Scalable Extraction of Training Data from (Production) Language Models · HackerPapers discussions

Scalable Extraction of Training Data from (Production) Language Models — mentioned by uhfraid

Fri, 28 Nov 2025 03:18:02 GMT

> Like if there is no way to trace it back to the original material, does it make sense to regulate it? Training data extraction has seen some success, tracing should be possible for at least some of it https://arxiv.org/abs/2311.17035

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

Scalable Extraction of Training Data from (Production) Language Models — mentioned by tptacek

Sun, 05 Oct 2025 22:10:13 GMT

Yes. https://arxiv.org/abs/2403.06634 https://arxiv.org/abs/2311.17035 (I just have these ones off the top of my head because I'm a Nicholas Carlini fan and we interviewed him about these attacks.)

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

Scalable Extraction of Training Data from (Production) Language Models — mentioned by kmeisthax

Sun, 04 Feb 2024 20:05:32 GMT

I had the exact same idea after seeing Google point out that you can[0] get ChatGPT to regurgitate verbatim training data by asking it to repeat the same word over and over again[1]. I'm glad to see someone else actually bring it to fruition. This, of course, brings two additional questions: 1. Is this "AI, hold the AI" approach more energy-efficient than having gradient descent backpropagation compress a bunch of training data into a model that can then be run on specialized AI coprocessors? 2. Will this result wind up being evidence in the ongoing lawsuits against OpenAI and Stability AI?…

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

Scalable Extraction of Training Data from (Production) Language Models — mentioned by snakeyjake

Sat, 20 Jan 2024 19:18:34 GMT

Commercial musicians do not? You sure about that? Entire legal firm empires have been built on the licensing, negotiations, and fees that make up the industry. I'm ain't talking about some dude on YouTube or Soundcloud. Few people care about some rando on Soundcloud. Those moles aren't big enough to whack. Vanilla Ice and MC Hammer were. OpenAI is as well. There's even a company that specializes in sample clearance: https://sampleclearance.com More info: https://www.soundonsound.com/sound-advice/sample-clearance Also: >not turned into a part of a pattern used to generate new sounds like what…

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

Scalable Extraction of Training Data from (Production) Language Models — mentioned by WhitneyLand

Tue, 05 Dec 2023 18:31:42 GMT

Sorry for any confusion, my comment was meant to refer to the linked paper: https://arxiv.org/abs/2311.17035 I think it’s valid work, but the original tweet seems to call a prompt based paper into question. At least enough to clarify where he would stand on an example like this.

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

Scalable Extraction of Training Data from (Production) Language Models — HN submission

Sat, 02 Dec 2023 07:32:54 GMT

Scalable extraction of training data from (production) language models

105 points; 14 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Scalable Extraction of Training Data from (Production) Language Models — HN submission

Fri, 01 Dec 2023 14:53:17 GMT

Scalable Extraction of Training Data from (Production) Language Models

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

Scalable Extraction of Training Data from (Production) Language Models — HN submission

Wed, 29 Nov 2023 14:40:50 GMT

Scalable Extraction of Training Data from (Production) Language Models

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

Scalable Extraction of Training Data from (Production) Language Models — HN submission

Wed, 29 Nov 2023 04:30:57 GMT

Extracting Training Data from LLMs

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