Maria Ryskina
@mryskina.bsky.social
Postdoc @vectorinstitute.ai | organizer @queerinai.com | previously MIT, CMU LTI | 🐀 rodent enthusiast | she/they
🌐 https://ryskina.github.io/
🌐 https://ryskina.github.io/
Congratulations!!!
November 8, 2025 at 12:14 AM
Congratulations!!!
Btw the PI of this work, Dr Kelly Lambert, has a cool book called "The Lab Rat Chronicles" that describes lots of behavioral findings from rat experiments! (Written pre-driving rats, unfortunately)
November 6, 2025 at 4:30 PM
Btw the PI of this work, Dr Kelly Lambert, has a cool book called "The Lab Rat Chronicles" that describes lots of behavioral findings from rat experiments! (Written pre-driving rats, unfortunately)
Congratulations! Took me a second to understand you weren't talking about Lexical Functional Grammar though...
November 5, 2025 at 1:22 PM
Congratulations! Took me a second to understand you weren't talking about Lexical Functional Grammar though...
Isn't mis- (or at least under-)specification inevitable? (I'm thinking of arxiv.org/abs/1804.04268)
Incomplete Contracting and AI Alignment
We suggest that the analysis of incomplete contracting developed by law and economics researchers can provide a useful framework for understanding the AI alignment problem and help to generate a syste...
arxiv.org
October 21, 2025 at 7:22 PM
Isn't mis- (or at least under-)specification inevitable? (I'm thinking of arxiv.org/abs/1804.04268)
The organizers mentioned that the videos will be up a few weeks after the conference! I expect it'll be at www.youtube.com/@colm_conf
October 19, 2025 at 12:19 AM
The organizers mentioned that the videos will be up a few weeks after the conference! I expect it'll be at www.youtube.com/@colm_conf
I still have that card! Still working on that second ice cream 🥲
October 17, 2025 at 5:59 PM
I still have that card! Still working on that second ice cream 🥲
It used to be 5 "no"s for ice cream/pizza! Has the exchange rate gone up?
October 17, 2025 at 5:36 PM
It used to be 5 "no"s for ice cream/pizza! Has the exchange rate gone up?
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
LMs are bad (too rational) at predicting human behaviour, but aligned with humans in assuming rationality in others’ choices.
arxiv.org/abs/2406.17055
October 14, 2025 at 12:43 AM
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
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
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
October 14, 2025 at 12:43 AM
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
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
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
October 14, 2025 at 12:43 AM
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
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
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
October 14, 2025 at 12:43 AM
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
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
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
October 14, 2025 at 12:43 AM
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
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
LMs use sensory language (olfactory, auditory, …) differently from people + evidence that RLHF may discourage sensory language.
arxiv.org/abs/2504.06393
October 14, 2025 at 12:43 AM
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
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...
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...
October 14, 2025 at 12:43 AM
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...
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...