“FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention” With how Long context LLMs are being bottlenecked by KV cache, because every old token keeps consuming GPU memory even when most of it is irrelevant, this paper turns long context into
MACHINE LEARNING
-
Need to Integrate VLAs and Models in Production Robotics
By
–
It's not VLAs versus World Models; production robotics needs both, in addition to model-based methods, all integrated by agentic coding. At ICRA last week, I presented a perspective on the divisions in our field, including
-
Gary Marcus questions unfalsifiable ‘real deal’ claim about LLMs
By
–
what does this even mean, @dwarkesh_sp, “the real deal”? is it even a falsifiable conjecture? what’s the evidence?
— Gary Marcus (@GaryMarcus) 9 juin 2026
and if you can agree that “If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it”, why acknowledge the… https://t.co/4DtoGvFCCJwhat does this even mean, @dwarkesh_sp
, “the real deal”? is it even a falsifiable conjecture? what’s the evidence? and if you can agree that “If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it”, why acknowledge the -
Claude Code battery drain solved by server-based coding
By
–
Nope, Claude Code sucks battery regardless, using it on my laptop I'd have to charge it after 3 hours! I only use Claude Code on VPS via Termius SSH now Now I only need to charge my MacBook daily, and I can close my MacBook whenever Nice benefits of AI coding on the server!
-
Gemini 3.5 Live Translate: natural real-time voice translation
By
–
What makes Gemini 3.5 Live Translate a massive step forward isn’t just the speed, it’s how it maintains natural pacing, pitch, and intonation over long conversations.
— Harold Sinnott 📱 (@HaroldSinnott) 9 juin 2026
Finally, real-time speech translation that actually sounds human. 🌍@GeminiApp @GoogleAI #AI https://t.co/uwSA1fhsCX pic.twitter.com/ZOTbgaBtw4What makes Gemini 3.5 Live Translate a significant advance is not just the speed, but the way it maintains a natural rhythm, pitch, and intonation over long conversations. Finally, a real-time voice translation that sounds
-
VQAScore: open-source framework for evaluating text-video/image/3D models
By
–
VQAScore, a simple framework for evaluating text-video/image/3d models is opensource and got many many features. Packaged with leading frontier and commonly used open-source multimodal models and simple/clean eval interface. https://t.co/8xy7SjGEqt pic.twitter.com/wrt6wBTQvQ
— Jean de Nyandwi (@Jeande_d) 9 juin 2026VQAScore, a simple framework for evaluating text-video/image/3d models is opensource and got many many features. Packaged with leading frontier and commonly used open-source multimodal models and simple/clean eval interface.
-
LangSmith Executive Assistant Template for Email and Calendar
By
–
LangSmith Fleet template spotlight: Executive Assistant Manages your inbox, calendar, and daily brief. An email agent that triages your inbox, briefs you every morning, schedules meetings, and learns your preferences over time. Try it today: https://
langchain.com/templates/exec
utive-assistant
… -
Claude Fable 5 now available on Databricks across AWS, Azure, GCP
By
–
.
@AnthropicAI
's Claude Fable 5 is now available on Databricks, governed by Unity AI Gateway, across AWS, Azure, and GCP! Fable 5 sets a new state of the art on OfficeQA Pro at 57.9% and is designed for work that was previously too complex or too long-running to hand to an AI. -
AI model predicts fire spread, redirects evacuees to safer exits
By
–
#AI model predicts building fire spread, redirecting evacuees to safer exits in real time
by National Institute of Standards and Technology @TechXplore_com Learn more: https://
bit.ly/4xcBdwo #ArtificialIntelligence #EmergingTech #Innovation #Technology #Tech -
Shower thoughts: autoencoders as good addition after LoRA
By
–
This was purely based on shower-thoughts :D.
As someone called out, autoencoders would have been a good addition after LoRA.