This is real. People get angry when dating data gets leaked because it shows, to no surprise, that people lean into their worst tendencies when an application makes an incentive structure out of it. Individual *ordered* preferences become cascading and turn dating into slanted unimodal distribution.
December 28, 2024 at 11:30 PM
This is real. People get angry when dating data gets leaked because it shows, to no surprise, that people lean into their worst tendencies when an application makes an incentive structure out of it. Individual *ordered* preferences become cascading and turn dating into slanted unimodal distribution.
The median eukaryote has 16 chromosomes. It’s perhaps surprisingly unimodal. All this and even more here: hanliconius.github.io/eukaryote-ge...
September 29, 2026 at 3:05 PM
The median eukaryote has 16 chromosomes. It’s perhaps surprisingly unimodal. All this and even more here: hanliconius.github.io/eukaryote-ge...
We present Gene Regulatory nETwork Analsyis (GRETA), a framework to infer, compare and evaluate gene regulatory networks #GRNs. With it, we have benchmarked multimodal and unimodal GRN inference methods. Check the results here 👇
Paper: doi.org/10.1101/2024.12.20.629764
Code: github.com/saezlab/greta
Paper: doi.org/10.1101/2024.12.20.629764
Code: github.com/saezlab/greta
December 23, 2024 at 8:43 AM
We present Gene Regulatory nETwork Analsyis (GRETA), a framework to infer, compare and evaluate gene regulatory networks #GRNs. With it, we have benchmarked multimodal and unimodal GRN inference methods. Check the results here 👇
Paper: doi.org/10.1101/2024.12.20.629764
Code: github.com/saezlab/greta
Paper: doi.org/10.1101/2024.12.20.629764
Code: github.com/saezlab/greta
New year, new preprint (+ last thesis paper!)
Macroscale gradients of intracranial EEG power spectra follow familiar axes: sensory-motor and unimodal-transmodal 🕵️♀️
Gradients show overall stronger association to distance, but multimodal information was best in transmodal areas 🧠
tinyurl.com/ykacndd5
Macroscale gradients of intracranial EEG power spectra follow familiar axes: sensory-motor and unimodal-transmodal 🕵️♀️
Gradients show overall stronger association to distance, but multimodal information was best in transmodal areas 🧠
tinyurl.com/ykacndd5
January 8, 2025 at 12:55 PM
New year, new preprint (+ last thesis paper!)
Macroscale gradients of intracranial EEG power spectra follow familiar axes: sensory-motor and unimodal-transmodal 🕵️♀️
Gradients show overall stronger association to distance, but multimodal information was best in transmodal areas 🧠
tinyurl.com/ykacndd5
Macroscale gradients of intracranial EEG power spectra follow familiar axes: sensory-motor and unimodal-transmodal 🕵️♀️
Gradients show overall stronger association to distance, but multimodal information was best in transmodal areas 🧠
tinyurl.com/ykacndd5
#New_accepted_paper
Jorge Olivares-Vinales and *Semin Yoo*,
Towers and Bratteli-Vershik systems in Fibonacci-like unimodal maps,
Int. Math. Res. Not. IMRN, accepted, 2026.
arxiv.org/abs/2602.21623
Jorge Olivares-Vinales and *Semin Yoo*,
Towers and Bratteli-Vershik systems in Fibonacci-like unimodal maps,
Int. Math. Res. Not. IMRN, accepted, 2026.
arxiv.org/abs/2602.21623
Towers and Bratteli-Vershik systems in Fibonacci-like unimodal maps
For a class of Fibonacci-like unimodal maps, the restriction to the $ω$-limit set of the unique turning point defines a minimal Cantor system. We construct these Cantor sets geometrically using a nest...
arxiv.org
October 1, 2026 at 3:26 AM
#New_accepted_paper
Jorge Olivares-Vinales and *Semin Yoo*,
Towers and Bratteli-Vershik systems in Fibonacci-like unimodal maps,
Int. Math. Res. Not. IMRN, accepted, 2026.
arxiv.org/abs/2602.21623
Jorge Olivares-Vinales and *Semin Yoo*,
Towers and Bratteli-Vershik systems in Fibonacci-like unimodal maps,
Int. Math. Res. Not. IMRN, accepted, 2026.
arxiv.org/abs/2602.21623
once again I'm begging the people who write these sorts of headlines to learn a crumb of statistics
"average" = mean, median, or mode (usually mean)
in all cases "average" depends on distribution
all assume a single mode
most things aren't unimodal
P.S. I hate how easy it was to make this graph
"average" = mean, median, or mode (usually mean)
in all cases "average" depends on distribution
all assume a single mode
most things aren't unimodal
P.S. I hate how easy it was to make this graph
October 15, 2025 at 4:46 AM
once again I'm begging the people who write these sorts of headlines to learn a crumb of statistics
"average" = mean, median, or mode (usually mean)
in all cases "average" depends on distribution
all assume a single mode
most things aren't unimodal
P.S. I hate how easy it was to make this graph
"average" = mean, median, or mode (usually mean)
in all cases "average" depends on distribution
all assume a single mode
most things aren't unimodal
P.S. I hate how easy it was to make this graph
Looked unimodal for a bit but now we're back
July 29, 2026 at 1:58 PM
Looked unimodal for a bit but now we're back
Who uses Bluesky? We look at the political leanings of Users, and investigate polarisations and political leanings on Bluesky. Bluesky users are almost exclusively center-left leaning with a unimodal distribution of political opinions.
June 20, 2024 at 7:08 PM
Who uses Bluesky? We look at the political leanings of Users, and investigate polarisations and political leanings on Bluesky. Bluesky users are almost exclusively center-left leaning with a unimodal distribution of political opinions.
Unimodal areas come online early and the MTG only much later
September 30, 2026 at 1:08 PM
Unimodal areas come online early and the MTG only much later
In “Better Together: Leveraging Unpaired Multimodal Data for Stronger Unimodal Models,” we study a question I’ve wanted to make progress on for years: can you learn useful multimodal representations from *unpaired* data?
5/9
5/9
October 10, 2025 at 10:13 PM
In “Better Together: Leveraging Unpaired Multimodal Data for Stronger Unimodal Models,” we study a question I’ve wanted to make progress on for years: can you learn useful multimodal representations from *unpaired* data?
5/9
5/9
It's Day 10 of the #30DayChartChallenge, and the prompt is "Multi-modal" 📊
📈 Showing distribution of % of wealth received by richest 1% from 1980-2023
🔴 = higher inequality
💻 Made with {ggridges} in #RStats
#Day10 #DataViz
📈 Showing distribution of % of wealth received by richest 1% from 1980-2023
🔴 = higher inequality
💻 Made with {ggridges} in #RStats
#Day10 #DataViz
April 10, 2025 at 9:08 AM
It's Day 10 of the #30DayChartChallenge, and the prompt is "Multi-modal" 📊
📈 Showing distribution of % of wealth received by richest 1% from 1980-2023
🔴 = higher inequality
💻 Made with {ggridges} in #RStats
#Day10 #DataViz
📈 Showing distribution of % of wealth received by richest 1% from 1980-2023
🔴 = higher inequality
💻 Made with {ggridges} in #RStats
#Day10 #DataViz
Ecologists recognize the importance of integrating across scales, yet our models often target specific scales. Our new paper presents models of population growth that scale up to explain community-level patterns, including interactions of temperature, light & nutrients. doi.org/10.1002/ecm....
December 9, 2025 at 2:41 PM
Ecologists recognize the importance of integrating across scales, yet our models often target specific scales. Our new paper presents models of population growth that scale up to explain community-level patterns, including interactions of temperature, light & nutrients. doi.org/10.1002/ecm....
As we progress from unimodal to multimodal w/foundation models of A.I., we're moving from the "scan era" to the "everything era" in medicine (with radiology the leading force)
brief summary of my @rsnasky.bsky.social talk #RSNA2024 today
www.auntminnie.com/resources/co...
@auntminnie.bsky.social
brief summary of my @rsnasky.bsky.social talk #RSNA2024 today
www.auntminnie.com/resources/co...
@auntminnie.bsky.social
Topol: AI gearing for 'everything era'
Eric Topol, MD, founder and director of the Scripps Research Translational Institute discussed AI’s progress toward the future at RSNA 2024.
www.auntminnie.com
December 2, 2024 at 10:34 PM
As we progress from unimodal to multimodal w/foundation models of A.I., we're moving from the "scan era" to the "everything era" in medicine (with radiology the leading force)
brief summary of my @rsnasky.bsky.social talk #RSNA2024 today
www.auntminnie.com/resources/co...
@auntminnie.bsky.social
brief summary of my @rsnasky.bsky.social talk #RSNA2024 today
www.auntminnie.com/resources/co...
@auntminnie.bsky.social
Check out our new preprint, which presents object-based retrieval processes in multisensory working memory. Here we show that unimodal feature probes incidentally reactivate untested tones and orientations of audiovisual objects. +
Neural reinstatement of features in audiovisual working memory indicatesobject-based retrieval https://www.biorxiv.org/content/10.64898/2026.02.06.704386v1
February 17, 2026 at 2:34 PM
Check out our new preprint, which presents object-based retrieval processes in multisensory working memory. Here we show that unimodal feature probes incidentally reactivate untested tones and orientations of audiovisual objects. +
3) They tried to fit a unimodal normal curve to the onset data but that didn't fit well (Hartigan's Dip Test was significant). Using three modes also caused problems but if they used a bimodal distribution, the means and standard deviations were almost the same in all countries.
March 23, 2026 at 9:34 AM
3) They tried to fit a unimodal normal curve to the onset data but that didn't fit well (Hartigan's Dip Test was significant). Using three modes also caused problems but if they used a bimodal distribution, the means and standard deviations were almost the same in all countries.
So, until/if we start imaging the surfaces of these worlds, what might we learn about them?
Well, we argue that the Solar System's rocky worlds provide a useful template to start from.
And that's the gist of the paper! So we start with Earth, Venus, Mars, Mercury, and the Moon.
Well, we argue that the Solar System's rocky worlds provide a useful template to start from.
And that's the gist of the paper! So we start with Earth, Venus, Mars, Mercury, and the Moon.
September 23, 2026 at 1:57 AM
So, until/if we start imaging the surfaces of these worlds, what might we learn about them?
Well, we argue that the Solar System's rocky worlds provide a useful template to start from.
And that's the gist of the paper! So we start with Earth, Venus, Mars, Mercury, and the Moon.
Well, we argue that the Solar System's rocky worlds provide a useful template to start from.
And that's the gist of the paper! So we start with Earth, Venus, Mars, Mercury, and the Moon.
Emmorey et al (2008) was wrong (underpowered)! New paper by Asanowicz et al. shows that bimodal bilinguals (Codas), like unimodal bilinguals, exhibit enhancements in specific aspects of executive control. Codas also showed unique effects of visuospatial attention.
psycnet.apa.org/fulltext/202...
psycnet.apa.org/fulltext/202...
APA PsycNet
psycnet.apa.org
October 17, 2025 at 3:11 PM
Emmorey et al (2008) was wrong (underpowered)! New paper by Asanowicz et al. shows that bimodal bilinguals (Codas), like unimodal bilinguals, exhibit enhancements in specific aspects of executive control. Codas also showed unique effects of visuospatial attention.
psycnet.apa.org/fulltext/202...
psycnet.apa.org/fulltext/202...
While unimodal sensory input to the hippocampal region was gradually eliminated with the increase in species brain size, transmodal input was relatively preserved.
So - does memory mean different things for different animals?? (2/3)
So - does memory mean different things for different animals?? (2/3)
September 8, 2026 at 4:46 PM
While unimodal sensory input to the hippocampal region was gradually eliminated with the increase in species brain size, transmodal input was relatively preserved.
So - does memory mean different things for different animals?? (2/3)
So - does memory mean different things for different animals?? (2/3)
I wonder what Ilhan’s looks like. Probably unimodal young I’d guess.
October 16, 2025 at 10:13 PM
I wonder what Ilhan’s looks like. Probably unimodal young I’d guess.
By modelling brain responses to brief visual stimuli, we define a functional hierarchy of time processing and perception: journals.plos.org/plosbiology/...
So very happy to see this work out in @plosbiology.org! Another great team effort with @gianfrancof.bsky.social and @dbueti.bsky.social
So very happy to see this work out in @plosbiology.org! Another great team effort with @gianfrancof.bsky.social and @dbueti.bsky.social
Neuronal populations across the cortex underlie discrete, categorical, and subjective representations of visual durations
Duration perception involves multiple brain regions, but how unimodal tuning across this network supports subjective timing remains unclear. This study shows that parietal, premotor, and caudal SMA en...
journals.plos.org
March 28, 2026 at 10:19 AM
By modelling brain responses to brief visual stimuli, we define a functional hierarchy of time processing and perception: journals.plos.org/plosbiology/...
So very happy to see this work out in @plosbiology.org! Another great team effort with @gianfrancof.bsky.social and @dbueti.bsky.social
So very happy to see this work out in @plosbiology.org! Another great team effort with @gianfrancof.bsky.social and @dbueti.bsky.social
New review out on facial signalling in conversation, making the argument that the notion of ‘speech acts’ is outdated due to its unimodal focus, especially in the light of the fundamental contributions of the face towards social action communication www.sciencedirect.com/science/arti...
Facial clues to conversational intentions
It has long been known that we use words to perform speech acts foundational to everyday conversation, such as requesting, informing, proposing, or co…
www.sciencedirect.com
May 12, 2025 at 7:10 AM
New review out on facial signalling in conversation, making the argument that the notion of ‘speech acts’ is outdated due to its unimodal focus, especially in the light of the fundamental contributions of the face towards social action communication www.sciencedirect.com/science/arti...
Human cortical dynamics reflect graded contributions of local geometry and network topography
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Human cortical dynamics reflect graded contributions of local geometry and network topography - Nature Communications
Royer and colleagues find a principal dimension of neural dynamic similarity dissociating unimodal and association cortices. Systematic variations in local and macroscale constraints across the cortex...
www.nature.com
August 21, 2026 at 1:16 PM
Human cortical dynamics reflect graded contributions of local geometry and network topography
www.nature.com/articles/s41...
www.nature.com/articles/s41...
How is the input of the different kinds of primary sensory neurons responding to heat and mechanical stimuli summarized in the spinal cord? Check out our recent study.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Neural ensembles that encode nocifensive mechanical and heat pain in mouse spinal cord - Nature Neuroscience
Zhang et al. identify unimodal neural representations in the spinal cord of cutaneous mechanical and heat stimuli gated by a shared feed-forward local inhibitory neuron type and a neural transition du...
www.nature.com
March 24, 2025 at 4:07 PM
How is the input of the different kinds of primary sensory neurons responding to heat and mechanical stimuli summarized in the spinal cord? Check out our recent study.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
(7/n) We found that multimodal objectives on egocentric data actually harm unimodal performance. This only improves when interleaving unimodal losses during training, which acts as a regularizer to preserve independent modality learning when vision-language alignment is weak.
June 26, 2026 at 10:51 PM
(7/n) We found that multimodal objectives on egocentric data actually harm unimodal performance. This only improves when interleaving unimodal losses during training, which acts as a regularizer to preserve independent modality learning when vision-language alignment is weak.