long
58 stories and discussions about long, aggregated from every source we track.
The play store does not run on my old devices, this led me accidentally experience some joy while using Android.
If you checked my profile recently, you probably noticed my page has been completely quiet since...
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Contribute to karmanyaahm/budget-tools development by creating an account on GitHub.
We've come a long, long way together. You changed. Where do we go from here?
Our last post on MAME's failure to build on Power yielded some interesting information in the comments which, in my copious spare time, see...
Obscura is now available as a native app on Linux, with official packages for Debian, Ubuntu, Fedora and Arch
What the durability and interface design of Bose QC25 headphones can teach us about building long-lasting software.
Achieving conformance with the C++ Standards has been a long road for the Visual C++ team. If you've seen us at any conferences lately, you've probably
Short article announcing my purchase of a Slimbook Executive 14-inch laptop, following the battery bulge problem in the five-year-old Slimbook Pro2, with the new device intended as a replacement, covering…
As a Long-Term Support (LTS) release, Qt 6.12 is a culmination of various efforts on cybersecurity, UI performance, and much more across the tech stack.
One open Norwegian mountain, from the snow line to the fjord. Real people, the mountain’s own weather, rare skies. Free in your browser.
NumPy on Emscripten-forge now links OpenBLAS in WebAssembly, with large speedups for matmul and linear algebra in the browser.
First, you want to form a habit. Second, you want to operate at peak productivity during your session. Third, you want to minimize the amount you forget between sessions.
I've been writing long before AI, starting in 2015 to publish on this very blog, and then AI came, and I still write manually.
Some observations on a fast-moving calamity
The return of AMD's ambidextrous strategy?
I often read about getting LLMs to work for a long time toward a goal. So, for example, Claude has a /goal command, and many of us (directly or indirectly) ...
The complete guide to AI agent memory: long-term memory for LLM agents, memory architectures from vector stores to knowledge graphs, and how to build one.
Long horizon language model agents continuously accumulate interaction history, increasing computational cost while making relevant information harder to preserve and reuse. Existing context management methods mainly…
How to prompt Opus 5.5, steer a long run, and check your results in Claude apps and Claude Code.
How to prompt Opus 5.5, steer a long run, and check your results in Claude apps and Claude Code.
The most repeated number in enterprise AI turns out to be a paraphrase of unnamed opinion surveys, quoted in a magazine profile of a vendor.
A lot of prior work addressed key-value (KV) cache selection and compression by sparse attention to enable long-context inference for transformer language models without excessive hardware budgets. We provide a new…
How long can you leave software development before you wake up in another world?
Agents often work on complex problems that require millions of tokens of context, which necessitates compacting across sessions due to limited context windows. We develop CliffCompaction, an autocompaction technique…
We followed 1,112 tables through the git history of nine open-source apps: 72.8% are still alive after five years, but half of the tables that die do so within three months.
A lightweight native desktop editor for writing and safely saving long prompts for coding tools. C++20, Qt 6 Widgets and Scintilla; local files, no AI service. - aaravchour/PromptPad
How much of SWE-bench Pro can 64 DeepSeek agents solve in a day?
Agents often work on complex problems that require millions of tokens of context, which necessitates compacting across sessions due to limited context windows. We develop CliffCompaction, an autocompaction technique…
An interactive atlas of human dispersal out of Africa over 300,000 years, a glowing ember reveal anchored to dated archaeological sites.
Efficient cloud environments for long-horizon agents
Trajectory-aware guardrails for LLM chatbots: catches the gradual delusion reinforcement that turn-local safety filters miss - nwjang/psychosis-guard
Agents that do the work and show it. Contribute to oguzbilgic/kern-ai development by creating an account on GitHub.
LLM agents are increasingly deployed in collaborative settings, yet long-term interaction may give rise to undesirable coordination. We study the emergence of collusion in a long-horizon multi-agent environment: two…
Some observations on a fast-moving calamity
From “Storing” to “Staying Current”: Why Agent Memory Needs a Shared Evaluation
Ruiyang Wang and colleagues present GAVEL, which verifies and repairs long-horizon LLM robot plans against an explicit graph world model holding object relation
Dell’s XPS Googlebook is just one of the devices Google needs to be great. In late 2010, an up-and-coming Google executive named Sundar Pichai took the stage in front of a bunch of reporters and explained that Google had a big new idea about laptops. In the future, he said, the only thing you'd need on your computer was a web browser - that was where all your apps, data, and connections would live going forward. Your computer didn't need much storage, Pichai explained, or any fancy hardware, because the cloud would do the work. The software designed for these new laptops, Chrome OS , was an enormous bet that the web would eventually overtake everything. Chrome OS (which eventually became ChromeOS ) has been, in many … Read the full story at The Verge.