EVOLVE-VLA: Test-Time Training from Environment Feedback for Vision-Language-Action Models
Achieving truly adaptive embodied intelligence requires agents that learn not just by imitating static demonstrations, but by continuously improving through environmental interaction, which is akin to how humans master skills through practice. Vision-Language-Action (VLA) models have advanced robotic manipulation by leveraging large language models, yet remain fundamentally limited by Supervised Finetuning (SFT): requiring hundreds of demonstrations per task, rigidly memorizing trajectories, and failing to adapt when deployment conditions deviate from training. We introduce EVOLVE-VLA, a test-time training framework enabling VLAs to continuously adapt through environment interaction with minimal or zero task-specific demonstrations. The key technical challenge is replacing oracle reward signals (unavailable at test time) with autonomous feedback. We address this through a learned progress estimator providing dense feedback, and critically, we design our framework to ``tame'' this inherently noisy signal via two mechanisms: (1) an accumulative progress estimation mechanism smoothing noisy point-wise estimates, and (2) a progressive horizon extension strategy enabling gradual policy evolution. EVOLVE-VLA achieves substantial gains: +8.6\% on long-horizon tasks, +22.0\% in 1-shot learning, and enables cross-task generalization -- achieving 20.8\% success on unseen tasks without task-specific demonstrations training (vs. 0\% for pure SFT). Qualitative analysis reveals emergent capabilities absent in demonstrations, including error recovery and novel strategies. This work represents a critical step toward VLAs that truly learn and adapt, moving beyond static imitation toward continuous self-improvements.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Test-Time Learning with an Evolving Library
We introduce EvoLib, a test-time learning framework that enables large language models to accumulate, reuse, and evolve knowledge across problem instances without parameter updates or external supervision. Instead of ada…
Mathematical ReasoningCode GenerationPACEvolve++: Improving Test-time Learning for Evolutionary Search Agents
Large language models have become drivers of evolutionary search, but most systems rely on a fixed, prompt-elicited policy to sample next candidates. This limits adaptation in practical engineering and research tasks, wh…
Sequential RecommendationReinforcement LearningAdapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents
LLM agents are shaped not only by their language models, but also by the runtime harness that mediates observation, tool use, action execution, feedback interpretation, and trajectory control. While existing agent adapta…
From Feedback Loops to Policy Updates: Reinforcement Fine-Tuning for LLM-Based Alpha Factor Discovery
Modern quantitative trading increasingly relies on systematic models to extract predictive signals from large-scale financial data, where alpha factor discovery plays a central role in transforming market observations in…
Effect of Feedback between Environment and Finite Population
Natural selection imply that any organisms including human being will evolve to improve its fitness advantage and the selected genotype or phenotype in equilibrium state will not vary over the time. However, evolutionary…