2026年10月6日 技术热点总结

项目简介：An open-source, clean-room reimplementation of Adobe Photoshop in pure Rust

技术栈：Rust、adobe、adobe-photoshop-2026、adobe-photoshop-2026-ai、art、image-editing、image-editing-software、image-editor、images

核心介绍：PhotoCraft Image editing; an open-source, clean-room reimplementation of Adobe Photoshop, rebuilt in pure Rust. Layers, masks, adjustment layers, layer styles, type, vectors, brushes and real PSD files, in a native app written entirely in Rust. Open source, offline, and yours. PhotoCraft on getartcraft.com · ArtCraft · All Crafting Apps A caption card with a drop shad…

项目简介：Turn off Apple Intelligence on macOS 27 and get its disk space back. One command, fully reversible.

技术栈：Swift、apple-intelligence、cli、debloat、macos、macos-27、privacy、siri、swift

核心介绍：Turn off Apple Intelligence on macOS 27 and remove its downloaded models. macOS 27 no longer has a single switch for Apple Intelligence, and its models stay on disk after the features are turned off. RemoveMacAI turns the features off, removes the models and prevents macOS from downloading them again. All changes can be reverted. MacRumors, AppleInsider and Help Net Security have written about R…

项目简介：Kill shader compilation stutter before you play. Compiles your games’ shaders into the GPU driver’s cache ahead of time. Windows, NVIDIA and AMD.

技术栈：C#、amd、directx12、nvidia、pc-gaming、shader-compilation、stutter、windows

核心介绍：SCSKiller Shader Compilation Stutter Killer. SCSKiller compiles your games’ shaders into your GPU driver’s cache before you play, so the game doesn’t stop to compile them mid-game. Free and open source, for Windows. Website Download Patreon Discord X Report a bug Request a game > The only official downloads are this repository’s Releases. > See of…

Humans often solve spatial problems by mentally simulating visual transformations. In contrast, conventional vision-language models (VLMs) reason primarily through language. We investigate whether VLMs can solve spatial problems by reasoning with both text and generated visual states. To this end, we introduce WM-VLM, which equips a pretrained VLM with a lightweight world model branch for generating intermediate visual states. Our two-stage training first teaches the model to generate the next visual state and then to use that state for reasoning. We programmatically construct spatial reaso…

Large language models (LLMs) increasingly serve as general-purpose interfaces to factual knowledge, but their parameters do not automatically reflect information that changes after pretraining. Knowledge editing (KE) provides a targeted alternative to costly retraining by modifying selected knowledge and preserving unrelated knowledge and general capabilities. Conventional KE uses structured factual triples, whereas unstructured KE (UKE) uses free-form passages containing multiple facts. Nonetheless, existing UKE editors exhibit a failure mode known as context reliance: edited LLMs can ofte…

Terminal agents are increasingly deployed beyond software engineering in science and other specialized domains. Constructing training environments requires executable reference behavior and a domain-specific verifier that distinguishes semantic correctness from superficially plausible artifacts. Authoring these components for each task requires repeated engineering and limits reuse. We introduce software-in-the-loop reconstruction, a self-supervised framework that obtains reference outputs and verification targets from existing software workflows, executable programs mapping structured inpu…

We introduce the Extender, a log-structured variant of the standard Transformer architecture. In a standard Transformer, each layer communicates with subsequent layers exclusively via the residual h, a superposition channel. The Extender adds a concatenation channel x: each layer ell emits both a residual update δ_ell which is added to h, and a much smaller extension ε_ell which is appended to x. While both the FFN and q see h, the attention kv projections take only x as input. As a result, the fully extended x contains the complete input for the kv projections of all layers, reducing the p…

Inspired by human vision, we introduce a framework using active gaze to enable fine-grained bimanual manipulation with only a single stereo camera. EyeRobot 2.0 physically attends to a 3D fixation point in the scene by swiveling two eye viewpoints to center their gaze on it. The resulting images are processed foveally by allocating more visual tokens to the image centers, focusing computation on task-relevant features. Such Active Visual Fixation (AVF) requires carefully coordinated gaze during task execution, which we accomplish hierarchically by first training a low-level gaze servoing po…

We introduce 4DCodeBench, a benchmark for 4D inverse graphics through code generation, in which agents reconstruct dynamic scenes from video as executable graphics programs. To accomplish this, agents must translate visual observations into compact representations of scene structure and dynamics, by implementing abstractions such as physical simulations to reproduce complex behavior. To evaluate this capability, we curate a set of real-world videos and construct synthetic scenes spanning diverse physical phenomena, including deformation, fluid flow, and fracture. We perform extensive benchm…