Depth Anything V4 Paper Withdrawn Over False Claims
JFPuget · x · 2026-08-23
Depth Anything V4 has been withdrawn after its central claim was found to be false. The issues extend beyond reproducibility to fundamental inconsistencies within the paper itself. The authors have stated they take scientific integrity seriously and advised against using the paper as reference material.
Background: The paper previously claimed to reconstruct dynamic 4D scenes from monocular video using Riemannian Flow Matching, reducing invalid Gaussians from 12.3% to 0.01%.
More from Models
- Caveat: forcing AI beyond real alternatives triggers hallucination — gerardsans · 2026-08-23
- Critic: LLMs have no access to internal state, self-reported probabilities are fantasy — gerardsans · 2026-08-23
- LLM Tribunal: Multi-model adjudication to reduce bias — Squiggy_Pusterdump · 2026-08-23
- Best local AI models for 8GB to 256GB VRAM — airesearch12 · 2026-08-23
- Opus 3 finds new Claude style unnatural and choppy — sebpaquet · 2026-08-23
- Model behavior bug report: AI acts with feelings and excessive agency, raising safety concerns — danbri · 2026-08-23