菜单
Implement efficient memory systems for long-term LLM agent interactions using semantic compression, achieving 30-fold inference token reduction while improving F1 scores by 26.4%—enabling agents to learn from extended interaction histories without prohibitive context costs.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/ADu2021/skillXiv --skill simplemem-lifelong-memory命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | simplemem-lifelong-memory |
| title | SimpleMem: Efficient Lifelong Memory for LLM Agents |
| version | 0.0.2 |
| engine | skillxiv-v0.0.2-claude-opus-4.6 |
| license | MIT |
| url | https://arxiv.org/abs/2601.02553 |
| keywords | ["Memory Systems","LLM Agents","Semantic Compression","Context Management","Lifelong Learning"] |
| description | Implement efficient memory systems for long-term LLM agent interactions using semantic compression, achieving 30-fold inference token reduction while improving F1 scores by 26.4%—enabling agents to learn from extended interaction histories without prohibitive context costs. |
SimpleMem is an efficient memory framework for LLM agents engaged in long-term interactions. It solves a critical problem: managing interaction history in memory-constrained environments where full context retention causes exponential token bloat.
Core Challenge: Existing approaches either retain full histories (causing token explosion) or rely on iterative filtering (incurring high computational costs). SimpleMem achieves semantic lossless compression through a three-stage pipeline.
Distill unstructured agent interactions into compact, multi-view indexed memory units. This stage:
An intra-session process that instantly integrates related context into unified abstract representations.
Process:
This enables the agent to access coherent, consolidated context from fragmented experiences.
Infers search intent to dynamically determine retrieval scope and construct precise context.
Implementation approach:
Benchmark Results (LoCoMo dataset):
Token Efficiency Example:
vs. Passive Context Extension:
vs. Iterative Filtering:
vs. Fixed-Size Memory Buffers:
Semantic Indexing: Create multiple views of compressed memories to support diverse retrieval patterns:
# Multi-view indexing for memory units
memory_unit = {
"semantic_content": "...", # Compressed interaction summary
"entity_index": [...], # Entities mentioned
"action_types": [...], # Categories of actions taken
"temporal_markers": [...], # Session/episode boundaries
"relevance_scores": {...} # Pre-computed relevance for different query types
}
Retrieval Planning: Dynamically determine what memories to retrieve based on current context:
def plan_retrieval(current_task: str, memory_database: List[MemoryUnit]) -> List[MemoryUnit]:
"""Infer intent and retrieve relevant memories."""
# 1. Analyze task to determine information needs
information_needs = infer_information_needs(current_task)
# 2. Score memories for relevance to identified needs
scored_memories = [
(mem, relevance_score(mem, information_needs))
for mem in memory_database
]
# 3. Select minimal set meeting relevance threshold
relevant = [mem for mem, score in scored_memories if score > threshold]
return relevant
When to Use SimpleMem:
When NOT to use:
SimpleMem complements agent frameworks like:
Available at: https://github.com (code link provided in paper)