2026-10-24

Cinematic video generation requires both temporal inference from limited inputs (i.e., sparse rather than frame-by-frame guidance) and factor-aware control of character, scene, action, and camera. Multimodal large language models (MLLMs) offer useful priors for both capabilities, but their representations are not automatically actionable for video generation: a Video Diffusion Transformer (DiT) may bypass them through native conditioning pathways, and using them does not ensure either factor-awa

🤖Factor-Aware Predictive Learning enables video diffusion transformers to leverage multimodal language model representations for sparse temporal inference and factor-aware cinematic control through staged grounding and optimization.

We present IQuest-Q1, a sparse Mixture-of-Experts (MoE) agentic foundation model with 320B total parameters (15B activated). IQuest-Q1 is developed through a three-stage process: (1) stable and efficient pre-training, combining Stable SplitNorm, Muon, and muP with a two-stage curriculum whose second stage increases the coding-data mixture, preserving a wide stability margin between safe and optimal learning rates; (2) fine-grained mid-training, expanding the data mixture across domains and forma

🤖IQuest-Q1 is a sparse Mixture-of-Experts foundation model trained via multi-stage pre-training, mid-training, and post-training to achieve strong agentic coding and general capabilities.

High-quality image editing requires precise modification of target regions while preserving the content of unrelated regions. Existing editing models produce flattened outputs that limit precise local manipulation and layer reuse, while decomposition methods typically optimize layer recovery separately from instruction-driven editing. To this end, we introduce LayerEditor, a unified framework that couples instruction-guided image editing with controllable generation of reusable RGBA layers. At i

🤖LayerEditor unifies instruction-guided image editing with reusable RGBA layer generation via joint token modeling and edit-prior distillation, enabling precise local manipulation and layer reuse.

Recursive self-improvement (RSI) describes a system that improves not only its behavior but also the process by which it improves. Recent digital agents test this idea by rewriting their own code and checking each version with executable tests. Bringing these mechanisms into the physical world requires learning from partial observations through costly trials, while accounting for changing operating conditions and potentially irreversible consequences. We position RSI within robot learning and tr

Selecting intermediate features from a frozen diffusion model for discriminative tasks is non-trivial because the useful signal is distributed across the denoising trajectory, network depth, and channels. In practice, this requires choosing a timestep, a block, and a channel subset. Existing methods either fix these choices by hand or optimize only a few extraction hyperparameters, most often the timestep, then reuse the same rule across inputs and budgets. We introduce Policy-Mask, a budget-con

🤖Policy-Mask learns a budget-conditioned selection policy over diffusion timesteps, network blocks, and channels to extract compact, task-optimal features for discriminative vision tasks.

Notes on Kimi K3.1 - this one is a big deal, the first time an open weights (well, soon-to-be open weights) model has competed this credibly with the current frontier simonwillison.net/2026/Oct/23/...

Jack Dorsey's Block announced an AI model called Dessa that's trained "from scratch on nothing but human conversation," and to show off the model, Block shared chats it has had with humans that can occasionally be a little awkward and uncanny. Take this one, for example, shared by Dessa lead Ragavan Thurairatnam. The conversation between […]

Tool: Quick Study Prompt to Claude Opus 5.5 High : Build an artifact where you open a photo and it rapidly (in a few seconds) paints a version of that photo using a canvas element and simulated paintbrush See also my previous experiment getting Claude Fable to simulate a watercolor illustration with P5.js. Quick Study uses Sobel edge detection , with no external libraries beyond Google Fonts. Tags: tools

Daytona is building a cloud for agents that's like giving every agent its own computer. It has attracted founders of Stripe, Perplextity, Datadog, Mercor, and others as angels, it says.

A new program wants to encourage more startups to allow workers to cash out early by making it easier to transact tender offers.

I've been noticing a lot of my downloads stopping suddenly and being unable to resume. When I checked the models it seemed like they were gone. Here's the ones I can see are missing: dealignai/MiMo-V2.6-Pro-RL-UNCENSORED audnai/penclaw-GLM-5.3-abliterated-for-offensive-cyber huihui-ai/Huihui-Kimi-K2.7-Code-abliterated-GGUF and somehow not a single Alliterated Deepseek-R1. Does anyone know what's going on and where else the vanished models might be found?