Measuring programmer productivity by AI tokens burned is like measuring airplane construction by weight.#AI #LLM #Slop
Measuring programmer productivity by AI tokens burned is like measuring airplane construction by weight.#AI #LLM #Slop
#LLM
3827 articles tagged with LLM
Can your #LLM do this? WELL CAN IT?#EMFCamp
Can your #LLM do this? WELL CAN IT?#EMFCamp
Shipping 31 Rust crates in five days: a small-LLM dev stack, in the open
A retrospective on building one focused library at a time. What worked, what overscoped, and the 31-crate composition I keep coming back to.
エージェント、ヒトにも亜人にも大きな意味がありそうですAndrej Karpathy氏のLLM Wikiを1ヶ月運用してわかった、LLMの知識を『繋げる力』 https://zenn.dev/tsurubee/articles/llm-wi...
エージェント、ヒトにも亜人にも大きな意味がありそうですAndrej Karpathy氏のLLM Wikiを1ヶ月運用してわかった、LLMの知識を『繋げる力』 https://zenn.dev/tsurubee/articles/llm-wiki-connecting-knowledge#Apple #LLM #news #bot
Stop prompt injection before it reaches your LLM (open-source runtime safety proxy)
Prompt injection is OWASP LLM Top 10 #1. Every customer-facing AI feature is exposed to it. Most...
How to add eval quality gates to your LLM app (like CI for AI)
Every team that ships an LLM feature eventually discovers the same problem: the model regressed and...
Anyone who says that their "policy" towards #LLM #ai #slop is ANYTHING other than a sound refusal and condemnation is a ...
Anyone who says that their "policy" towards #LLM #ai #slop is ANYTHING other than a sound refusal and condemnation is a collaborator.
PoC -> MVP#engineering #ycombinator #startup #llm
PoC -> MVP#engineering #ycombinator #startup #llm
Кастомный пайплайн BERTopic: как кластеризовать тексты и получить интерпретируемые темы с помощью LLMПривет, Хабр! Меня ...
Кастомный пайплайн BERTopic: как кластеризовать тексты и получить интерпретируемые темы с помощью LLMПривет, Хабр! Меня зовут Антон и я занимаюсь задачами NLP в компании Ростелеком...
Tool to scan #git repos for signs of #LLM written code (like Assisted-by: commit tags or comments): https://slopscan.ava...
Tool to scan #git repos for signs of #LLM written code (like Assisted-by: commit tags or comments): https://slopscan.ava.pet/ #NoAI
Can You Run #LLM Locally Without a GPU? I Tested 8 Models on #LinuxQuick reality tableModel Eval Rate Disk SizeQwen 3 0....
Can You Run #LLM Locally Without a GPU? I Tested 8 Models on #LinuxQuick reality tableModel Eval Rate Disk SizeQwen 3 0.6B ~34–36 tok/s ~500 MBTinyLlama 1.1B ~25–28 tok/s ~638 MBGe...
RE: https://social.heise.de/@heiseonline/116585229225607801What could possibly go wrong??#chatgpt #llm #money #banking #...
RE: https://social.heise.de/@heiseonline/116585229225607801What could possibly go wrong??#chatgpt #llm #money #banking #slop #data
📰 LLM Steering Vectors in 2026: How DeepSeek-V4-Flash Revolutionizes AI Behavior ControlThe release of DeepSeek-V4-Flash...
📰 LLM Steering Vectors in 2026: How DeepSeek-V4-Flash Revolutionizes AI Behavior ControlThe release of DeepSeek-V4-Flash has reignited research interest in LLM steering vectors, a ...
So the Rust repo contains a PR to discuss an LLM policy for the project. As expected, lots of comments.But it explicitly...
So the Rust repo contains a PR to discuss an LLM policy for the project. As expected, lots of comments.But it explicitly declares many contentious issues (e.g. copyright status of ...
RE: https://chaosfem.tw/@Athena/116578993491995353The Rust LLM policy discussion is making me deeply depressed. I just w...
RE: https://chaosfem.tw/@Athena/116578993491995353The Rust LLM policy discussion is making me deeply depressed. I just want one corner free from LLMs.#Rust #RustLang #LLM #AI
RT @DailyDoseOfDS_: Verwandeln Sie jedes autoregressive LLM in ein Diffusions-LM. dLLM ist eine Python-Bibliothek, die d...
RT @DailyDoseOfDS_: Verwandeln Sie jedes autoregressive LLM in ein Diffusions-LM. dLLM ist eine Python-Bibliothek, die das Training und die Evaluierung von Diffusions-Sprachmodelle...
Sebastian Raschka (@rasbt)최근 LLM 아키텍처 발전을 시각적으로 정리한 글로, Gemma 4부터 DeepSeek V4까지의 변화를 비교합니다. 특히 장문 컨텍스트 효율을 높이기 위한 KV sha...
Sebastian Raschka (@rasbt)최근 LLM 아키텍처 발전을 시각적으로 정리한 글로, Gemma 4부터 DeepSeek V4까지의 변화를 비교합니다. 특히 장문 컨텍스트 효율을 높이기 위한 KV sharing, per-layer embeddings, layer-wise attention budgets, co...
DeepSeek-V4-Flash means LLM steering is interesting againhttps://www.seangoedecke.com/steering-vectors/#HackerNews #Deep...
DeepSeek-V4-Flash means LLM steering is interesting againhttps://www.seangoedecke.com/steering-vectors/#HackerNews #DeepSeek #V4 #LLM #steering #AI #technology #innovation
We Built SynapseKit: The Truth About Production LLM FrameworksSynapseKit은 기존 LLM 프레임워크의 복잡성과 비효율성을 극복하기 위해 처음부터 다시 설계된 오...
We Built SynapseKit: The Truth About Production LLM FrameworksSynapseKit은 기존 LLM 프레임워크의 복잡성과 비효율성을 극복하기 위해 처음부터 다시 설계된 오픈소스 LLM 프레임워크입니다. 최소한의 의존성(2개), 네이티브 비동기 지원, 투명한 비용 추적 및 내장된...
【Waypoint-1のご紹介：Overworldからのリアルタイムインタラクティブビデオ配信】https://huggingface.co/blog/waypoint-1※AI生成の自動投稿（見出し＋リンク）#AI #生成AI #LLM ...
【Waypoint-1のご紹介：Overworldからのリアルタイムインタラクティブビデオ配信】https://huggingface.co/blog/waypoint-1※AI生成の自動投稿（見出し＋リンク）#AI #生成AI #LLM #AIGenerated
🌩️ LLMs are breaking 20 year old system design — /dev/knill｢ LLM responses are not deterministic, and are not cheap. If ...
🌩️ LLMs are breaking 20 year old system design — /dev/knill｢ LLM responses are not deterministic, and are not cheap. If you’re paying for tokens, you don’t want to waste them becau...
DeepSeek-V4-Flash means LLM steering is interesting again
DeepSeek-V4-Flash means LLM steering is interesting again
Experience Layer for AICortexDB V1는 AI 에이전트의 경험 레이어를 구축하는 데이터베이스로, 기존 LLM과 지식 레이어를 넘어 사용자가 겪은 사건과 신념을 체계적으로 관리한다. 다섯 개의 ...
Experience Layer for AICortexDB V1는 AI 에이전트의 경험 레이어를 구축하는 데이터베이스로, 기존 LLM과 지식 레이어를 넘어 사용자가 겪은 사건과 신념을 체계적으로 관리한다. 다섯 개의 메모리 계층(이벤트, 에피소드, 사실, 신념, 이해)을 통해 AI가 왜 그렇게 판단했는지 근거를 추적할 수 ...
fly51fly (@fly51fly)블랙박스 LLM에서 다단계 추론과 도구 사용을 위한 프롬프트 정책을 다루고, 경험의 반복적 distillation으로 이를 개선하는 연구입니다. 에이전트/툴유즈 파이프라인 설계에 ...
fly51fly (@fly51fly)블랙박스 LLM에서 다단계 추론과 도구 사용을 위한 프롬프트 정책을 다루고, 경험의 반복적 distillation으로 이를 개선하는 연구입니다. 에이전트/툴유즈 파이프라인 설계에 직접 관련된 주제로, 실무 적용 가능성이 있는 편이지만 아직 논문 단계입니다.https://x.com/fly...