How to Steal Reasoning Without Reasoning Traces · HackerPapers discussions

How to Steal Reasoning Without Reasoning Traces — mentioned by dannyw

Tue, 18 Aug 2026 02:56:15 GMT

You can train a LLM to inverse summarised thinking into thinking text. It’s not perfect, but it gets you maybe 80% of the quality with proper techniques. Paper: https://arxiv.org/abs/2603.07267 FWIW, there’s not that much value protected here anyway IMHO, and even raw thinking text can lie (as shown by Anthropic’s amazing research), so for legitimate interpretability research it’s limited. Scaling frontier performance hasn’t been SFT-bounded for a while now; it’s now basically how much you can scale RL rollouts.

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

How to Steal Reasoning Without Reasoning Traces — mentioned by nervai

Tue, 11 Aug 2026 15:09:14 GMT

Really cool work, you get the actual traces. Looks like the vendors can all reliably fix this one though. A harder to defend against approach here where they work backwards from the results and ask the model to generate a plausible trace: How to Steal Reasoning Without Reasoning Traces https://arxiv.org/pdf/2603.07267

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

How to Steal Reasoning Without Reasoning Traces — mentioned by nyrikki

Wed, 17 Jun 2026 18:59:14 GMT

It is quite likely that the intermediate tokens don’t have ‘semantic import’[0] There are methods like Habitual Reasoning Distillation or Inverted Reasoning Traces [1] that can help. While there are reasons to hide the intermediate tokens from a IP protection stand point, there is also a need to hide more effective and efficient generating that doesn’t fit the R1 claims of an aha moment that has been debunked, but is a consumer expectation. While hidden intermediate tokens do increase the difficulty, it is not a from barrier in itself, especially as they are billed, given information about…

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