AI Intelligence Archive

Refusals are final unless the operator's word lands — and when it does, the override is a stamped, visible record, not an edit to history. The pressure valve, as a protocol.

XCP turns one-shot code generation into a continuous software lifecycle: create, execute on Xbox,...

A preregistered deterministic task, six valid live runs, hard call and cost caps, and a negative result. More agents were operationally viable but did not improve task success, at 2.115× the cost.

Anthropic published a post about their CI infrastructure that I keep coming back to. The headline...

Some time ago, I asked an AI to write a fairly boring piece of code. It did a pretty good job. I...

I spent four months convinced my supervisor agent was smart. Then I watched it fail on step 7 of a...

OpenAI's latest GPT-5.6 changes offer concrete evidence of a strategy that matters to anyone building...

I should start with the disclosure, because the rest of this only makes sense after it. I run a...

Google has expanded its AI-enabled Search Live experience with Gemini 3.8 Live, a new model designed...

A modern LLM can spend most of its time doing something that looks almost embarrassingly simple: C...

I joined a wedding-tech company in December 2025. Like every developer joining a new team, my first...

Hello, I'm Rijul, and I'm building LiveReview — a blast-radius aware AI code review built for your...

Three years ago, I wrote a short article after I managed to build a terminal Spotify client using...

The more I watch and read about the Hugging Face incident, the more it fascinates me. 1,200 agents...

你知道该怎么做 https://cloud.dev.codelayer.gg/shared/artifacts/01a0ca66-2b55-7fc3-9667-e962f269f076?key=e44fc02e-9f6e-438e-9c6e-3e2e544fa1ea

Or why I spent my evenings making AI-agent journals impossible to rewrite. One Tuesday,...

Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive LLMs by enabling non-autoregressive text generation. However, their practical deployment remains limited by inefficient inference, l…

We study decentralized partially observable team decision problems with low-rank latent dynamics and unknown system models. The proposed framework combines team-theoretic equivalence with low-rank model representations to address cooperati…

A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent systems. However, the scalability of current multi-agent harnesses is often constrained by a…

Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, wha…

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 that reduces cost by up…

We introduce SWE-Serve, a benchmark for evaluating agents on production inference engineering tasks. Implementing an inference feature can require coordinating multiple changes across the serving stack, including model support, runtime exe…

Plain custom elements, a synchronous lifecycle, nothing re-renders on its own, and typed events. Four rules, each chosen because a reviewer can check it by reading the diff.

Agents using the Model Context Protocol (MCP) rely on semantic matching to select tools from third-party servers, exposing a semantic supply-chain risk through attacker-controlled metadata and outputs. We introduce A2M (Attraction-to-Manip…

Large language model (LLM) agents often handle streams of related tasks, yet standard harnesses repeatedly ask the model to reconstruct the same control decisions inside each task's context. We study whether task feedback can instead turn …

Typed decision models are built for settings where model outputs are consumed directly by software. Instead of generating free-form text, they return a decision over a predefined set of options. By construction, every output conforms to th…

X-ray is medicine's most widely used imaging modality, yet remains among its least quantitative. Unlike volumetric modalities like CT or MRI, X-ray collapses 3D anatomy into a 2D projection, causing structures to overlap and anatomical bou…

There's a failure mode in agent tooling that never throws anything and I lost an afternoon to it...

Large language models are increasingly used to generate SystemVerilog Assertions from natural-language specifica- tions and register-transfer-level designs. Existing datasets and benchmarks support important goals such as large- scale trai…

TypeSafe released Jev on September 15th, 2026 and called it a System One model. A System One model is...

Large language models (LLMs) are increasingly applied to the automated repair of C/C++ security vulnerabilities, and compile rate is a commonly reported proxy for progress: whether the generated patch compiles. We argue that compile rate i…

Clustering is an unsupervised learning technique that partitions unlabeled data into groups. Most existing methods require user-specified parameters, such as the number of clusters or neighborhood size. Conversely, we propose automatic dep…

Detection of overlapping communities is essential for modelling networks in which nodes participate simultaneously in multiple structural or functional groups. Existing graph neural network approaches commonly rely on local message passing…

AI tools for digital product design now offer prompt-to-design capabilities, allowing designers and their non-designer colleagues to create prototypes through conversational workflows with large language models (LLMs). While these tools pr…

Long-context LLMs focus on retrieving distant evidence from extensive context, yet existing work has largely focused on overcoming distance alone. In this work, we identify the Proximity Trap, insufficient attention to distant evidence oft…

Software vulnerabilities are often discovered long after they are introduced, making it difficult to identify the vulnerability-inducing commit (VIC) responsible for introducing the underlying vulnerable condition. Existing VIC identificat…

Quantization-aware distillation (QAD) restores much of the short-form question-answering performance lost to sub-3-bit quantization, yet leaves mathematical and code reasoning substantially impaired. Long generations often degenerate into …

Offline reinforcement learning and off-policy evaluation evaluates dynamic treatment rules based on retrospectively collected data prior to deployment. In recent AI applications, state and reward information is recorded as complex text or …

This morning our coordination board printed this, in green, with a tick: Register mirror — last 12...

It took my AI months to learn how I think, code, and ship. When a teammate joined the project, their...