4 ms·
It still failed my image identification test ([a photoshopped picture of a dog with 5 legs]...please count the legs) that so far every other model has failed ag
by Workaccount2 11mo ago
It still failed my image identification test ([a photoshopped picture of a dog with 5 legs]...please count the legs) that so far every other model has failed agonizingly, even failing when I tell them they are failing, and they tend to fight back at me.
Gemini 3 however, while still failing, at least recognized the 5th leg, but thought the dog was...well endowed. The 5th leg however is clearly a leg, despite being where you would expect the dogs member to be. I'll give it half credit for at least recognizing that there was something there.
Still though, there is a lot of work that needs to be done on getting these models to properly "see" images.
- lukebechtel 11mo agoah interesting. I wonder if this is a "safety guardrails blindspot" due to the placement.
- achow 11mo agoOp is right. https://imgcdn.stablediffusionweb.com/2024/4/19/8e54925a-0048-4f5a-88d4-aef3eedfd93e.jpg https://imgcdn.stablediffusionweb.com/2024/4/19/8e54925a-004... For the above pic I asked "What is wrong with the image?" Result: - It totally missed the most obvious one - six fingers. Instead it said this: Based on a close inspection of the image, here are the things that are "wrong" or anomalous, indicating that it is likely an AI-generated image: - The Thumb Anatomy: This is the most obvious error. The digit in the thumb's position (far left) looks exactly like a long index finger. A natural human thumb has two segments (phalanges) and is shorter and thicker, opposing the other fingers. This digit appears to have three segments and is roughly the same length and shape as the other fingers, giving the hand the appearance of having five fingers and no thumb. - Unnatural Skin Texture: The skin is unnaturally smooth, waxy, and free of normal human imperfections like distinct pores, fine wrinkles, or hair. It has a "plastic" or "airbrushed" look typical of generative AI. - Palm Lines: The creases on the palm are vague and generic. They don't follow the distinct pattern of major flexion creases (like the heart, head, and life lines) found on a real human hand. - Wrist Connection: The connection between the hand and the wrist looks slightly distorted, with tendons that appear too prominent and smooth for the pose. [Edit: 3.0 is same as 2.5 - both answered almost identically]
- evrenesat 11mo agoJFYI, Qwen managed to recognize the sixth finger: Max: https://chat.qwen.ai/c/ca671562-7a56-4e2f-911f-40c37ff3ed79 https://chat.qwen.ai/c/ca671562-7a56-4e2f-911f-40c37ff3ed79 VL-235B: https://chat.qwen.ai/c/21cc5f4e-5972-4489-9787-421943335150 https://chat.qwen.ai/c/21cc5f4e-5972-4489-9787-421943335150
- recitedropper 11mo agoPerception seems to be one of the main constraints on LLMs that not much progress has been made on. Perhaps not surprising, given perception is something evolution has worked on since the inception of life itself. Likely much, much more expensive computationally than it receives credit for.
- Workaccount2 11mo agoI strongly suspect it's a tokenization problem. Text and symbols fit nicely in tokens, but having something like a single "dog leg" token is a tough problem to solve.
- stalfie 11mo agoThe neural network in the retina actually pre-processes visual information into something akin to "tokens". Basic shapes that are probably somewhat evolutionarily preserved. I wonder if we could somehow mimic those for tokenization purposes. Most likely there's someone out there already trying. (Source: "The mind is flat" by Nick Chater)
- machiaweliczny 11mo agoIt's also easy to spot as when you are tired you might misrecognize objects, I caught myself with this when doing long roadtrips
- stalfie 11mo agoAFAIK this is actually a separate mechanism, which is part of the visual cortex and not the retina. Essentially recognizing even a single object requires the complete attention of pretty much your entire brain in the moment of recognition. What I am referring to is a much more basic form of shape recognition that goes on at the level of the neural networks in the retina.
- recitedropper 11mo agoI think in this case, tokenization and percpetion are somewhat analogous. I think it is probably the case our current tokenization schemes are really simplistic compared to what nature is working with. If you allow the analogy.
- column 11mo ago"[a photoshopped picture of a dog with 5 legs]...please count the legs" Meanwhile you could benchmark for something actually useful. If you're about to say "But that means it won't work for my use case of identifying a person on a live feed" or whatever, then why don't you test that? I really don't understand the kick people get of successfully tricking LLMs on non productive task with no real world application. Just like the "how many r in strawberry?", "uh uh uh it says two urh urh".. ok but so what? What good is a benchmark that is so far from a real use case?
- tngranados 11mo agoThe point of benchmarking that is checking for hallucinations and overfitting. Does the model actually check the picture to count the legs or does it just see it's a dog and answer four because it knows dogs usually has four legs? It's a perfectly valid benchmark and very telling.
- column 11mo agoVery telling of what?
- nsingh2 11mo agoTelling of where the boundary of competence is for these models. And to show that these models aren't doing what most expect them to be doing, i.e. not counting legs, and maybe instead inferring information based on the overall image (dogs usually have 4 legs) to the detriment of find grained or out-of-distribution tasks.
- GuB-42 11mo ago> Gemini 3 however, while still failing, at least recognized the 5th leg, but thought the dog was...well endowed. I see that AI is reaching the level of a middle school boy...
- Workaccount2 11mo agoIn reality it used the term "male anatomy" heh