BestHub
Sign in
Tagged articles

hierarchical memory

9 articles · Page 1 of 1
Data Party THU
Data Party THU
Jul 22, 2026 · Artificial Intelligence

MemSlides Enables Precise Local Edits in AI‑Generated PPTs

MemSlides, a collaboration between Tsinghua, Shanghai Jiao Tong, and Beijing University of Posts and Telecommunications, introduces a hierarchical memory framework that lets a Slides Agent remember long‑term user preferences, temporary constraints, and tool experience, achieving accurate local PPT modifications and significantly improving personalization and edit efficiency.

AI PPTMemSlideshierarchical memory
0 likes · 13 min read
MemSlides Enables Precise Local Edits in AI‑Generated PPTs
DataFunSummit
DataFunSummit
Jul 20, 2026 · Artificial Intelligence

Why More Context, Tools, and Memory Make Agents Unstable—and How to Fix It

The article explains that in long‑running autonomous agents, larger context windows, excessive tool sets, and unstructured memory cause slower, costlier, and error‑prone behavior, and it proposes six design principles—dense context, minimal toolkits, task‑driven skill growth, hierarchical memory, action‑validated experience, and efficiency‑focused evaluation—to achieve stable, self‑evolving agents.

Agentic AIautonomous agentscontext management
0 likes · 17 min read
Why More Context, Tools, and Memory Make Agents Unstable—and How to Fix It
DataFunTalk
DataFunTalk
Jul 16, 2026 · Artificial Intelligence

Why Agents Slow Down and Cost More? Achieving True Self‑Evolution by Subtraction

In long‑running tasks agents often become slower, more expensive, and error‑prone because context explodes, tools proliferate, and memory becomes chaotic; the article argues that true self‑evolution requires reducing context to high‑density information, using a minimal yet composable tool set, and structuring memory hierarchically to let experience grow through validated actions.

Agentic AIEfficiency MetricsTool Optimization
0 likes · 18 min read
Why Agents Slow Down and Cost More? Achieving True Self‑Evolution by Subtraction
Machine Heart
Machine Heart
Jul 5, 2026 · Artificial Intelligence

Eliminating Fragmented Memory with Mandol: An Open‑Source Lightweight In‑Memory Agent System

Mandol tackles the fragmented memory problem of LLM agents by unifying representation, storage, and retrieval in a memory‑native architecture; benchmarked on LoCoMo and LongMemEval it achieves up to 92.21% accuracy, 5× faster latency, and runs efficiently on consumer‑grade hardware without external databases.

Agent MemoryLLMOpen Source
0 likes · 14 min read
Eliminating Fragmented Memory with Mandol: An Open‑Source Lightweight In‑Memory Agent System
Architect's Tech Stack
Architect's Tech Stack
Jun 4, 2026 · Artificial Intelligence

How TencentDB Agent Memory Cuts Token Usage by 61% and Boosts Task Success

TencentDB Agent Memory, an open‑source hierarchical memory system for long‑running AI agents, offloads tool calls, structures short‑term and four‑layer long‑term memories, and reduces token consumption by 61% while raising task success rate 51% and persona accuracy from 48% to 76%, all running locally with SQLite and no API keys.

AI AgentsOpenClawSQLite
0 likes · 4 min read
How TencentDB Agent Memory Cuts Token Usage by 61% and Boosts Task Success
PaperAgent
PaperAgent
May 18, 2026 · Artificial Intelligence

How MemWeaver Combines Behavioral and Cognitive Memory to Rebuild LLM Personalization

MemWeaver introduces a hierarchical memory that fuses behavior‑level and cognition‑level user signals, enabling large language models to generate more personalized content across multiple tasks, with extensive experiments, ablations, and an efficient incremental update mechanism demonstrating superior performance over strong baselines.

Incremental UpdateLLM personalizationLaMP benchmark
0 likes · 12 min read
How MemWeaver Combines Behavioral and Cognitive Memory to Rebuild LLM Personalization
DataFunTalk
DataFunTalk
Apr 30, 2026 · Artificial Intelligence

How GenericAgent Cuts Token Costs by 10× While Boosting AI Agent Performance

The technical report on GenericAgent, a self‑evolving LLM‑based agent, shows that by maximizing context information density and using a minimal atomic toolset with hierarchical memory, it achieves up to ten‑fold token savings, 100% task accuracy, and progressive efficiency gains across multiple benchmarks.

AI benchmarksGenericAgentLLM
0 likes · 15 min read
How GenericAgent Cuts Token Costs by 10× While Boosting AI Agent Performance
Ma Wei Says
Ma Wei Says
Apr 12, 2026 · Artificial Intelligence

Designing AI Agent Memory: From Simple Context Concatenation to a Cognitive System (Part 1)

The article analyzes why conventional AI‑agent memory—mere context concatenation—fails in long or cross‑session interactions, then proposes a four‑capability, three‑layer hierarchical memory framework (L1‑L3) with concrete design strategies, progressive summarisation, semantic retrieval, conflict resolution and decay mechanisms.

AI AgentLLMMemory Architecture
0 likes · 15 min read
Designing AI Agent Memory: From Simple Context Concatenation to a Cognitive System (Part 1)
Machine Heart
Machine Heart
Apr 5, 2026 · Artificial Intelligence

What Gaps Must Spatial AI Agents Fill to Achieve Action in 2026?

The article analyzes spatial intelligence as a core AI frontier, outlines the 2026 bottleneck of agents lacking spatial‑scale capabilities, reviews recent industry and academic advances such as World Labs' Marble model, hierarchical memory, GNN‑LLM integration, and world‑model research directions.

2026 AI ResearchAgentic CapabilityGNN-LLM Integration
0 likes · 7 min read
What Gaps Must Spatial AI Agents Fill to Achieve Action in 2026?