This week in Multimodal

Back to Pixels: Lifting Latent Video Diffusion into Pixel SpaceNov 12
Unbundling Multimodal Vision EncodingNov 12 · ▲ 2
TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis ModelsNov 12 · ▲ 19

This week

Get the morning digest

The top stories in the fields you follow, at 08:00 your time.

Latest in AI and developer tools

Today's trail

updated 24m ago

Last hour Today

Pulse · cacm.acm.org

cacm.acm.org · Nov 13

ChatGPT’s new Wallet lets AI agents shop and check out for you

Pulse

Top picks this weekarXiv

arXiv

Nov 12

Back to Pixels: Lifting Latent Video Diffusion into Pixel Space

Frontier video generation models work in a VAE latent space that compresses video in both space and time. This compression made training tractable, but now limits high-resolution, controllable, and interactive generation -- doubling resolution quadruples token count, conditions apply to groups of frames rather than individual ones, and each response must wait for the rest of its multi-frame block. We present PixVid, a simple yet effective method that lifts a pretrained latent video model to directly generate every frame in pixel space. PixVid keeps the diffusion transformer and replaces the VAE decoder with a convolutional head that renders its outputs into pixels. We find that predicting clean video rather than velocity, together with a larger noise shift, is key to quality. Finetuned from Wan2.1-14B, PixVid is competitive with the latent baseline within $\sim 1.3$M training clips. Working framewise in pixels, it generates at $2\times$ higher resolution with the same per-frame token count, keeps depth-conditioned video editing aligned under fast motion, and responds to each action at the next frame. Project page: https://chenshuo20.github.io/PixVid/

The Decoder

Nov 13

Anthropic's Claude for Manufacturing plugs AI directly into factory software to automate industrial workflows

Anthropic's new Claude for Manufacturing plugin connects the AI directly to Autodesk Fusion, Oracle NetSuite, and Infor. Its 17 specialized skills let the AI draft quotes, compare suppliers, and place backup orders when deliveries are delayed, all after human approval. The…

Top picks

All arXiv Hugging Face GitHub Hacker News

Latest Top this week

GitHub

Nov 19

vsmutok/FuckCaptcha

Open field guide to 39 CAPTCHA types: identification signals, free and open-source bypass methods, and paid alternatives. Includes a searchable web version.

WSU-Carbon-Lab/refl-analysis

Analysis code for "Beyond the diffraction limit to track molecular states at buried interfaces" (resonant X-ray reflectivity of ZnPc thin films; free-tensor vs. ML-DFT slab models).

Today's trail

Most active this week

large language models · machine translationFrontier LLMs40 this week
gpu memory · gpusInference & serving39 this week
coding agents · coding agentagentsNew37 this week
agents · toolagents36 this week
code generation · language models llmsagents33 this week
radiology report generation · radiomicsclinical reasoning32 this week
long term memory · memory constructionFrontier LLMs32 this week
indirect prompt injection · prompt injectionagents31 this week
pose estimation · camera poseMultimodal29 this week
question answering · answeringFrontier LLMs29 this week

Rising topics

Rising tools

openclaude+176% downloads · 5.8k/wk+176.1%
better-auth+111% downloads · 1.6M/wk+111.4%
Open WebUI+107% downloads · 388k/wk+107.4%
agent-native+95% downloads · 79.6k/wk+94.7%
DesktopCommanderMCP+75% downloads · 724.3k/wk+74.8%
remotion+105% downloads · 160.7k/wk+105.1%
OpenUI+117% downloads · 722.9k/wk+116.9%
notion-mcp-server+97% downloads · 387.2k/wk+97.3%

All tools