Prompt Repetition Improves Non-Reasoning LLMs · HackerPapers discussions

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by jeff_ciesielski

Wed, 30 Sep 2026 17:37:43 GMT

Howdy folks. I was inspired by the really neat "Jev in 25 lines of Python" post a few days ago (and System One models in general), and after playing with the idea I tried whipping up something a little more complex, and Lichen is what popped out: https://github.com/Mushroom-Systems/lichen My question after reading that post (and I know I wasn't the only one; antirez came right out and said it in the comments) was whether Google Research's prompt repetition paper for non-reasoning models (https://arxiv.org/abs/2512.14982) would improve the outcomes. Turns out it did a fair bit. After some…

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by stellalo

Wed, 23 Sep 2026 18:05:11 GMT

Prompt Repetition Improves Non-Reasoning LLMs: https://arxiv.org/abs/2512.14982

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by hasteg

Mon, 06 Jul 2026 23:06:47 GMT

Really interesting paper, thanks for the share. The point their making in that paper reminds me of this paper some people shared around work earlier this year, https://arxiv.org/pdf/2512.14982 (Prompt Repetition Improves Non-Reasoning LLMs)... I wonder how OPs question would fare (or the questions presented in the paper you posted) given double repetition.

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by 0xbadcafebee

Tue, 31 Mar 2026 15:33:35 GMT

There's an ancient paper that shows repetition improves non-reasoning weights: https://arxiv.org/html/2512.14982v1

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Fri, 27 Mar 2026 17:49:50 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by skyde

Tue, 24 Mar 2026 15:37:18 GMT

Actually not surprised. I guess this is for the same reason “say it twice” [1] is working. Because LLm are trained as causal language model, past token cannot attend to future token. One copy of the layer set solve this. [1]https://arxiv.org/html/2512.14982v1

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by ar0b

Thu, 12 Mar 2026 19:16:41 GMT

"Prompt Repetition Improves Non-Reasoning LLMs " - https://arxiv.org/pdf/2512.14982 What instance of ChatGPT are you doing that with? (Reasoning?)

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Sat, 28 Feb 2026 15:00:47 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Thu, 26 Feb 2026 20:10:28 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — mentioned by sReinwald

Tue, 24 Feb 2026 08:39:38 GMT

Interestingly, when I apply the "simply repeat the prompt" technique [1], Sonnet 4.6 on the website got it right every time, both with and without extended thinking. Not repeating the prompt got a mix of walk and drive answers. I love how prompt engineering is basically techno-alchemy 1: https://arxiv.org/pdf/2512.14982

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Fri, 20 Feb 2026 20:19:41 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Fri, 20 Feb 2026 03:26:24 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Thu, 19 Feb 2026 04:09:43 GMT

Prompt Repetition Improves Non-Reasoning LLMs [pdf]

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Wed, 18 Feb 2026 07:24:46 GMT

Prompt Repetition Improves Non Reasoning LLM

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Tue, 20 Jan 2026 18:27:08 GMT

Repeating your prompt twice before sending it to an LLM improves accuracy

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Mon, 19 Jan 2026 17:28:16 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Sat, 17 Jan 2026 19:58:51 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Wed, 14 Jan 2026 21:24:53 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Wed, 14 Jan 2026 08:58:14 GMT

Prompt Repetition Improves Non-Reasoning LLMs

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

Prompt Repetition Improves Non-Reasoning LLMs — HN submission

Sat, 27 Dec 2025 16:28:10 GMT

Prompt Repetition Improves Non-Reasoning LLMs

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.