Effective asynchronous planning, or the ability to efficiently reason and plan over states and actions that must happen in parallel or sequentially, is essential for agents that must account for time delays, reason over diverse long-horizon tasks, and collaborate with other agents. While large language model (LLM) agents show promise in high-level task planning, current benchmarks focus primarily on short-horizon tasks and do not evaluate such asynchronous planning capabilities. We introduce Robotouille, a challenging benchmark environment designed to test LLM agents' ability to handle long-horizon asynchronous scenarios. Our synchronous and asynchronous datasets capture increasingly complex planning challenges that go beyond existing benchmarks, requiring agents to manage over- lapping tasks and interruptions Our results show that ReAct (gpt-4o) achieves 47% on synchronous tasks but only 11% on asynchronous tasks, highlighting significant room for improvement. We further analyze failure modes, demonstrating the need for LLM agents to better incorporate long-horizon feedback and self-audit their reasoning during task execution.

We address the problem of code generation from multi-turn execution feedback. Existing methods either generate code without feedback or use complex, hierarchical reinforcement learning to optimize multi-turn rewards. We propose a simple yet scalable approach, $\mu$CODE, that solves multi-turn code generation using only single-step rewards. Our key insight is that code generation is a one-step recoverable MDP, where the correct code can be recovered from any intermediate code state in a single turn. $\mu$CODE iteratively trains both a generator to provide code solutions conditioned on multi-turn execution feedback and a verifier to score the newly generated code. Experimental evaluations show that our approach achieves significant improvements over state-of-the-art baselines. We provide analysis of the design choices of the reward models and policy, and show the efficacy of $\mu$CODE at utilizing the execution feedback.

We investigate what it might look like for a robot to work with a human on a need-finding design task using an affinity diagram. While some recent projects have examined how human–robot teams might explore solutions to design problems, human–robot collaboration in the sensemaking aspects of the design process has not been studied. Designers use affinity diagrams to make sense of unstructured information by clustering paper notes on a work surface. To explore human–robot collaboration on a sensemaking design activity, we developed HIRO, an autonomous robot that constructs affinity diagrams with humans. In a within-user study, 56 participants affinity-diagrammed themes to characterize needs in quotes taken from real-world user data, once alone and once with HIRO. Users spent more time on the task with HIRO than alone, without strong evidence for corresponding effects on cognitive load. In addition, a majority of participants said they preferred to work with HIRO. From post-interaction interviews, we identified eight themes leading to four guidelines for robots that collaborate with humans on sensemaking design tasks: (1) account for the robot’s speed, (2) pursue mutual understanding rather than just correctness, (3) identify opportunities for constructive disagreements, and (4) use other modes of communication in addition to physical materials.

Language instructions and demonstrations are two natural ways for users to teach robots personalized tasks. Recent progress in Large Language Models (LLMs) has shown impressive performance in translating language instructions into code for robotic tasks. However, translating demonstrations into task code continues to be a challenge due to the length and complexity of both demonstrations and code, making learning a direct mapping intractable. This paper presents Demo2Code, a novel framework that generates robot task code from demonstrations via an extended chain-of-thought and defines a common latent specification to connect the two. Our framework employs a robust two-stage process: (1) a recursive summarization technique that condenses demonstrations into concise specifications, and (2) a code synthesis approach that expands each function recursively from the generated specifications. We conduct extensive evaluation on various robot task benchmarks, including a novel game benchmark Robotouille, designed to simulate diverse cooking tasks in a kitchen environment.

Multi-tenancy for latency-critical applications leads to resource interference and unpredictable performance. Core reconfiguration opens up more opportunities for application colocation, as it allows the hardware to adjust to the dynamic performance and power needs of a specific mix of co-scheduled services. However, reconfigurability also introduces challenges, as even for a small number of reconfigurable cores, exploring the design space becomes more time- and resource-demanding.We present CuttleSys, a runtime for reconfigurable multicores that leverages scalable and lightweight data mining to quickly identify suitable core and cache configurations for a set of co-scheduled applications. The runtime combines collaborative filtering to infer the behavior of each job on every core and cache configuration, with Dynamically Dimensioned Search to efficiently explore the configuration space. We evaluate CuttleSys on multicores with tens of reconfigurable cores and show up to 2.46× and 1.55× performance improvements compared to core-level gating and oracle-like asymmetric multicores respectively, under stringent power constraints.