Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! · HackerPapers discussions

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by a2ff6eeb0

Mon, 07 Sep 2026 17:38:39 GMT

Still not meaningful -- https://arxiv.org/pdf/2504.09762; even for local models, the reasoning traces are often filtered and summarized to sound sensible to humans. And even if not, they don't necessarily represent what the model is thinking.

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by porridgeraisin

Sat, 05 Sep 2026 18:13:01 GMT

Literally many previous instance used one. Right from alphaevolve onwards. I'm not some "LLM is just a next token predictor guy" (GP seems to have a thing against LLMs), but to use LLMs properly you genuinely do need a grounded verifier and a planner. Coding harnesses for example are exactly that. For some plans, you can AR generate the search tree and that's what subagents being planned around by high level (LLM)agents and such are. Coding agents even with subagents are imperfect even on verifiable tasks only because of that. If you can put a human to simply guide it, it becomes a full…

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by stymaar

Fri, 04 Sep 2026 00:28:30 GMT

Also 2504.09762: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! https://arxiv.org/abs/2504.09762

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by ionioagnio

Thu, 03 Sep 2026 19:30:42 GMT

I thought it was pretty well understood amongst experts that trying to interpret so-called "chain of thought" is anthopomorphization and that the actual "thought process" of models is already opaque. https://arxiv.org/html/2504.09762v2

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by stymaar

Wed, 19 Aug 2026 18:46:39 GMT

Related: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces![1] > Our findings consistently challenge the prevailing narrative that intermediate tokens constitute a semantically meaningful reasoning process. First, we observe a pronounced lack of correlation between solution correctness and trace validity—models frequently produce invalid reasoning traces even when they arrive at correct solutions. Second, and more strikingly, models trained on corrupted or semantically irrelevant traces achieve performance comparable to, and often exceeding, that of models trained on…

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — HN submission

Wed, 19 Aug 2026 11:35:26 GMT

Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)

316 points; 281 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — HN submission

Fri, 26 Jun 2026 06:46:56 GMT

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Trac

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking 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.

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by YeGoblynQueenne

Thu, 13 Nov 2025 13:21:13 GMT

The discussion has become contentious and that's very unfortunate because there's clearly some confusion about Prolog and that's always a great opportunity to learn. You say: >> Yes, though, it traverses the code tree by depth first walk. Here's what I suggest: try to think what, exactly, is the data structure searched by Depth First Search during Prolog's execution. You'll find that this structure is what we call and SLD-Tree. That's a tree where the root is a Horn goal that begins the proof (i.e. the thing we want to dis-prove, since we're doing a proof by refutation); every other node is a…

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — mentioned by drakythe

Thu, 09 Oct 2025 20:36:09 GMT

I'm almost positive reasoning is not an emergent behavior considering the reasoning models have specific architecture. As a source: https://arxiv.org/html/2504.09762v1

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — HN submission

Wed, 04 Jun 2025 12:19:20 GMT

Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — HN submission

Tue, 06 May 2025 14:29:31 GMT

(How) Do reasoning models reason?

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

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces! — HN submission

Mon, 21 Apr 2025 13:27:21 GMT

(How) Do reasoning models reason?

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