JOjonnycomputer·May 31, 2023
In Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?FranxYao
chain-of-thought-hub
Jupyter Notebook2023–2023
FranxYao / chain-of-thought-hub
chain-of-thought-hub
Benchmarking large language models' complex reasoning ability with chain-of-thought prompting
3Distinct people
3Mentions
0Explicit recommenders
2.8KGitHub stars
What to know before exploring
What it is
Benchmarking large language models' complex reasoning ability with chain-of-thought prompting
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 5:04:57 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by hour3 dated mentions
Observed mentions: 2023-05-30 21:52:56 UTC to 2023-05-31 11:49:20 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| May 30, 21:00 | 2023-05-30 21:00:00 UTC | 2023-05-30 22:00:00 UTC | 1 | 1 | 0 |
| May 30, 22:00 | 2023-05-30 22:00:00 UTC | 2023-05-30 23:00:00 UTC | 0 | 0 | 0 |
| May 30, 23:00 | 2023-05-30 23:00:00 UTC | 2023-05-31 00:00:00 UTC | 0 | 0 | 0 |
| May 31, 00:00 | 2023-05-31 00:00:00 UTC | 2023-05-31 01:00:00 UTC | 0 | 0 | 0 |
| May 31, 01:00 | 2023-05-31 01:00:00 UTC | 2023-05-31 02:00:00 UTC | 0 | 0 | 0 |
| May 31, 02:00 | 2023-05-31 02:00:00 UTC | 2023-05-31 03:00:00 UTC | 0 | 0 | 0 |
| May 31, 03:00 | 2023-05-31 03:00:00 UTC | 2023-05-31 04:00:00 UTC | 0 | 0 | 0 |
| May 31, 04:00 | 2023-05-31 04:00:00 UTC | 2023-05-31 05:00:00 UTC | 0 | 0 | 0 |
| May 31, 05:00 | 2023-05-31 05:00:00 UTC | 2023-05-31 06:00:00 UTC | 0 | 0 | 0 |
| May 31, 06:00 | 2023-05-31 06:00:00 UTC | 2023-05-31 07:00:00 UTC | 0 | 0 | 0 |
| May 31, 07:00 | 2023-05-31 07:00:00 UTC | 2023-05-31 08:00:00 UTC | 0 | 0 | 0 |
| May 31, 08:00 | 2023-05-31 08:00:00 UTC | 2023-05-31 09:00:00 UTC | 0 | 0 | 0 |
| May 31, 09:00 | 2023-05-31 09:00:00 UTC | 2023-05-31 10:00:00 UTC | 0 | 0 | 0 |
| May 31, 10:00 | 2023-05-31 10:00:00 UTC | 2023-05-31 11:00:00 UTC | 1 | 0 | 1 |
| May 31, 11:00 | 2023-05-31 11:00:00 UTC | 2023-05-31 12:00:00 UTC | 1 | 0 | 1 |
What Hacker News says
3 sourcesIt's not just you. Here's a bit of research you can cite:
> GPT-4 from its website and Bubeck et al Mar 2023. Note that the version that Bubeck uses is GPT-4 Early which is supposedly to be more powerful than GPT-4 Launch (OpenAI paid a lot of alignment tax to make GPT-4 safer).
https://github.com/FranxYao/chain-of-thought-hub
Anecdotally, there seemed to be a golden set of weeks in late April to early May that seemed like "peak GPT" (GPT-4), followed by heavy topic and knowledge mitigation since, then -- just this week -- adding back some "chain of thought" or "show your work" ("lets go step by step" style) for math. I say anecdotally because I could just be prompting it wrong.
In Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?Chain-of-Thought Hub: Measuring LLMs' Reasoning Performance
114 points · Open discussion