REVIEW 24 cited by
Yell At Your Robot: Improving On-the-Fly from Language Corrections
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Hierarchical policies that combine language and low-level control have been shown to perform impressively long-horizon robotic tasks, by leveraging either zero-shot high-level planners like pretrained language and vision-language models (LLMs/VLMs) or models trained on annotated robotic demonstrations. However, for complex and dexterous skills, attaining high success rates on long-horizon tasks still represents a major challenge -- the longer the task is, the more likely it is that some stage will fail. Can humans help the robot to continuously improve its long-horizon task performance through intuitive and natural feedback? In this paper, we make the following observation: high-level policies that index into sufficiently rich and expressive low-level language-conditioned skills can be readily supervised with human feedback in the form of language corrections. We show that even fine-grained corrections, such as small movements ("move a bit to the left"), can be effectively incorporated into high-level policies, and that such corrections can be readily obtained from humans observing the robot and making occasional suggestions. This framework enables robots not only to rapidly adapt to real-time language feedback, but also incorporate this feedback into an iterative training scheme that improves the high-level policy's ability to correct errors in both low-level execution and high-level decision-making purely from verbal feedback. Our evaluation on real hardware shows that this leads to significant performance improvement in long-horizon, dexterous manipulation tasks without the need for any additional teleoperation. Videos and code are available at https://yay-robot.github.io/.
Forward citations
Cited by 24 Pith papers
-
Zero2Skill: Bootstrapping Robot Skills through Autonomous Data Collection, Training, and Deployment
Corrective Memory lets a robot data collector reuse natural-language corrections across rounds, cutting human time to 16% of teleoperation while matching its success rate and downstream policy performance.
-
FailSafe: Reasoning and Recovery from Failures in Vision-Language-Action Models
A VLM trained on auto-generated failure trajectories with executable correction actions helps VLA models detect and fix manipulation errors, lifting success rates by up to 22.6 percentage points.
-
Robix: A Unified Model for Robot Interaction, Reasoning and Planning
A three-stage-trained VLM unifies robot planning and dialogue, and beats commercial VLMs on the authors' interactive-task benchmarks.
-
CueLearner: Bootstrapping and local policy adaptation from relative feedback
CueLearner learns a relative-feedback model from a small number of human directional corrections and uses it to guide off-policy RL exploration or refine a deployed policy.
-
FEAST: A Flexible Mealtime-Assistance System Towards In-the-Wild Personalization
FEAST is a mealtime assistance robot that uses LLM-editable behavior trees and modular tools to let care recipients personalize feeding, drinking, and mouth wiping in real home settings.
-
SwitchVLA: Execution-Aware Task Switching for Vision-Language-Action Models
SwitchVLA trains a vision-language-action policy to handle mid-execution instruction changes by conditioning on contact state and a three-way behavior mode, using only existing single-task demonstrations.
-
Learning Compositional Behaviors from Demonstration and Language
BLADE learns structured, planable action representations from language-annotated demonstrations and composes them with a symbolic planner, outperforming latent and LLM/VLM baselines on new manipulation tasks.
-
Universal Actions for Enhanced Embodied Foundation Models
UniAct learns a shared discrete codebook of universal actions for many robots, decodes them with per-robot heads, and reports gains over larger baselines in robot manipulation.
-
Diffusion-VLA: Generalizable and Interpretable Robot Foundation Model via Self-Generated Reasoning
A robot policy generates its own language reasoning before acting and injects it into a diffusion action decoder, outperforming several VLA baselines on real-robot manipulation.
-
Inference-Time Policy Steering through Human Interactions
A stochastic MCMC sampling method, applied to frozen diffusion policies, best aligns generated robot trajectories with human interaction inputs while minimizing distribution shift.
-
A Few Words Go a Long Way: Language Guided Robot Policy Synthesis
Interactive LLM program synthesis plus a persistent skill library from natural-language corrections outperforms zero-shot VLAs and one-shot code policies on complex real-robot manipulation.
-
Scene Graph-Guided Proactive Replanning for Failure-Resilient Embodied Agent
A robot replanner that compares scene graphs to successful demonstrations before each subtask, triggering LLM-based replanning on mismatch, raises task success in AI2-THOR.
-
A Human-in-the-loop Approach to Robot Action Replanning through LLM Common-Sense Reasoning
A human-in-the-loop system lets users refine vision-generated robot behavior trees through natural-language requests to GPT-4o, correcting errors and adapting plans before execution.
-
Tactile-VLA: Unlocking Vision-Language-Action Model's Physical Knowledge for Tactile Generalization
Tactile-VLA fuses tactile sensing into a vision-language-action model so force-related instructions and corrective reasoning transfer to new contact-rich tasks with few demonstrations.
-
ProVox: Personalization and Proactive Planning for Situated Human-Robot Collaboration
A personalized, proactive LLM planner suggests helpful next actions during human-robot lunch packing, reporting 38.7% faster task execution at the cost of an extra 5.6-minute setup phase.
-
AMO: Adaptive Motion Optimization for Hyper-Dexterous Humanoid Whole-Body Control
A hybrid trajectory-optimization and RL framework lets a humanoid robot flex its torso and legs to reach and manipulate objects beyond the range of prior controllers.
-
RoboReflect: A Robotic Reflective Reasoning Framework for Grasping Ambiguous-Condition Objects
RoboReflect couples GPT-4V planning with a self-reflection loop, a discussion module, and a memory of successful strategies, claiming improved grasping success on ambiguous-condition objects over AnyGrasp, ReKep, and ...
-
CoA-VLA: Improving Vision-Language-Action Models via Visual-Textual Chain-of-Affordance
Adding a visual and textual chain-of-affordance reasoning step to a vision-language-action model improves robot manipulation success rates and generalization in the paper's evaluations.
-
Can Large Language Models Help Developers with Robotic Finite State Machine Modification?
Two LLMs, given old FSM code and an automatically generated change summary, reproduced five of six real robot FSM modifications from a RoboCup team's git history.
-
Large Model Empowered Embodied AI: A Survey on Decision-Making and Embodied Learning
A review that categorizes large-model-empowered embodied AI into hierarchical and end-to-end decision-making, imitation and reinforcement learning, and world models.
-
Steering Robots with Inference-Time Interactions
Frozen imitation policies can be steered at inference time via user interactions, with a diffusion-sampling method and a constraint-enforcing framework that provides formal task guarantees.
-
LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks
A unified vision-language-action model that emits a sub-task description followed by a discrete action token outperforms modular and action-only baselines on simulated long-horizon tabletop tasks.
-
ALPHA-$\alpha$ and Bi-ACT Are All You Need: Importance of Position and Force Information/Control for Imitation Learning of Unimanual and Bimanual Robotic Manipulation with Low-Cost System
Using force information from bilateral control improves imitation learning on unfamiliar objects, and a new low-cost ALPHA-alpha platform supports bimanual tasks.
-
A Survey On Enhancing Reinforcement Learning in Complex Environments: Insights from Human and LLM Feedback
A survey of prior work on using human and LLM feedback to improve reinforcement learning, plus attention-based methods for large state spaces.
Discussion (0). Continue with ORCID to comment.