgoSafetyBench shows VLMs reliably spot hazard-containing videos but miss specific contextual hazards and are degraded by misleading in-scene text.
SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
VLA models from VLM adaptation can be pruned 12-30% via multi-module joint scheme based on divergence signals while keeping ~90% performance on LIBERO without post-pruning recovery, unlike standard criteria that collapse.
A prompt-only attack called command-preserving trajectory redirection can steer VLA robot behavior to attacker-chosen physical outcomes while the text still appears to match the intended task.
Introduces LIBERO-Occ benchmark showing VLA performance drop under occlusion and Viewpoint Imagination method that generates complementary views to improve robustness without extra hardware.
ReCoVLA improves VLA policy reliability by using a VLM as a semantic reward selector to train residual recovery policies in simulation, raising average success from 36.7% to 66.7% in sim and achieving 61.7% in zero-shot sim-to-real physical tests.
Introduces Colosseum V2 benchmark for evaluating VLA model generalization in robotic manipulation with 28 tasks, revealing limitations in current methods and sim-real correlations.
VGenST-Bench is a new video benchmark for MLLM spatio-temporal reasoning built via generative synthesis, a multi-agent pipeline with human oversight, a 3x2x2 taxonomy, and hierarchical tasks separating perception from reasoning.
Aero-World adapts a pretrained latent diffusion transformer for action-conditioned aerial video generation by injecting inertial action tokens and using a frozen latent-space Physics Probe for inertial consistency supervision during LoRA finetuning, with a new AeroBench benchmark showing improved AA
PRISM is a tiered benchmark with 300 human-verified tasks across five photorealistic apartments that diagnoses embodied agent failures in basic ability, reasoning ability, and long-horizon ability using an agent-agnostic API.
ECHO organizes VLA experiences into a hierarchical memory tree in hyperbolic space via autoencoder and entailment constraints, delivering a 12.8% success-rate gain on LIBERO-Long over the pi0 baseline.
KinDER is a new open-source benchmark that demonstrates substantial gaps in current robot learning and planning methods for handling physical constraints.
AeroBridge-TTA achieves +22 pt average gains on out-of-distribution UAV dynamics mismatches by updating a latent state online from observed transitions in a language-conditioned policy.
LLM planners for robots often produce dangerous plans even when planning succeeds, with safety awareness staying flat as model scale improves planning ability.
Creates the first egocentric screen-view movie emotion benchmark and demonstrates that cinematic models drop sharply in Macro-F1 on realistic robot-like viewing conditions while domain-specific training improves robustness.
Mosaic uses cross-modal clusters as the unit for KVCache organization in VLMs to achieve up to 1.38x speedup in streaming long-video understanding.
Large multimodal models display emerging but limited spatial action capabilities in goal-oriented urban 3D navigation, remaining far from human-level performance with errors diverging rapidly after critical decision points.
KITE is a training-free method that uses keyframe-indexed tokenized evidence including BEV schematics to enhance VLM performance on robot failure detection, identification, localization, explanation, and correction.
KITScenes LongTail supplies multimodal driving data and multilingual expert reasoning traces to benchmark models on rare scenarios beyond basic safety metrics.
AR-VLA introduces a standalone autoregressive action expert with long-lived memory that generates context-aware continuous actions for VLAs, replacing chunk-based heads with smoother trajectories and maintained task success.
Mema adds a stateful memory module to vision encoders that accumulates hierarchical visual features across layers and selectively injects portions back via feedback to preserve fine-grained cues, yielding consistent gains on multimodal benchmarks.
citing papers explorer
-
From Prompt to Physical Actuation: Holistic Threat Modeling of LLM-Enabled Robotic Systems
A unified threat model for LLM-enabled robots reveals three cross-boundary attack chains from user input to unsafe physical actuation due to missing validations and unmediated crossings.
-
MME-RealWorld: Could Your Multimodal LLM Challenge High-Resolution Real-World Scenarios that are Difficult for Humans?
MME-RealWorld is the largest manually annotated high-resolution benchmark for MLLMs, where even the best models achieve less than 60% accuracy on challenging real-world tasks.
-
Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.
-
EgoSafetyBench: A Diagnostic Egocentric Video Benchmark for Evaluating Embodied VLMs as Runtime Safety Guards
EgoSafetyBench shows VLMs reliably spot hazard-containing videos but miss specific contextual hazards and are degraded by misleading in-scene text.
-
Adapting Generalist Robot Policies with Semantic Reinforcement Learning
SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
-
Revisiting Parameter Redundancy in Vision-Language-Action Models: Insights from VLM-to-VLA Adaptation
VLA models from VLM adaptation can be pruned 12-30% via multi-module joint scheme based on divergence signals while keeping ~90% performance on LIBERO without post-pruning recovery, unlike standard criteria that collapse.
-
Trajectory-Level Redirection Attacks on Vision-Language-Action Models
A prompt-only attack called command-preserving trajectory redirection can steer VLA robot behavior to attacker-chosen physical outcomes while the text still appears to match the intended task.
-
LIBERO-Occ: Evaluating and Improving Vision-Language-Action Models under Scene-Induced Occlusion via Viewpoint Imagination
Introduces LIBERO-Occ benchmark showing VLA performance drop under occlusion and Viewpoint Imagination method that generates complementary views to improve robustness without extra hardware.
-
ReCoVLA: VLM-Guided Reward Compilation for Failure Recovery in Vision-Language-Action Policies
ReCoVLA improves VLA policy reliability by using a VLM as a semantic reward selector to train residual recovery policies in simulation, raising average success from 36.7% to 66.7% in sim and achieving 61.7% in zero-shot sim-to-real physical tests.
-
Colosseum V2: Benchmarking Generalization for Vision Language Action Models
Introduces Colosseum V2 benchmark for evaluating VLA model generalization in robotic manipulation with 28 tasks, revealing limitations in current methods and sim-real correlations.
-
VGenST-Bench: A Benchmark for Spatio-Temporal Reasoning via Active Video Synthesis
VGenST-Bench is a new video benchmark for MLLM spatio-temporal reasoning built via generative synthesis, a multi-agent pipeline with human oversight, a 3x2x2 taxonomy, and hierarchical tasks separating perception from reasoning.
-
Aero-World: Action-Conditioned Aerial Video Generation from Inertial Controls
Aero-World adapts a pretrained latent diffusion transformer for action-conditioned aerial video generation by injecting inertial action tokens and using a frozen latent-space Physics Probe for inertial consistency supervision during LoRA finetuning, with a new AeroBench benchmark showing improved AA
-
PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments
PRISM is a tiered benchmark with 300 human-verified tasks across five photorealistic apartments that diagnoses embodied agent failures in basic ability, reasoning ability, and long-horizon ability using an agent-agnostic API.
-
ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models
ECHO organizes VLA experiences into a hierarchical memory tree in hyperbolic space via autoencoder and entailment constraints, delivering a 12.8% success-rate gain on LIBERO-Long over the pi0 baseline.
-
KinDER: A Physical Reasoning Benchmark for Robot Learning and Planning
KinDER is a new open-source benchmark that demonstrates substantial gaps in current robot learning and planning methods for handling physical constraints.
-
AeroBridge-TTA: Test-Time Adaptive Language-Conditioned Control for UAVs
AeroBridge-TTA achieves +22 pt average gains on out-of-distribution UAV dynamics mismatches by updating a latent state online from observed transitions in a language-conditioned policy.
-
Using large language models for embodied planning introduces systematic safety risks
LLM planners for robots often produce dangerous plans even when planning succeeds, with safety awareness staying flat as model scale improves planning ability.
-
Watching Movies Like a Human: Egocentric Emotion Understanding for Embodied Companions
Creates the first egocentric screen-view movie emotion benchmark and demonstrates that cinematic models drop sharply in Macro-F1 on realistic robot-like viewing conditions while domain-specific training improves robustness.
-
Mosaic: Cross-Modal Clustering for Efficient Video Understanding
Mosaic uses cross-modal clusters as the unit for KVCache organization in VLMs to achieve up to 1.38x speedup in streaming long-video understanding.
-
How Far Are Large Multimodal Models from Human-Level Spatial Action? A Benchmark for Goal-Oriented Embodied Navigation in Urban Airspace
Large multimodal models display emerging but limited spatial action capabilities in goal-oriented urban 3D navigation, remaining far from human-level performance with errors diverging rapidly after critical decision points.
-
KITE: Keyframe-Indexed Tokenized Evidence for VLM-Based Robot Failure Analysis
KITE is a training-free method that uses keyframe-indexed tokenized evidence including BEV schematics to enhance VLM performance on robot failure detection, identification, localization, explanation, and correction.
-
LongTail Driving Scenarios with Reasoning Traces: The KITScenes LongTail Dataset
KITScenes LongTail supplies multimodal driving data and multilingual expert reasoning traces to benchmark models on rare scenarios beyond basic safety metrics.
-
AR-VLA: True Autoregressive Action Expert for Vision-Language-Action Models
AR-VLA introduces a standalone autoregressive action expert with long-lived memory that generates context-aware continuous actions for VLAs, replacing chunk-based heads with smoother trajectories and maintained task success.
-
Mema: Memory-Augmented Adapter for Enhanced Vision-Language Understanding
Mema adds a stateful memory module to vision encoders that accumulates hierarchical visual features across layers and selectively injects portions back via feedback to preserve fine-grained cues, yielding consistent gains on multimodal benchmarks.
-
PhysMem: Scaling Test-Time Memory for Embodied Physical Reasoning
PhysMem enables VLM-based robot planners to learn and verify physical properties through test-time interaction and hypothesis testing, raising success on a brick insertion task from 23% to 76%.
-
UniLACT: Depth-Aware RGB Latent Action Learning for Vision-Language-Action Models
UniLACT improves VLA models by adding depth-aware unified latent action pretraining that outperforms RGB-only baselines on seen and unseen manipulation tasks.
-
ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs
ST-BiBench reveals a coordination paradox in which MLLMs show strong high-level strategic reasoning yet fail at fine-grained 16-dimensional bimanual action synthesis and multi-stream fusion.
-
Large Video Planner Enables Generalizable Robot Control
A video foundation model trained on human demonstrations generates zero-shot plans that convert to executable robot actions on novel scenes and tasks.
-
From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
-
WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs
WorldSense provides the first benchmark requiring synergistic audio-video-text understanding on 1,662 real-world videos and 3,172 QA pairs, where the best current multimodal LLM reaches only 65.1% accuracy.
-
Modality-Inconsistent Continual Learning of Multimodal Large Language Models
The paper introduces the MICL scenario for MLLMs with modality and task shifts and proposes MoInCL using pseudo-target generation and instruction-based distillation, reporting gains over continual learning baselines on six tasks.
-
3D-VLA: A 3D Vision-Language-Action Generative World Model
3D-VLA is a new embodied foundation model that uses a 3D LLM plus aligned diffusion models to generate future images and point clouds for improved reasoning and action planning in 3D environments.
-
RT-H: Action Hierarchies Using Language
RT-H learns robot policies by first predicting language motions as an intermediate representation and then mapping those plus the high-level task to actions, yielding more robust multi-task performance and the ability to learn from language interventions.
-
Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Models
SuSIE uses a finetuned InstructPix2Pix diffusion model to propose subgoal images that guide a low-level goal-conditioned policy, achieving SOTA zero-shot performance on CALVIN and real-world manipulation.
-
Learning Interactive Real-World Simulators
UniSim learns a universal real-world simulator from orchestrated diverse datasets, enabling zero-shot deployment of policies trained purely in simulation.
-
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation
A new shared video-image tokenizer enables large language models to surpass diffusion models on standard visual generation benchmarks.
-
VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models
VoxPoser uses LLMs to compose 3D value maps via VLM interaction for model-based synthesis of robust robot trajectories on open-set language-specified manipulation tasks.
-
Voyager: An Open-Ended Embodied Agent with Large Language Models
Voyager achieves superior lifelong learning in Minecraft by combining an automatic exploration curriculum, a library of executable skills, and iterative LLM prompting with environment feedback, yielding 3.3x more unique items and 15.3x faster milestone unlocks than prior methods while generalizing技能
-
LLM+P: Empowering Large Language Models with Optimal Planning Proficiency
LLM+P lets LLMs solve planning problems optimally by converting them to PDDL for classical planners and back to natural language.
-
Visual Instruction Tuning
LLaVA is trained on GPT-4 generated visual instruction data to achieve 85.1% relative performance to GPT-4 on synthetic multimodal tasks and 92.53% accuracy on Science QA.
-
Path Planning in Physically Viable World Models
A physically viable world model augments 3D Gaussian splats with physics simulation to assess robot route feasibility under simulated terrain changes like flooding, revealing failures not visible in static maps.
-
ReShift: Aha-Moment-Driven Reasoning-Level Backdoor Attacks on Vision-Language Models
ReShift is a reasoning-level backdoor framework for VLMs that uses poisoned data construction and joint optimization to shift CoT trajectories on trigger while preserving surface coherence.
-
Automating the Design of Embodied AgentArchitectures
Automated architecture search for embodied agents produces directional success-rate gains on vision-language and manipulation tasks while exposing limits from simulation noise and incomplete credit assignment.
-
Trust Your Instincts: Confidence-Driven Test-Time RL for Vision-Language-Action Models
T^2VLA is a test-time reinforcement learning framework for VLAs that uses internal confidence to define intrinsic rewards via similarity to high-confidence expert demonstrations and a dual-expert bootstrapping mechanism.
-
Enhancing Part-Level Point Grounding for Any Open-Source MLLMs
A plug-in Q-Synth Module plus Attention-to-Point Decoder converts text-conditioned attention in frozen MLLMs into point heatmaps, improving part-level grounding accuracy on multiple datasets.
-
Improving Vision-Language-Action Model Fine-Tuning with Structured Stage and Keyframe Supervision
StaKe adds lightweight auxiliary heads for manipulation stage identification and next-gripper-transition keyframe prediction to VLA fine-tuning, reporting relative success rate gains of 14% in bimanual simulation and 56% on single-arm real-robot tasks.
-
SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation
SSI-Policy uses an RGB-only Structured Scene Interface to improve LIBERO benchmark performance by nearly 15% with only 10 demonstrations per task compared to prior methods.
-
Graph it first! Enabling Reasoning on Long-form Egocentric Videos through Scene Graphs
Egocentric Scene Graphs convert long videos into short structured text so MLLMs can answer questions about entire sequences, achieving SOTA on HD-EPIC VQA.
-
Steering Vision-Language Models with Joint Sparse Autoencoders
JSAE jointly factorizes pooled vision and language activations in VLMs into aligned interpretable features, revealing layer-dependent asymmetry in additive steering versus suppression on three models.
-
Dense Reward for Multi-View 3D Reasoning with Global Maps and Local Views
DR-MV3D decomposes MV3D-VQA into global map construction, question-conditioned view planning, and egocentric grounding, supervised by global consistency and local trajectory rewards optimized via GRPO.