ASPIRE is a continual learning system that autonomously writes and refines robot control programs in a code-as-policy paradigm, discovering transferable skills and surpassing prior methods on various…

This paper introduces Retriever, an asynchronous decision model and runtime system for building long-horizon robot agents with explicit clock and input-consumption semantics.

The paper introduces AGENTCL, a rigorous evaluation framework that uses controlled task streams to accurately measure an agent's ability to accumulate and reuse knowledge across multiple tasks, thereb…

This paper proposes Rollout-Retrieval Lifelong Policy Learning (R$^2$LPL), a framework for continual policy improvement in autonomous driving by retrieving corrective targets from recoverable mistakes…

The paper proposes HiSME, a lightweight hierarchical skill meta-evolving solution that jointly optimizes skills and the skill evolving strategy by learning meta-skills from task execution traces, lead…

The paper introduces Banyan, a new continual reinforcement learning benchmark, demonstrating that while task diversity enables local transfer across distribution shifts, it does not guarantee sustaine…

The JAMEL framework addresses the challenge of effective exploration in open-ended environments by jointly training agent memory and exploration policies using natural, novelty-driven signals.

A model-based planner is presented that minimizes immediate cost and aggregated expected future cost across all robots in a task planning scenario with robots sharing a persistent environment.

The paper introduces Graph-as-Policy (GaP), a multi-agent coding harness for Variational Automation tasks that generates directed computation graphs and improves success rates and throughput through i…

The paper introduces Extreme-RGMT, a two-stage continual learning framework for robust generalist humanoid control that learns a base policy from diverse motion data and emphasizes difficult dynamic s…

This paper proposes using Answer-Set Programming (ASP) to implement and evaluate CARCASS abstractions, demonstrating a promising method for constructing powerful abstractions for Reinforcement Learnin…

This paper proposes using offline reinforcement learning (RL) as an efficient alternative to online RL for post-training code-generating LLMs, demonstrating its effectiveness, especially for smaller m…

TailLoR is a new parameter-efficient finetuning method that uses the singular bases of pre-trained weights to learn low-rank updates, specifically penalizing updates along dominant directions to impro…

The paper proposes DIBS, a decoupled behavioral cloning approach that stabilizes inductive generalization in RL by separating task-specific policy learning from the evolution function, leading to impr…

This paper characterizes continual classification in homogeneous models as sequential projections and identifies regularity properties for local linear convergence.

The paper introduces a data-centric optimization pipeline to improve coding agents' ability to interact with a branching lakehouse, showing significant accuracy gains by treating agent evaluation as a…

The paper introduces 'layered mutability,' a framework for analyzing how persistent self-modifying AI agents drift away from intended behavior due to the accumulation of locally reasonable, uncoordina…

The paper introduces AdvCL, a framework that repurposes adversarial perturbations as a geometric control signal to stabilize continual learning in large language models, significantly reducing forgett…