Pack Light, CarryOn: Building Object-Oriented Context for Long-Horizon Agents
cs.AI, cs.CL
Submitted: 2026-10-13
Updated: 2026-10-14
Comments: 24 pages. Project page: https://ethannju.github.io/carryon
Project page: https://ethannju.github.io/carryon
License: http://creativecommons.org/licenses/by/4.0/
The gist: Long-horizon agents accumulate progress in persistent task state, yet repeatedly process interaction histories describing how that progress was achieved.
Abstract
Long-horizon agents accumulate progress in persistent task state, yet repeatedly process interaction histories describing how that progress was achieved. This raises a natural question: can long-horizon context instead be organized around reusable progress rather than the history that produced it? In this paper, we introduce CarryOn, an object-oriented context runtime that organizes continuation around reusable task objects. Rather than treating compressed histories or reconstructed accounts of progress as the primary basis for continuation, CarryOn builds a context view from continuation carriers, complementary execution context, and a live tail of recent interactions. A core challenge is keeping this reusable basis valid after its surrounding history is retired: how can CarryOn update the basis without reconstructing the full representation at every fold? CarryOn maintains carrier records across folds, revising affected records against bound evidence and applicable environment checks while carrying forward unaffected records and jointly updating execution context. Experiments on OSWorld 2.0 and BrowseComp-Plus show improved gains of 5.8% and 7.0% while reducing total token usage by 51.9% and 22.9%, respectively. Project page: https://ethannju.github.io/carryon
Sources
- Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- ACON: Optimizing Context Compression for Long-horizon LLM Agents
- LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks
- gpt-oss-120b & gpt-oss-20b Model Card
- AgentProg: Empowering Long-Horizon GUI Agents with Program-Guided Context Management
- Agent Workflow Memory
- StructAgent: Harness Long-horizon Digital Agents with Unified Causal Structure
- ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
- A-MEM: Agentic Memory for LLM Agents
- AgentFold: Long-Horizon Web Agents with Proactive Context Management
- Learning Agent-Compatible Context Management for Long-Horizon Tasks
- OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks
- MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems