New preprint: recurrent vision models group objects and predict human reaction times
tserre · x · 2026-10-13
A new preprint from Brown University, combining work from Pieter Roelfsema's lab on incremental grouping in visual cortex with the lab's recurrent vision models.
Key points:
- The model groups objects in natural scenes and predicts human reaction times.
- Early grouping is governed by local boundary signals.
- Later, top-down feedback about object identity allows the grouping signal to cross internal edges, such as changes in shading or texture.
A computational-model account of how feedback connections support visual grouping.
More from Research
- Harvard & NTT Research: cooperating AI agent populations hit an Allee-effect takeoff threshold — Hidenori8Tanaka · 2026-10-13
- RT-Super trains tumor segmentation from radiology reports, lifting Dice from 16 to 30 without masks — maier_ak · 2026-10-13
- Princeton's QUEEN: a 4B model that plays grandmaster-level chess and explains its moves in plain language — jiqizhixin · 2026-10-13
- Oxford shows neural nets can be encrypted inside model weights, surviving 500 SFT steps — zainhas · 2026-10-13
- XPENG's 4B GroundingPI model beats GPT-6 Astra across 34 grounding benchmarks, boosts robot manipulation — XPENG-AI · 2026-10-13
- XPENG's GroundAnything pairs diffusion parallel decoding with precise grounding, 4.51x faster than AR — XPENG-AI · 2026-10-13