LLMs Assume People Are More Rational Than We Really Are by Ryan Liu* & Jiayi Geng* et al.:

LMs are bad (too rational) at predicting human behaviour, but aligned with humans in assuming rationality in others’ choices.

arxiv.org/abs/2406.17055

Neologism Learning by John Hewitt et al.:

Training new token embeddings on examples with a specific property (e.g., short answers) leads to finding “machine-only synonyms” for these tokens that elicit the same behaviour (short answers=’lack’).

arxiv.org/abs/2510.08506

Hidden in Plain Sight by Stephanie Fu et al. [Outstanding paper award]:

VLMs are worse than vision-only models on vision-only tasks – LMs are biased and underutilize their (easily accessible) visual representations!

hidden-plain-sight.github.io

UnveiLing by Mukund Choudhary* & KV Aditya Srivatsa* et al.:

Linguistic olympiad problems about certain linguistic features (e.g., morphological ones) are tougher for LMs, but morphological pre-tokenization helps!

arxiv.org/abs/2508.11260

A Taxonomy of Transcendence by Natalie Abreu et al.:

LMs outperform the experts they are trained on through skill denoising (averaging out experts’ errors), skill selection (relying on the most appropriate expert), and skill generalization (combining experts’ knowledge).

arxiv.org/abs/2508.17669

The Zero Body Problem by @rmmhicke.bsky.social et al.:

LMs use sensory language (olfactory, auditory, …) differently from people + evidence that RLHF may discourage sensory language.

arxiv.org/abs/2504.06393

Readability ≠ Learnability by Ivan Lee & Taylor Berg-Kirkpatrick:

Developmentally plausible LM training works not because of simpler language but because of lower n-gram diversity! Warning against anthropomorphizing / equating learning in LMs and in children.

openreview.net/pdf?id=AFMGb...

We also compare the representational geometries of the models and the brain using RSA, and find significant alignment in all models. For VLMs, it further increases when both text and image stimuli are used, especially in the ventral ROI:

2. Within our ROIs, LM predictivity is correlated with semantic consistency, even where response to meaningful language (preference for sentences over non-words) is low:

In our brain encoding experiments (using LM representations to predict brain responses), we find that:

1. Across the whole brain, areas with higher semantic consistency – plausibly representing concepts across modalities – are better predicted by LMs:

…we define semantic consistency – a measure of how consistently a brain area responds to the same concepts across paradigms – and use it to identify three regions of interest (ROI) where semantically consistent voxels are found:

Using an fMRI dataset of brain responses to stimuli that encode concepts in different paradigms (sentences, pictures, word clouds) and modalities…

Interested in language models, brains, and concepts? Check out our COLM 2025 🔦 Spotlight paper!

(And if you’re at COLM, come hear about it on Tuesday – sessions Spotlight 2 & Poster 2)!

ICML update: met The Transformer at the Bill Reid Gallery!