"Crucially, it tells the agent not to rely on its internal training data (which might be hallucinated or refer to a different version of the game) but to ground its knowledge in what it observes. "
Does this even have any effect?
"Crucially, it tells the agent not to rely on its internal training data (which might be hallucinated or refer to a different version of the game) but to ground its knowledge in what it observes. "
Does this even have any effect?
Yes, at least to some extent. The author mentions that the base model knows the answer to the switch puzzle but does not execute it properly here.
"It is worth noting that the instruction to "ignore internal knowledge" played a role here. In cases like the shutters puzzle, the model did seem to suppress its training data. I verified this by chatting with the model separately on AI Studio; when asked directly multiple times, it gave the correct solution significantly more often than not. This suggests that the system prompt can indeed mask pre-trained knowledge to facilitate genuine discovery."
My issue with this is that the LLM could just be roleplaying that it doesn't know.
Of course it is. It's not capable of actually forgetting or suppressing its training data. It's just double checking rather than assuming because of the prompt. Roleplaying is exactly what it's doing. At any point, it may stop doing that and spit out an answer solely based on training data.
It's a big part of why search overview summaries are so awful. Many times the answers are not grounded in the material.
It may actually have the opposite effect - the instruction to not use prior knowledge may have been what caused Gemini 3 to assume incorrect details about how certain puzzles worked and get itself stuck for hours. It knew the right answer (from some game walkthrough in its training data), but intentionally went in a different direction in order to pretend that it didn't know. So, paradoxically, the results of the test end up worse than if the model truly didn't know.
Doesn't know what? This isn't about the model forgetting the training data, of course it can't do that any more than I can say "press the red button. Actually, forget that, press whatever you want" and have you actually forget what I said.
Instead, what can happen is that, like a human, the model (hopefully) disregards the instruction, making it carry (close to) zero weight.
To test would just need to edit the rom and switch around the solution. Not sure how complicated that is, likely depends on the rom system.
I don't know why people still get wrapped around the axle of "training data".
Basically every benchmark worth it's salt uses bespoke problems purposely tuned to force the models to reason and generalize. It's the whole point of ARC-AGI tests.
Unsurprisingly Gemini 3 pro performs way better on ARC-AGI than 2.5 pro, and unsurprisingly it did much better in pokemon.
The benchmarks, by design, indicate you can mix up the switch puzzle pattern and it will still solve it.
I'm wondering about this too. Would be nice to see an ablation here, or at least see some analysis on the reasoning traces.
It definitely doesn't wipe its internal knowledge of Crystal clean (that's not how LLMs work). My guess is that it slightly encourages the model to explore more and second-guess it's likely very-strong Crystal game knowledge but that's about it.
The model probably recognizes the need for a grassroots effort to solve the problem, to "show it's work".
It's hard to say for sure because Gemini 3 was only tested with this prompt. But for Gemini 2.5, which is who the prompt was originally written for, yes this does cut down on bad assumptions (a specific example: the puzzle with Farfetch'd in Ilex Forest is completely different in the DS remake of the game, and models love to hallucinate elements from the remake's puzzle if you don't emphasize the need to distinguish hypothesis from things it actually observes).
It will definitely have some effect. Why won't it? Even adding noise into prompts (like saying you will be rewarded $1000 for each correct answer) has some effect.
Whether the 'effect' something implied by the prompt, or even something we can understand, is a totally different question.
I very much doubt it
Do we have examples of this in promps in other contexts?
I would imagine that prompting anything like this will have an excessively ironic effect like convincing it to suppress patterns which it would consider to be pre-knowledge.
If you looked inside they would be spinning on something like "oh I know this is the tile to walk on, but I have to only rely on what I observe! I will do another task instead to satisfy my conditions and not reveal that I have pre-knowledge.
LLMs are literal douche genies. The less you say, generally, the better
If they trained the model to respond to that, then it can respond to that, otherwise it can't necessarily.
I think you got a point here. These companies are injecting a lot of datasets every day into it.
What I meant is more like, if you write tests for something you know it works, and if you don't write tests you don't know that.
It might get things wrong on purpose, but deep down it knows what it's doing