Tools Need Not Speak in Text: Latent Context Compression for Agents
cs.AI
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/THU-Agent/Coca-cola
Terminology
Sources
- CoACT: Action-Preserving Observation Compression for Coding Agents
- See What I See, Know What I Think: Dense Latent Communication Across Heterogeneous Agents
- Compressed Chain of Thought: Efficient Reasoning Through Dense Representations
- Implicit Chain of Thought Reasoning via Knowledge Distillation
- In-context Autoencoder for Context Compression in a Large Language Model
- GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
- Training Large Language Models to Reason in a Continuous Latent Space
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- Agent Primitives: Reusable Latent Building Blocks for Multi-Agent Systems
- ACON: Optimizing Context Compression for Long-horizon LLM Agents
- Context Embeddings for Efficient Answer Generation in RAG
- LongCodeZip: Compress Long Context for Code Language Models
- ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- MEMORYLLM: Towards Self-Updatable Large Language Models
- SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
- RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation
- ReAct: Synergizing Reasoning and Acting in Language Models
- The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
- MemGen: Weaving Generative Latent Memory for Self-Evolving 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