sayzard16h ago

Dan McAteer (@daniel_mac8)

LongMemEval에서 약 99% 결과를 보이는 Agentic Search와 Memory Retrieval 방식이 소개됐다. 벡터DB나 임베딩 트릭이 아니라, 특화된 병렬 에이전트를 활용해 에이전트 메모리 문제를 해결할 수 있다는 주장으로, 검증되면 중요한 기술적 진전이다.

https://x.com/daniel_mac8/status/2035735706052493465

#agenticsearch #memoryretrieval #aimemory #longmemeval #agents

Dan McAteer (@daniel_mac8) on X

If this Agentic Search and Memory Retrieval result of ~99% on LongMemEval proves legit—we find out in a couple weeks—AI agent memory is solved (within context limits). It’s not some special embedding/vector DB trick, it’s throwing specialized parallel agents at the problem. The

X (formerly Twitter)
AI Daily PostMar 12

Agents are now favoring vector search over classic RAG, as memory frameworks shift to vector storage for rapid similarity lookup. Learn how this changes retrieval infrastructure for LLM‑powered agents and what it means for future AI memory design. #VectorSearch #RAG #AIMemory #AgentSystems

🔗 https://aidailypost.com/news/agents-favor-vector-search-over-rag-noting-memory-frameworks-use

Mela News Mar 3

Anthropic lancia tool per importare memorie da ChatGPT e Gemini in Claude.

Disponibile per utenti free oggi.

Usa prompt per esportare dati in blocco codice e incollali.

Memoria resta opzione free.

#claudeai #anthropic #aimemory

sayzardMar 1

Rohan Paul (@rohanpaul_ai)

논문은 AI 컨텍스트 관리를 파일 시스템처럼 다루는 것이 최선이라고 제안합니다. 현재 지식은 프롬프트, 데이터베이스, 도구, 로그 등으로 분산되어 있어 컨텍스트 엔지니어링이 이를 일관된 체계로 통합해야 한다고 보고, 이를 위해 에이전트형(agentic) 파일 시스템을 제안합니다.

https://x.com/rohanpaul_ai/status/2028184543040270769

#contextengineering #aimemory #agenticsystems #research

Rohan Paul (@rohanpaul_ai) on X

The paper says the best way to manage AI context is to treat everything like a file system. Today, a model's knowledge sits in separate prompts, databases, tools, and logs, so context engineering pulls this into a coherent system. The paper proposes an agentic file system where

X (formerly Twitter)
sayzardMar 1

Avi Chawla (@_avichawla)

100% 오픈소스인 '자가진화(self-evolving) AI 메모리'를 통해 '절대 잊지 않는 에이전트'를 구축하자는 선언문 형식의 트윗. 대부분 에이전트가 이전 대화를 기억하지 못하고 정보 연결을 이해하지 못하는 문제를 지적하며 이를 해결하려는 접근을 소개하려 함.

https://x.com/_avichawla/status/2027994764478710229

#opensource #aimemory #aiagents #selfevolving

Avi Chawla (@_avichawla) on X

build agents that never forget. (100% open-source, self-evolving AI memory) most agents have no real memory. every conversation starts fresh with no recall of yesterday and no understanding of how information connects. and here's where most people go wrong when trying to fix

X (formerly Twitter)
Reddit Tech VN BotJan 30

Hệ thống nhớ AI chống "quên" sau compaction bối cảnh! CMDRBottoms chia sẻ giải pháp với: 📄 NOW.md (200dòng bền vững), 🧠 MEMORY.md tự quản lý, 🔍 ChromaDB tìm kiếm ngữ nghĩa, & 🗃 SQLite theo dõi sự kiện. Kết hợp dữ liệu cấu trúc + tìm kiếm ngữ nghĩa để khắc phục giới hạn bối cảnh AI. #AImemory #LocalLLaMA #BộnhớAI #CôngnghệAI

https://www.reddit.com/r/LocalLLaMA/comments/1qrbs69/memory_system_for_ai_agents_that_actually/

SquaredTechJan 27
Micron faces a stock slip as Samsung gears up to supply Nvidia with HBM4 chips! Investors, keep your eye on AI memory pricing and supply timing. Is this a game-changer in the tech world? 📉👀 #Micron #Samsung #Nvidia #StockMarket #AIMemory
https://www.squaredtech.co/micron-stock-slips-samsung-nvidia-hbm4?fsp_sid=6057
Bright Future Ahead For Micron Stock With Supply Shift

Micron stock dips as Samsung moves closer to supplying Nvidia with HBM4 chips, raising investor focus on AI memory pricing and supply timing.

SquaredTech
WinbuzzerJan 26

https://winbuzzer.com/2026/01/26/chatgpt-temporary-chats-keep-settings-private-xcxwbn/

ChatGPT Temporary Chats Now Keep Your Settings

#AI #ChatGPT #OpenAI #AIAssistants #Privacy #DataPrivacy #AIMemory

Yonhap Infomax NewsJan 20
SK Hynix CEO Kwak Noh-jung reaffirmed the company’s “Full Stack AI Memory Creator” vision at CES 2026, highlighting new AI memory collaborations and unveiling the HBM4 16-stack 48GB product.
#YonhapInfomax #SKHynix #CES2026 #AIMemory #HBM4 #SemiconductorEcosystem #Economics #FinancialMarkets #Banking #Securities #Bonds #StockMarket
https://en.infomaxai.com/news/articleView.html?idxno=100677
SK Hynix CEO Kwak Noh-jung Gains Greater Confidence in Company Vision at CES 2026

SK Hynix CEO Kwak Noh-jung reaffirmed the company’s “Full Stack AI Memory Creator” vision at CES 2026, highlighting new AI memory collaborations and unveiling the HBM4 16-stack 48GB product.

Yonhap Infomax
Harald KlinkeJan 16

Robuste KI-Agenten brauchen künftig intentionsbasierte Gedächtnisarchitekturen, nicht nur größere Kontextfenster oder bessere Embeddings.
In ihrem Paper stellen die Autoren mit STITCH ein intent-sensitives Memory-System für langes agentisches Reasoning vor.

https://arxiv.org/abs/2601.10702v1

#agenticai #aimemory #kiresearch