AI Intelligence Archive

Automated identification of individual animals is moving from small curated studies toward shared benchmarks and reusable foundation models. This progress is useful, but current evaluation leaves the evidence behind a match unresolved, i.e…

Hierarchical text embedding models offer variable-length representations, while quantization reduces numerical precision; both aim to improve computational efficiency. This work is an empirical benchmark study of how these two orthogonal, …

Recursive self-improvement requires AI systems to improve the models and infrastructure used by subsequent systems. LLM compression offers a concrete efficiency objective: retain a pretrained model's behavior within a smaller model storage…

We introduce LongBanana, to our knowledge the first expert-designed and expert-verified benchmark for long-context multi-reference image synthesis in the age of agent-driven image generation. The task requires retrieving dispersed visual e…

Computer-use agents (CUAs) have recently achieved superhuman performance on several long-horizon benchmarks. However, their computational cost limits their use in long-horizon tasks and real-time domains such as video games. A key bottlene…

Common methods for monitoring features in large language models (LLMs) such as linear probes and sparse autoencoders (SAEs) assume that features maintain a consistent representation across contexts. LLMs, however, can flexibly adopt differ…

Faithfulness requires that algorithmic explanations accurately reflect the model's behavior. Yet existing operationalizations of faithfulness can disagree about whether the same explanation is faithful, leaving developers and auditors to c…

Grain growth during heat treatment strongly affects the microstructure of polycrystalline materials and, consequently, their mechanical properties. Although numerical methods such as front-tracking or front-capturing simulations can accura…

On 28 October I posted about ClaudeArena, the free practice engine I built for Anthropic's Claude...

By Spacexcx · November 2026 I built a small tool that reads a git diff and flags possible logic...

A content management system can make routine SEO work either manageable or unnecessarily dependent on...

AI-generated search summaries may answer the first question, but they do not necessarily settle the...

Forums and community threads were the most trusted and most-clicked research format in Fractl’s Quiet...

Most multi-model routers treat dispatch as a winner-take-all classification problem. A study across 25 benchmarks shows why that destroys performance, and how outcome regression fixes it.

AI-generated summaries may be the first thing many searchers read, but they are not necessarily the...

Google Classroom and Search have different AI rollouts. Schools can choose Search separately; its AI-free configuration remains unpromised.

Fayaz argues why fundamental human knowledge is important, no matter how advanced AI becomes.

Reddit plans to begin fully testing a logged-in-only experience in select markets outside the United...

Sam Altman is running a betting pool on the first one-person billion-dollar company. Meanwhile, a randomized trial found experienced devs were 19% slower with AI while believing they were 20% faster. What solo AI leverage actually survives…

Hey there, dev team! I'm off to Athens for CityJS Athens tomorrow, so this week's post is a little...

The 4K checkbox costs one credit. 1K is 3 credits, 2K is 4, 4K is 5. So the jump from a screen-sized...

Code reviews are a cornerstone of healthy software engineering teams. They catch bugs, share...

Code review is broken. Between context switching, subjective nitpicks, and the ever-growing backlog...

Chances are you're already using open-source software without realizing it. In this episode of...

Most teams spend a year perfecting an AI feature and ten minutes deciding what to charge for it. Then...

Most tutorials about vocal separation start with "run Demucs on your GPU". That is a fine answer if...

Short-interval wireless schedulers must return a complete, feasible beam-and-resource allocation before each execution deadline, and a late or invalid output gives no utility in the current control cycle. Searching longer can increase tran…

We study non-parametric bandits in which arms are evaluated using Cumulative Prospect Theory (CPT) rather than expected reward. We consider continuous action spaces with arm-dependent noise and propose Non-Parametric CPT Bandits ($\mathtt{…

Model merging combines separately fine-tuned experts into a single model without retraining. Standard practice trains every expert at the same LoRA rank, which couples the rank choice across experts even though each domain has its own best…

While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth $K$ and interaction depth $T$ via $Pass@(K,T)$,…

If you run long Claude Code sessions you know the moment. The context hits the auto-compact...

Graph-based retrieval-augmented generation (GraphRAG) answers questions by retrieving over a knowledge graph built from a corpus. Yet retrieval strategies are designed around the structure their own construction pipeline produces, and whet…

Setting. This is an empirical characterization study; we propose no new method and run no comparisons against alternative objectives. We fine-tune Qwen3.5-4B with sparse LoRA and length-normalized DPO plus an SFT retention term (the RPO ob…

Let $d_1,\dots,d_m$ be directions in $\mathbb{R}^k$, where $k\ge 2$. A multiset $X$ of $n$ points of $\mathbb{R}^k$ projects to the $m$ multisets of numbers $\{\{x\cdot d_j : x\in X\}\}$, or equivalently to the $m$ sorted lists of these nu…

Microfluidic droplet generation is critical for many applications, yet efficient device design in high-dimensional parameter spaces remains computationally prohibitive due to complex multiphase flow dynamics. This work presents a unified m…

During my vacation, I decided to spend some time learning more about the Model Context Protocol, or...

Background/Objectives: CNNs perform well in retinal disease classification from spectral-domain optical coherence tomography (SD-OCT), but whether EfficientNet's natural-image scaling hierarchy transfers to medical-image transfer learning …

Traditional document parsing systems use Markdown as their output representation, which is effective for linear text but does not fully preserve fine-grained spatial structure, table geometry, visual hierarchy, or rendering semantics. We s…

Long-horizon language-model agents accumulate increasingly large interaction histories, making context compaction essential for efficient execution. Existing learned compaction methods often rely on downstream task outcomes or solver behav…

Skills let language agents reuse experience across tasks as natural-language documents. Recent work evolves skills from interaction, but existing methods have limitations when skills evolve online. Prior methods update a skill offline from…