From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning · HackerPapers discussions

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning — mentioned by Xmd5a

Sat, 22 Nov 2025 22:37:09 GMT

Related: From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning – https://arxiv.org/pdf/2505.17117 (Lecun/Jurafsky) > Large Language Models (LLMs) demonstrate striking linguistic capabilities that suggest semantic understanding (Singh et al., 2024; Li et al., 2024). Yet, a critical question remains unanswered: Do 1arXiv:2505.17117v5 [cs.CL] 25 Sep 2025LLMs navigate the compression-meaning trade-off similarly to humans, or do they employ fundamentally different representational strategies? This question matters because true understanding, which goes beyond surface-level…

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

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning — mentioned by mbando

Tue, 21 Oct 2025 01:43:28 GMT

This also touches on the contrast between how human beings and LLM's trade compression for nuance. Human beings have enormous resources devoted to long-tailed distribution of information, for example in lexical items. Word distributions follow Zipf's Law, so like in the million word FROWN corpus, roughly half the words only occur one time. Like when's the last time you use the word chrysanthemum, or corpulent? But did you have any difficulty recognizing them? So while human beings have limited scale compared to machines, we do have an enormous capacity for nuanced, communication and…

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

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning — mentioned by Xmd5a

Wed, 02 Jul 2025 20:45:45 GMT

I don't think this cursory look at the abstracts and first pages of these papers is achieving the effect you intended. The first quote summarize the findings (page 4) of Tversky & Shafir, 1992. Google indicates this paper has 978 citations. >is STONE LION a STONE This is called a typicality test. Jurafsky, Lecun & two more fellows dropped this paper recently: https://arxiv.org/pdf/2505.17117 (emphasis mine) >This work aims to bridge this gap by integrating cognitive psychology, information theory, and modern NLP. We pose three central research questions to guide our investigation: [RQ1]: To…

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

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning — HN submission

Thu, 05 Jun 2025 07:59:21 GMT

From tokens to thoughts: How LLMs and humans trade compression for meaning

124 points; 25 discussion replies on HN. Paper and source history. A submission is not an endorsement.

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning — HN submission

Fri, 30 May 2025 14:48:46 GMT

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning

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