https://iclr.cc/virtual/2025/oral/31888
Our poster was popular:
poster: https://iclr.cc/media/PosterPDFs/ICLR%202025/30358.png?t=174...
oral presentation (watch me roast yoshua bengio on this topic and then have him be the first questioner, 2nd speaker starting around 19:30 min mark. My slides for the presentation are there too and really funny.): https://iclr.cc/virtual/2025/session/31936
paper: https://arxiv.org/abs/2407.01082
As one of the min_p authors, I can confirm that Top N sigma is currently the best general purpose sampler by far. Also, temperature can and should be scaled far higher than it is today. Temps of 100 are totally fine with techniques like min_p and top N sigma.
Also, the special case of top_k = 2 with ultra high temperature (one thing authors recommend against near the end) is very interesting in its own right. Doing it leads to spelling errors every ~10th word - but also seems to have a certain creativity to it that's quite interesting.
Does MBR (minimal bayes risk) sampling count?
Also there was this paper at ICLR which is relevant to this question: https://arxiv.org/abs/2410.03968
This paper basically claims that non-heuristic methods (like beam search) are harmful compared to the heuristic ones.
Certain samplers described here like repetition penalty or DRY are just like this - the model could repeat itself in a myriad of ways, the only way to prevent all of them is better training, not n-gram search or other classic NLP methods. This is basically trying to plug every hole with a finger. How many fingers do you have?
Hacking the autoregressive process has some some low-hanging fruits like Min-P that can make some improvement and certain nifty tricks possible, but if you're doing it to turn a bad model into a good one, you're doing it wrong.
Top n sigma has been around since mid 2024, min_p around since 2023 and we are still waiting for these innovations to be integrated outside of open source stuff (i.e. outside of HF/vllm). It's being done slowly on purpose by API providers because they don't want to deal with the risk of models being "too creative" (also high temp likely breaks their watermarking)
One other thing - making models aware of their own sampling settings is super easy if you just feed it back to the model every token or generation (say, using structured generation). Models can control their own sampling settings and thus "have access to its internal states" with just a tiny bit of extra programming (the model can write that code for you now lol)
But the bigger problem is that the concepts are expressed before they're decoded into the output distribution. You can steer them to a degree by hacking the autoregressive transport, but if the model itself learned that this concept corresponds to that particular concept, not a set of concepts (and RL tends to do exactly that), fixing it with sampling is usually hard to impossible, you'll just lose accuracy/make it dumber as you basically force out-of-distribution outputs.
I see this sentiment a lot, there's even people that swear by samplers like XTC (which sounds counter intuitive af) but it's always on "creative" tasks. On math tasks, with a clear correct/incorrect answer, none of the "creative" samplers come on top, not even min_p (except for crazy temperatures, and even there the overall accuracy is still lower than normal temps w/ normal sampling)...
The main problem is that "creativity" is such a subjective measure that it's hard to score properly.
You're right in general on this post, but I think you underestimate how many coomers/erp folks there are and how much they use LLMs. XTC was made for them to give some notion of slop removal. It's probably not quite as good at that task as the antislop sampler (from Sam Peach, EQ bench creator) - but I find XTC to be quite good at adding "spice" to outputs.
re: difficulty to measure "creativity" is especially true - especially around the difficulty of scoring it! We have some nitpickers of our own whispering into our ears about this. You don't happen to be at Stanford do you? IFYKYK...
Not that I think that goes against your point -- I think it's rather a problem with the bitter lesson.
The bitter lesson critique is that the human designed heuristics were not free, and harmed the notion of "letting the computer figure it out" by slowing down training. High temp sampling is very important for half-way decent synthetic data generation and thus enabling "letting the computer figure it out" for natural language. Better sampling is the only way to make high temperature generations coherent.
For instance, why not use whole words as tokens? Make a "robot" with a limited "robot dialect." Yes, no capacity for new words or rare words, but you could modify the training data and input data to translate those words into the existing vocabulary. Now you have a much smaller mapping that's literally robot-like and kind of gives the user an expectation of what kind of answers the robot can answer well, like C-3PO.
Word-only tokenizers what people did in the RNN/LSTM days. There's no functional improvement over tokenization schemes like BPE or even WordPiece/SentencePiece, and it results in worse quality since you can't use meaningful semantic hints such as punctuation.
The sampling, in this framework, should not happen near the output level ("what will the next spoke word be").
So, articles like this submission - while interesting from many points of view - make the elephant in the room more evident.
> You cannot define an idea as a training loss objective
What tells you so? If you see a technical limit, note e.g. that sentences and paragraphs can have their own position in an embedding space.
These one-hot encoded vectors are then fed through a linear layer that encodes the token vector down into the hidden state size of the model. e.g. you might have a token vocabulary of 10-100k but a hidden state size of 0.5-2k. Everything else in the model works in hidden state space[1], which has all sorts of higher-level concepts in it.
Now, if we were to remove tokenization, then the encoder needs to do more work in order to get to the same hidden state space we're used to. It might be able to find a more efficient encoding from unpaired bytes to the hidden space, but that seems unlikely, given that the tokenization most models use is already based on the statistical properties of the training set. If we don't automatically pair "anti" or "ism" into a single token before handing it off to the model, then the attention heads on the lower layers in the model have to do the same work.
Given that we used to train models on character sequences, and then moved to tokenization because it was more efficient, I suspect the trade-off is never going to be worth it.
[0] That is, you can't just give it a list of token IDs, because there's no mathematical meaning to token 123.25, nor any meaning to increasing or decreasing token IDs.
[1] This improves performance but makes interperability harder. Most notably, the hidden space's basis vectors are not directly correlated to words or concepts, instead all the concepts exist on a sort of N-dimensional ring.
What I mean is an extra neural network that comes before the input of the llm, which converts characters (or simple 1-hot vectors which correspond to charactes) into tokens (or whatever it is you would call the internal representation of the network). The advantage would be a more unified way of representing the llm, and I guess one downside would be that you'd get a lot of replication in the NN, but perhaps these parts can be merged (have shared weights).
If you’re going to make a criticism like that, you might want to check a dictionary first:
> modern, adj. designed and made using the most recent ideas and methods
— https://dictionary.cambridge.org/us/dictionary/english/moder...
That’s exactly what this article is describing. There’s been a lot of development in this space over the last seven years or so, and e.g. GPT 1, 2, and 3 are certainly very outdated at this point, i.e. not modern in the above sense.
Anyone know who wrote it? It's not credited and it's pubished on a free Markdown pastebin.
The section on DRY - "repetition penalties" - was interesting to me. I often want LLMs to deliberately output exact copies of their input. When summarizing a long conversation for example I tend to ask for exact quotes that are most illustrative of the points being made. These are easy to fact check later by searching for them in the source material.
The DRY penalty seems to me that it would run counter to my goal there.
Most techniques are not made available by API providers because they enable alignment breaking. It's the only explanation for why we are still stuck with only top_p, top_k, and temp of 0-2.
If you want proper sampler settings to be available, your options are oobabooga, sillytavern (dependent on your backend, so vllm backend for example doesn't have top-n sigma yet), or directly running huggingface code. There might be some marginal options here too but in general, sampling innovation is firmly in the hands of open source coomers right now and not in the hands of academics.
Does min_p help with non-creative-writing tasks as well, such as maths or coding? Is there any way to improve/tune the performance for these with sampling?
Do you have a recommendation/guide on tuning the sampling parameters with what little the frontier model providers expose?
All I have ever seen is the same old very old advice of "t=0 for determinism, t=1 for creativity", with little treatment of top_k/top_p. But even the temp has become quite opaque nowadays. Is this really the softmax temperature that we're controlling, or some proxy for it? Claude 3.7 Sonnet doesn't allow me to use t != 1.0 for thinking mode, while Gemini 2.5 Pro, also a reasoner, will happily accept it. Does it apply only to the non-thinking tokens at the end?
If I get good results with t=0.0 (for e.g. coding), do I lose some "capability" by keeping it at 0.0?
Thanks in advance.
