51–60 of 138 posts
[1] https://jdsemrau.substack.com/p/nemotron-vs-qwen-game-theory...
This test pairs o1 with itself, which means the serializer is the deserializer. So while it's impressive that it can link 4 words, most humans could also easily link 4 with as much stretching! We just don't tend to because we can't guarantee that the other human will make the same connections we did.
Earlier quoted context omitted.
This is the take I thought I'd have, but in the last example, the guesser model reaches the correct conclusion using a different reasoning than the clue giver model. The clue giver justifies the link of Paper and Log as "written records", and between Paper and Line as "lines of text". But the guesser model connects Paper and Log because "paper is made from logs" (reaching the conclusion through a different meaning of…
> But the guesser model connects Paper and Log because "paper is made from logs" (reaching the conclusion through a different meaning of Log) No, it doesn't. It reaches the conclusion because of vector similarity (simplified explanation): these explanations are post-hoc .
A small weakness in this test is that one of the keys to strategic Codenames play is understanding your partner. You're not just trying to connect the words, you're trying to connect them in a way that will be obvious to your partner. As a computing analogy: you're trying to serialize a few cards in a way that will be deserializable by the other player. This test pairs o1 with itself, which means the serializer is th…
A small weakness in this test is that one of the keys to strategic Codenames play is understanding your partner. You're not just trying to connect the words, you're trying to connect them in a way that will be obvious to your partner. As a computing analogy: you're trying to serialize a few cards in a way that will be deserializable by the other player. This test pairs o1 with itself, which means the serializer is th…
lol I played this game with my family and they said my wife and I were cheating because I kept using inside jokes that made no sense to them but she would get immediately.
A small weakness in this test is that one of the keys to strategic Codenames play is understanding your partner. You're not just trying to connect the words, you're trying to connect them in a way that will be obvious to your partner. As a computing analogy: you're trying to serialize a few cards in a way that will be deserializable by the other player. This test pairs o1 with itself, which means the serializer is th…
This task could probably be solved nearly just as well with old school word 2 vec embeddings
I don’t find this “super good”. It’s mostly giving 2 clues which is the most basic level of competence. The paper 4 clue is reasonable but a bit lucky (eg Jack is also a good guess). I also don’t see it actually using tactics properly, which I would consider part of being “super good”. The game isn’t just about picking a good clue each round! Now obviously it’s still pretty decent at finding the clues. Probably bette…
Can you elaborate on some of the more advanced tactics? When I play, it's mostly about getting a good 2 clue each time. Then if you can opportunistically get a 3 or 4, that's awesome. Some tactics come in for choosing the right pairs of 2's so you don't end up mismatched, or leaving clues that might be ambiguous with your opponent's... But that's mostly it. It'll be fun for multiplayer! Just like how in other online…
If you want to get nasty, you learn to abuse the fact that the tile layouts follow rules and that you can rule out certain tiles without considering the words.
Earlier quoted context omitted.
> But the guesser model connects Paper and Log because "paper is made from logs" (reaching the conclusion through a different meaning of Log) No, it doesn't. It reaches the conclusion because of vector similarity (simplified explanation): these explanations are post-hoc .
Sorry, I’m uninformed. Do you mean thaw the explanation could be completely unrelated to the actual “reason”
If an LLM is told to do reasoning and then state the answer, it follows that the answer is basically guaranteed to be derived from the previously generated reasoning.
Earlier quoted context omitted.
Can you elaborate on some of the more advanced tactics? When I play, it's mostly about getting a good 2 clue each time. Then if you can opportunistically get a 3 or 4, that's awesome. Some tactics come in for choosing the right pairs of 2's so you don't end up mismatched, or leaving clues that might be ambiguous with your opponent's... But that's mostly it. It'll be fun for multiplayer! Just like how in other online…
Other advanced tactics involve giving a broad clue that matches 3-4 of your own and just one other (either your opponents or a civilian). Your team can pick up all the matches across several turns and the one off doesn't hurt as much as the plus four helps
Example: The clue is "places 4" and the guessers choose 1 correctly and then 1 wrong answer, but they had achieved consensus about 2 others (and are confused about only the remaining 1). So the turns ends but they inform the clue giver to inflate by 2 next turn. That clue giver (after the other team goes) will then say the clue is "people 5" and the guessers will know that they shall select 2 places and 3 people.
This can cascade beyond just a pair of turns.