MIRAGE: The Illusion of Visual Understanding · HackerPapers discussions

MIRAGE: The Illusion of Visual Understanding — mentioned by riahi

Mon, 29 Jun 2026 13:22:07 GMT

You can see it in just this PDF report. It's multiple things. It never shows the subscapularis in the way that people actually look the tendon. It hyper fixates on the axial when I find the sagittal much more useful for subscapularis. Figure 7. There's an arrow pointing "to the acromial undersurface". The arrow is not pointed to that location. Figure 5. "thin bursal fluid". This is within physiologic variation, but is calling bursitis. It keeps bringing up irrelevant normal things like the shape of the coracromipal arch, I assume because lots of websites have information about that as a…

Paper and discussion history. A mention is not an endorsement.

MIRAGE: The Illusion of Visual Understanding — mentioned by gpm

Sun, 03 May 2026 19:29:25 GMT

I'd be very very hesitant to trust studies like this. It's very easy to mess up these benchmarks. See for example this recent paper where AI managed to beat radiologists on interpreting x-rays... when the AI didn't even have access to the x-rays: https://arxiv.org/pdf/2603.21687 (on a pre existing "large scale visual question answering benchmark for generalist chest x-ray understanding" that wasn't intentionally messed up). And in interpreting x-ray's human radiologists actually do just look at the x-rays. In the context the article is discussing the human doctors don't just look at the notes…

Paper and discussion history. A mention is not an endorsement.

MIRAGE: The Illusion of Visual Understanding — HN submission

Mon, 30 Mar 2026 05:25:05 GMT

Stanford study reveals AI vision models invent images they never see

49 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

MIRAGE: The Illusion of Visual Understanding — mentioned by fernly

Sun, 29 Mar 2026 18:32:34 GMT

From the Conclusion section of the PDF[1], "Multimodal AI systems are increasingly deployed on the assumption that their benchmark performance reflects genuine visual understanding. Our results fundamentally challenge these assumptions. Across every model-benchmark pair tested, the accuracy that frontier models achieved without any access to images exceeded the additional accuracy they gained when images were provided. Moreover, a text-only 3-billion-parameter model, trained solely on question-answer pairs stripped of images, outperformed all frontier multimodal systems and human radiologists…

Paper and discussion history. A mention is not an endorsement.

MIRAGE: The Illusion of Visual Understanding — HN submission

Sun, 29 Mar 2026 16:51:13 GMT

Mirage Reasoning: The Illusion of Visual Understanding

12 points; 4 discussion replies on HN. Paper and source history. A submission is not an endorsement.