Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters · HackerPapers discussions

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — HN submission

Sun, 16 Feb 2025 05:48:45 GMT

Scaling Test-Time Compute Can Be More Effective Than Scaling Parameters (2024)

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — mentioned by dimitry12

Fri, 20 Dec 2024 21:32:38 GMT

Verifier is trained with soft values of reward-to-go for each solution-prefix, obtained from monte-carlo rollouts of step-by-step solutions sampled from the "base" model. In other words: 1) sample step-by-step solutions from "base" model; 2) do it at non-zero temperature so that you can get multiple continuation from each solution-prefix; 3) use MATH-labels to decide if full solution (leaf/terminal node in MC rolloout) has reward `1` or `0`; 4) roll up these rewards to calculate reward-to-go for each intermediate step. Yes, verifier trained in this manner can be used to score…

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

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — HN submission

Fri, 13 Sep 2024 20:46:49 GMT

Scaling LLM Test-Time can be More Effective than Scaling Parameters

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — mentioned by wesleyyue

Thu, 12 Sep 2024 20:13:45 GMT

Just added o1 to https://double.bot if anyone would like to try it for coding. --- Some thoughts: * The performance is really good. I have a private set of questions I note down whenever gpt-4o/sonnet fails. o1 solved everything so far. * It really is quite slow * It's interesting that the chain of thought is hidden. This is I think the first time where OpenAI can improve their models without it being immediately distilled by open models. It'll be interesting to see how quickly the oss field can catch up technique-wise as there's already been a lot of inference time compute papers recently…

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

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — mentioned by sharemywin

Thu, 29 Aug 2024 12:59:36 GMT

https://arxiv.org/abs/2408.03314v1 Enabling LLMs to improve their outputs by using more test-time computation is a critical step towards building generally self-improving agents that can operate on open-ended natural language. In this paper, we study the scaling of inference-time computation in LLMs, with a focus on answering the question: if an LLM is allowed to use a fixed but non-trivial amount of inference-time compute, how much can it improve its performance on a challenging prompt? Answering this question has implications not only on the achievable performance of LLMs, but also on the…

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

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — mentioned by sanxiyn

Tue, 27 Aug 2024 17:42:03 GMT

Fast inference can substitute for larger models in some circumstances. As you said, you can run multiple times. DeepMind had a detailed look, see https://arxiv.org/abs/2408.03314.

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

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters — HN submission

Wed, 07 Aug 2024 15:54:10 GMT

Scaling LLM Test-Time Compute More Effective Than Scaling Model Parameters

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.