What do these badges mean?
- 🚀ShippingCode exists. Multiple GitHub repos already reference this paper — people are building on it.
- 📈ClimbingCitation velocity is rising. Researchers are starting to pick it up.
- 💤QuietPublished but no notable signal yet. Most papers live here — could become anything later.
- 🎭HypeHeavy social buzz but no shipping signal. The counter-signal — defer until Twitter/X data is wired up.
- 💤Quiet2605.18746·May 18, 2026·~13 mincs.CVcs.AIcs.CL
ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop
Yining Hong, Jiageng Liu, Han Yin, Manling Li, +4
⭐ 14 stars / 7 repos📚 0 citesELI5A benchmark that tests whether AI agents can actively explore and understand 3D spaces by moving around and manipulating objects, rather than just looking at static pictures. It measures whether agents can figure out what's hidden, how things work, and what to do next—like how a human child learns by poking at toys.
Problem solvedMost spatial reasoning benchmarks give AI agents perfect information upfront. Real robots and embodied agents need to decide where to look and what to touch to understand a space. This benchmark exposes that current AI systems are 'action blind'—they make poor decisions about how to explore, leading to incomplete observations and cascading failures.
- 💤Quiet2605.18743·May 18, 2026·~9 mincs.AI
Actionable World Representation
Kunqi Xu, Jitao Li, Jianglong Ye, Tianshu Tang, +3
⭐ 2 stars / 5 repos📚 0 citesELI5A system that learns to understand how real objects change and can be manipulated by watching video or 3D scans, creating a digital twin of each object that knows its possible states and how actions affect it.
Problem solvedCurrent world models either generate videos or reconstruct scenes, but don't explicitly track how objects change state when acted upon. This makes it hard to build AI that understands the physical consequences of actions on specific objects.