Articles & Guides

NEEDLE is a training-free defence that targets a known backdoor direction while preserving refusal-related representations and limiting changes to normal model behaviour.

KaliBench targets a neglected part of cybersecurity AI: translating analyst intent into precise, executable Kali Linux commands. Its fine-grained dataset and multi-stage verification separate tool selection from argument construction.

Rules to Tools converts written scientific requirements into callable executable checks for coding agents. The study finds higher repair scores on some SciCode cohorts, but the gains are task-dependent and can come with higher CPU usage.

World Observer separates acting from observing by jointly generating an agent-centric actor view and one or more panoramic observer views. This lets out-of-view objects continue evolving and improves their state and appearance when they re-enter the actor’s view.

GraphForge introduces a framework that grounds agent tasks and evaluation rubrics in real files connected through an evidence graph. Its reported results suggest that traceable verification can improve training for agents that handle files, tools, and deliverables.

PyRUA-Lean replaces repeated tool calls with executable Python cells that compose robot primitives, perform local checks and request only necessary observations. Across 700 simulated instances, it raised success from 63.1% to 71.7% under the same LLM-call budget while cutting input tokens by 65% on jointly solved tasks.

Argo-Bench moves beyond text-to-SQL by testing whether agents can reconstruct facts from a large enterprise warehouse, analyze them, and take actions with measurable consequences. Its results suggest that current models still struggle with end-to-end business decision-making.

LOCI combines key-value caching with recurrent linear memory to help video world models reproduce previously observed regions when a camera returns. Its geometry-aware memory improves scene retrieval while reducing the cost of processing long videos.

ScholarCatalyst moves scientific retrieval beyond topical similarity by asking which earlier papers could genuinely advance a research project. Its author-annotated results show that current agents still struggle to match expert judgment.

HeteroFold targets a key bottleneck in heterogeneous multi-agent LLMs: the receiver repeatedly prefills context that the sender has already processed. The paper reports substantially faster cross-family cache transfer, including a 10.7× speedup over native prefill for one 32K-context direction.

MemFold compresses query-relevant textual memories into a fixed set of continuous vectors and optimizes them through the behavior they enable. Its training combines task-level rewards with confidence-gated on-policy distillation from a frozen textual-memory teacher.

A new study examines whether text-only interfaces can be used to infer and influence the generation behavior of aligned language models. Its approach concentrates costly interventions on a small number of high-risk positions.

Stability AI is shifting its center of gravity from image generation to tools for music professionals. The company has raised funding from Sony, Warner and Universal, alongside licenses to use their catalogs for training.

Meta has released code and SDKs that let developers connect its Muse AI agent to displays, Raspberry Pi boards, ESP32 hardware, and other components. The company is also testing a limited Muse Home Link device for home-control tasks.

Amazon says it will invest more than $1 billion in communities near its data centers over five years and end the use of NDAs for new projects. Environmental advocates welcome the transparency measure but argue that the plan sidesteps pollution, water use, and backup-power concerns.

PyTorch’s integration of Helion into vLLM’s linear backend explores whether a high-level kernel DSL can replace part of the manual specialization traditionally required for quantized GEMM. On NVIDIA Hopper, shape-specific tuning and hybrid dispatch produced end-to-end gains across the evaluated workloads, exceeding 10% in some cases.

US prosecutors accuse Earthmade Computer CEO Greg Lui of using false paperwork and Asian transit routes to divert Nvidia GPU servers to China. The arrest highlights how difficult it is to enforce export controls once chips are embedded in complete systems.

Apple says it will add stronger controls around macOS Full Disk Access as desktop AI agents become capable of handling files, messages, mail, and browsing data. The change is aimed at clearer, more explicit consent rather than a blanket reduction of the permission.

OpenAI’s Dots platform combines a cloud computer, software access, desktop control, and voice interaction. Hands-on testing suggests it is more useful for controlled creative and work tasks than for shopping, account management, or fully automated web errands.

AI is gaining political attention and increasingly friendly product interfaces, yet consumer willingness to pay remains limited. TechCrunch’s latest Equity episode examines why enterprise customers still matter more to the AI economy.

HC-DLM combines discrete token readouts with a persistent continuous latent state to address broken dependencies in parallel diffusion decoding. The paper reports gains over matched-size discrete and continuous diffusion baselines on Sudoku, Countdown, and LM1B.