When Replanning Becomes the Bottleneck: Budgeted Replanning for Embodied Agents

• URL: http://arxiv.org/abs/2608.01428v1
• Date: Sun, 02 Aug 2026 18:17:35 GMT
• Title: When Replanning Becomes the Bottleneck: Budgeted Replanning for Embodied Agents
• Abstract summary: We present BRACE, a controller that formulates replanning as a budgeted control loop.<n>We introduce E-RECAP, a cost-aware progressive token pruning method.<n>Across Meta Habitat, RoboFactory, and AirSim, BRACE with E-RECAP reduces replanning-call token counts by 62-92%.
• Score: 1.7674345486888505
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Embodied agents replan frequently to recover from execution drift, partial observability, and coordination hazards, but each LLM-based replanning call can consume an accumulated textual context that grows over time and across agents. Once this context becomes large, replanning latency develops heavy tails and can miss real-time deadlines even when task success remains high, a failure mode that is hard to detect from average latency or success alone. We present BRACE, a controller that formulates replanning as a budgeted control loop by deciding whether to replan, selecting a replanning mode, and allocating an explicit token budget and latency service-level objective (SLO) while accounting for optional efficiency modules. As a reusable component, we introduce E-RECAP, a cost-aware progressive token pruning method that predicts token utility and prunes replanning contexts across transformer layers while preserving critical head and tail tokens. Across Meta Habitat, RoboFactory, and AirSim, BRACE with E-RECAP reduces replanning-call token counts by 62-92% and SLO violation rates from 85.5-100.0% to 4.7-50.0% in settings where task success is already saturated. In a harder RoboFactory setting where open-loop, frozen-plan, and No BRACE all fail, BRACE + E-RECAP reaches 80.0% success with 4.6% SLO violations, demonstrating that tail-aware per-call budgeting is effective across embodied platforms.

Related papers

• LePlanner: An Iterative Amortized Controller For World Models [0.0]
  We propose LePlanner, an amortized iterative controller that learns to construct and refine latent action sequences through a frozen world-model predictor.<n>LePlanner is trained with an arrival-and-hold objective that encourages the controller to reach the goal at the earliest feasible horizon and remain there.<n>It achieves success rates of 98% on PushT, 100% on Reacher, 100% on TwoRooms, and 92% on the OGBench Cube task.
  arXiv  Detail & Related papers  (2026-09-12T10:01:21Z)
• Planning or Learning: Reliability and Cost in Multi-Asset Maintenance [0.8258557806969798]
  We empirically compare planning and reinforcement learning for multi-asset bearing maintenance using run-to-failure data.<n>Planning enforces reliability as a hard constraint and produces zero-failure policies whose total cost is largely insensitive to the magnitude of failure penalties.<n>This study clarifies the trade-offs between reliability and cost in multi-asset maintenance and suggests that planning and RL are complementary approaches.
  arXiv  Detail & Related papers  (2026-09-11T22:10:59Z)
• A Self-Triggered Agentic Push Recommendation System [77.13502692161426]
  Push notification is a critical recommendation scenario on large-scale platforms.<n>We propose a proactive, Self-Triggered End-to-end Agentic Push Recommendation System.<n>STEPS is already fully deployed at Douyin with over 1 billion users.
  arXiv  Detail & Related papers  (2026-08-03T09:20:53Z)
• SPIKE: An Adaptive Dual Controller Framework for Cost-Efficient Long-Horizon Game Agents [74.84742205422825]
  We present SPIKE, an adaptive dual controller framework for cost-efficient long-horizon game control.<n>Its Strategic Controller performs low-frequency global planning, failure analysis, and recovery, while its Reactive Controller handles fast local execution under a strict token budget.<n>An Event Trigger monitors visual change, task progress, repeated actions, and failure signals to decide when control should stay reactive or escalate to strategic reasoning.<n>This design reuses strategic proposals over multiple reactive steps, supports local override when plans become stale, and reserves expensive reasoning for moments where extra deliberation is useful.
  arXiv  Detail & Related papers  (2026-05-18T16:43:32Z)
• RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models [90.39703013636868]
  RePO-VLA is a recovery-driven policy optimization framework.<n>It assigns distinct roles to success, recovery, and failure trajectories.<n>It raises adversarial success from 20% to 75% on average and up to 80% in scaled real-world trials.
  arXiv  Detail & Related papers  (2026-05-10T08:24:05Z)
• MAFIG: Multi-agent Driven Formal Instruction Generation Framework [17.52930666720453]
  Emergency situations in scheduling systems often trigger local functional failures that undermine system stability and even cause system collapse.<n>Existing methods rely on robust scheduling or reactive scheduling, handling emergencies through predefined rules or rescheduling strategies.<n>We introduce span-focused loss-driven loss-driven local distillation mechanism to transfer the decision-making capability of powerful Cloud Large Language Models to lightweight local models.
  arXiv  Detail & Related papers  (2026-04-13T04:47:20Z)
• LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications [51.56484100374058]
  We present a readiness harness for LLM and RAG applications that turns evaluation into a deployment decision workflow.<n>The system combines automated benchmarks, OpenTelemetry observability, and CI quality gates under a minimal API contract.<n>We evaluate the harness on ticket-routing and BEIR tasks with full Azure matrix coverage.
  arXiv  Detail & Related papers  (2026-03-28T18:03:32Z)
• When Should a Robot Think? Resource-Aware Reasoning via Reinforcement Learning for Embodied Robotic Decision-Making [68.12864562049957]
  Embodied robotic systems increasingly rely on large language model (LLM)-based agents to support high-level reasoning.<n>This raises a fundamental question for embodied agents: when should the agent reason, and when should it act?<n>We propose RARRL (Resource-Aware Reasoning via Reinforcement Learning), a hierarchical framework for resource-aware orchestration of embodied agents.
  arXiv  Detail & Related papers  (2026-03-17T15:38:50Z)
• iScheduler: Reinforcement Learning-Driven Continual Optimization for Large-Scale Resource Investment Problems [30.109981943437006]
  Scheduling precedence-constrained tasks under shared renewable resources is central to modern computing platforms.<n>We present iScheduler, a reinforcement-learning-driven iterative scheduling framework.<n>Experiments show that iScheduler attains competitive resource costs while reducing time to feasibility by up to 43$times$ against strong commercial baselines.
  arXiv  Detail & Related papers  (2026-01-30T11:20:58Z)
• Accelerating Multi-modal LLM Gaming Performance via Input Prediction and Mishit Correction [4.323124094061299]
  Real-time sequential control agents are often bottlenecked by inference latency.<n>We propose a framework that adapts the predict-then-verify philosophy of speculative execution to model-based control with TD-MPC2.<n>We show that our method reduces the number of planning inferences from 500 to 282, improves end-to-end step latency by 25 percent, and maintains strong control performance with only a 7.1 percent return reduction.
  arXiv  Detail & Related papers  (2025-12-19T05:34:52Z)
• No-Regret Learning Under Adversarial Resource Constraints: A Spending Plan Is All You Need! [56.80767500991973]
  We focus on two canonical settings: $(i)$ online resource allocation where rewards and costs are observed before action selection, and $(ii)$ online learning with resource constraints where they are observed after action selection, under full feedback or bandit feedback.<n>It is well known that achieving sublinear regret in these settings is impossible when reward and cost distributions may change arbitrarily over time.<n>We design general (primal-)dual methods that achieve sublinear regret with respect to baselines that follow the spending plan. Crucially, the performance of our algorithms improves when the spending plan ensures a well-balanced distribution of the budget
  arXiv  Detail & Related papers  (2025-06-16T08:42:31Z)
• ALAS: A Stateful Multi-LLM Agent Framework for Disruption-Aware Planning [2.1331883629523634]
  We present Adaptive LLM Agent System (ALAS), a framework that tackles four fundamental LLM deficits.<n>ALAS decomposes each plan into role-specialized agents, equips them with automatic state tracking, and coordinates them through a lightweight protocol.<n>On real-world, large-scale job-shop scheduling benchmarks, ALAS sets new best results for static sequential planning and excels in dynamic reactive scenarios with unexpected disruptions.
  arXiv  Detail & Related papers  (2025-05-18T17:27:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.