Rethinking Agent Trajectory Compression: A Workflow-Aware Perspective
cs.AI
Submitted: 2026-11-06
Updated: 2026-11-09
Code: https://github.com/openai/codex
Terminology
Sources
- ReAct: Synergizing Reasoning and Acting in Language Models
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
- Lost in the Middle: How Language Models Use Long Contexts
- LCM: Lossless Context Management
- Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents
- Reflexion: Language Agents with Verbal Reinforcement Learning
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
- Learning to Compress Prompts with Gist Tokens
- Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector Perspective
- Telegraph English: Semantic Prompt Compression via Structured Symbolic Rewriting
- Dynamic Long Context Reasoning over Compressed Memory via End-to-End Reinforcement Learning
- BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
- Odysseys: Benchmarking Web Agents on Realistic Long Horizon Tasks
- SnapKV: LLM Knows What You are Looking for Before Generation
- PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- AgentFold: Long-Horizon Web Agents with Proactive Context Management
- From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
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