Folio: Recoverable Multi-Resolution Context Folding for Long-Horizon Language Agents
cs.AI
Submitted: 2026-11-12
Updated: 2026-11-12
Terminology
Sources
- LCM: Lossless Context Management
- SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
- Self-Compacting Language Model Agents
- The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management
- SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness
- Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks
- CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents
- MemGPT: Towards LLMs as Operating Systems
- Contextual Memory Virtualisation: DAG-Based State Management and Structurally Lossless Trimming for LLM Agents
- Context Language Models
- AgentRM: An OS-Inspired Resource Manager for LLM Agent Systems
- Scaling Long-Horizon LLM Agent via Context-Folding
- Context Compaction Theory
- When Can Agents Forget Their Reasoning? ICLR for Long-Horizon Agent Context Compression
- Token Reduction Is Not Cost Reduction
- LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
- Persistent Context Graphs for Efficient Memory Compaction in LLM Agents
- AgentFold: Long-Horizon Web Agents with Proactive Context Management
- LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- 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
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents