This week in Multimodal

Occlusion for Free: Detecting Occlusion from the Disagreement Between Two Monocular Distance CuesNov 4
PixelSpline: Continuous Space-Time Video Super-Resolution via Endpoint-Anchored Shared Pixel TrajectoriesNov 4
LongBanana: An Expert-Verified Benchmark for Long-Context Multi-Reference Image SynthesisNov 4

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 11m ago

Last hour Today

Pulse · computing.co.uk

computing.co.uk · Nov 5

Revolut turns to self-hosted AI to cut coding costs tenfold over Claude

Pulse · wazzup.ph

wazzup.ph · Nov 5

Tencent Cloud Expands Global Growth and Enterprise AI

Overseas customer base doubles amid triple-digit growth in demand for AI products and services. Tencent previews upcoming Hy4 model releases and expands its enterprise AI portfolio. Tencent has unveiled the enterprise capabilities of its AI office productivity agent, WorkBuddy, while offering a preview of its upcoming Hy4 and Hy4 Flash models at the 2026 Tencent […]

Pulse

Top picks this weekarXiv

arXiv

Nov 4

From Pixels to Actions - Towards Native Vision-Language-Action Policies

Vision-language-action models are typically built on a simple premise: perception is first compressed by visual encoders, and action reasoning begins only afterwards. Yet manipulation often hinges on precisely the visual signals most vulnerable to this abstraction, e.g., local geometry, boundaries, depth, and subtle state changes. This raises a fundamental question: should action policies inherit a fixed visual endpoint, or directly access evolving visual computation? We introduce NEO-VLA, a structurally encoder-free vision-language-action framework. Images are tokenized by lightweight patch convolutional layers and evolve jointly with language throughout the backbone, allowing a flow-matching action expert to draw from instruction-conditioned visual representations at different depths. This simple change reveals a consistent pattern. Native architectures outperform encoder-based counterparts under controlled comparisons; full fine-tuning reduces but does not erase the gap, and additional scale alone cannot recover it. More strikingly, layer-wise probing shows that the action-critical signals often peak well before the final visual representation. Across diverse manipulation…