Knowledge Distillation of Black-Box Large Language Models (2024)

by babelfish • 24 days ago

123 18 24 days ago

duendefm• 24 days ago

The Chinese are really going strong on destroying the American AI economy bubble. Honestly, despite the fact that I'm totally pro USA and anti China, I think we should help them crashing the American AI bubble. They are controlling everything and we can't even buy a new computer nowadays while getting no benefit from this. I wish some influential programmers stimulated coders everywhere to skip Claude and Chatgpt subscriptions for Chinese ones, at scale. If we programmers united we could help this bubble burst, I'm sure.

nozzlegear• 24 days ago

skip Claude and Chatgpt subscriptions for Chinese ones, at scale. If we programmers united we could help this bubble burst, I'm sure.

I'm doing my part!

anax32• 24 days ago

The US "product machine" is so strong. They really know how to do frictionless signup and vendor lock-in on the corporate side.

laichzeit0• 24 days ago

The US government will do the job of destroying the American AI economy through their export controls.

Roark66• 23 days ago

If we programmers united we had a clouded code alternative that didn't suck :-)

But we're not far.

My requirements: - a terminal app without advanced tui, not written like "a browser running in a terminal" or a game. There is no need to overcomplicated. - ability to manage prompts per model, compress context using alternate models, and minimise token costs better - like the YouTube's Sentdex's Minion mini harness (in fact I'm building on top of his as we speak). - support for agent work fanout - support for MCP, but switchable off/on depending if needed (I use a single MCP aggregator anyway so mcp tool use doesn't eat my context) - support for lsp/tree-sitter, again switchable when needed. - support for OpenAI api and written easily enough so other ones like deepinfra are easy to add.

Nice to have: - have some sort "prompt library" that would store tweaked versions of prompts for different models so it adjusted the harness as needed depending on which model we call.

That's it.

addedGone• 23 days ago

"anti China", why so? have you lived there?

linolevan• 24 days ago

Can we note that this is a 2024 paper in the title?

Alifatisk• 24 days ago

Why is this published again? Is this a reference to recent events?

babelfish• 24 days ago

I just saw some post about it on Threads and found it interesting so decided to share!

tough2• 24 days ago

My best guess is this is a reference to the recent accusations from Anthropic of chinese labs ¨distilling¨ on their models

dmezzetti• 24 days ago

Well-Read Students Learn Better: On the Importance of Pre-training Compact Models

Related paper that's a good read: https://arxiv.org/abs/1908.08962

phantompeace• 24 days ago

Considering the very small difference between just SFT on the student model as compared to SFT + DPO on a proxy, doesn't it make sense to concentrate on ensuring the SFT dataset is perfect rather than sorry about DPO etc? And just train directly on the student model?

StreamCtx• 24 days ago

“Relevant to anyone building failure-attribution systems for agent pipelines — black-box distillation techniques here could feed into causal attribution models without needing white-box access to the underlying model.”

adrian_b2• 23 days ago

That is easy when you can control the teacher model yourself and you want to transfer its capabilities to a smaller model.

If the teacher model is run by an external entity, e.g. Anthropic or OpenAI, then the number of queries to the blackbox model that is required is so great that it should be easy for the owner of the teacher LLM to detect and stop any such attempts.

potus_kushner• 23 days ago

probably more interesting (from 01/2026) https://arxiv.org/pdf/2511.10643 "Black-Box On-Policy Distillation of Large Language Models". they got a qwen 2.5 14B model trained to GPT5 level using the described technique "Generative Adversarial Distillation (GAD)".

Knowledge Distillation of Black-Box Large Language Models (2024)

Related Stories