Stable Geometry with Divergent Task Evidence for Efficient Long-Horizon Agent Compression

arXiv:2609.27332 · cs.AI · Submitted 2026-09-23 · Read on arXiv

cs.AI

Submitted: 2026-09-23

Updated: 2026-09-23

Terminology

• Core — 17× in this paper · explained in 2 episodes
• Ablation — 13× in this paper · explained in 1 episode
• Redundancy — 10× in this paper · explained in 2 episodes
• Alignment — 6× in this paper · explained in 1 episode
• Embeddings — 6× in this paper · explained in 2 episodes
• Aggregation — 4× in this paper · explained in 1 episode
• Anisotropy — 3× in this paper · explained in 3 episodes
• Benchmark — 3× in this paper · explained in 2 episodes
• Pruning — 3× in this paper · explained in 4 episodes
• Geometry-Guided Evidence-Preserving Memory — 2× in this paper
• RAG — 2× in this paper · explained in 3 episodes
• Reconstruction Error — 2× in this paper · explained in 2 episodes
• Block Encoding — 1× in this paper · explained in 4 episodes
• Contrastive Learning — 1× in this paper · explained in 5 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Semantic Similarity — 1× in this paper · explained in 2 episodes
• State Management — 1× in this paper · explained in 3 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• ReAct: Synergizing Reasoning and Acting in Language Models
• ACON: Optimizing Context Compression for Long-horizon LLM Agents
• SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
• SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
• Self-Compacting Language Model Agents
• All-but-the-Top: Simple and Effective Postprocessing for Word Representations
• Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents
• Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
• Self-GC: Self-Governing Context for Long-Horizon LLM Agents
• CoMem: Context Management with A Decoupled Long-Context Model
• ACM: Agentic Context Management for Long Horizon Tasks
• MemGPT: Towards LLMs as Operating Systems
• LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference
• RULER: What's the Real Context Size of Your Long-Context Language Models?
• Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction
• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
• In-context Autoencoder for Context Compression in a Large Language Model

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
• 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
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection