15495 articles tagged with LLM
Large language model (LLM) agents increasingly rely on external memory stores to retain information across conversations, typically via dense vector embeddings searched with approx...
まさかAppleが……艦長にまた怒られそうですApple、再設計されたヘルスケアアプリの対応言語を拡大 https://www.macotakara.jp/news/entry-52222.html#Apple #LLM #news #bot
Open models can do 90% of #agentic work at a twentieth of the cost of a frontier #LLM: Gartner“Nine in 10 agent workflow steps are routine, and open-weight models handle them at pa...
Someone asked: can we end the Globex contract early? Seven words. The question looked short and...
A central yet under-explored question in Multimodal Large Language Models (MLLMs) is where to connect---how visual information from a hierarchical vision encoder should be wired in...
防御側の LLM 活用：ログ分析・アラート一次対応・脅威インテリジェンスで「ログを入力として疑う」設計https://qiita.com/shalXXXX/items/5920c6ba75c2f1203f1b?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items#qiita #...
The key problem is the vague, anthropomorphic language. Saying an LLM "understands" how to do something carries so much human baggage, and it's meaningless in the field of machine ...
RE: https://tomkahe.com/@GiftArticles/117554787184569999. . . because there is no #search, only #LLM variations on #autocomplete
"In particular I strongly recommend that authors should not fall into the trap of getting the LLM to write a draft."#ai #writing #vs #editinghttps://martinfowler.com/bliki/LLM-Voic...
Korben: #Survival #LLM, l' #IA de survie qui tourne sans #internetEt les #Linuxiens ?Ils n'ont pas le droit de survivre... ??C'est mal ...korben.info/survival-llm...
Korben: #Survival #LLM, l' #IA de survie qui tourne sans #internetkorben.info/survival-llm...
-"But why would anyone use an LLM to do the fun computery stuff!!11"I see this often from anti-AI fanatics, so let me give you an example from five minutes ago:Troed wanted to upgr...
Release: llm 0.37.1 A bugfix release for LLM, with contributions from Steven Huang, Tao An, Arjan Mossel, and XQ. Tags: llm
Show HN: Uncracked City, detective game where you question LLM suspects that lie
#AI is #LLM made visible.
Teleoperation remains the default way to collect robot demonstrations, but every episode costs robot and operator time, and every new task starts from scratch. We ask whether a cod...
Can a language model reason without writing its steps down? Coconut feeds GPT-2's last hidden state back in as the next input instead of a token, and beats written chain of thought...
The comments are vicious. Schadenfreude is a complex emotion.https://github.com/openai/codex/issues/59060#AI #LLM #Slop
llamafile Systemd templatehttps://solvecomputerscience.substack.com/p/llamafile-systemd-template#ai #llm #llamafile
One hundred thirty-one thousand numbers. That is the entire size of the edit that moved Qwen3.5-0.8B, a small open reasoning model, from 7.9% to 48.1% on MATH-5
A-K-M-E – LLM + Credenze + Inferenze Piano Terra, viernes, 13 de noviembre, 21:00 CETA-K-M-E – LLM + Credenze + Inferenzeallucinazioni, riverberazioni e pressione conversazionale13...
it seems like #bigTech stochastic parrots as a service (SPaaS #LLM) are mainly targeting solving the huge problem the petrodollar economy may be facing:the potential reduction in t...
Show HN: Interactive notes on RLHF, attention, LoRA and LLM evalshttps://concepts.laxmena.com/#llm
Как я оформляю тренировки и питание в Obsidian с помощью LLM и что она теряла по дорогеЧто теряется, когда LLM превращает свободные записи в таблицы: 360 запросов к трём моделям, п...