nick ichien

ucla cogsci postdoc

[Liuba Papeo @ljuba_pi]

The September issue @CurrentBiology What do infants think when they see social interactions? 👶🏻
The representation of interacting people as agent and patient reveals a language-of-thought before language development
https://t.co/jfG7Lz0Epz
@JR_Hochmann @ISC_MJ @BabyLyon11 🧵👇

[Alon Hafri @AlonHafri]

Curious about the origins of structured thought? In the latest @CurrentBiology I share some thoughts on @ljuba_pi @JR_Hochmann & colleagues' new study showing that infants grasp complex social relations, w/ implications for language & thought. Take a look!
https://t.co/hVIQ79qb5n

[nick ichien @IchienNick]

@rachelfheaton Congrats Rachel!

[Iris van Rooij 💭 @IrisVanRooij]

🚨Our paper `Reclaiming AI as a theoretical tool for cognitive science' is now forthcoming in the journal Computational Brain & Behaviour (@CompBrainBeh).

Below a thread summary 🧵1/n

#metatheory #AGI #AIhype #cogsci #theoreticalpsych #criticalAIliteracy

[Abhishek Dedhe | अभिषेक देढे @AbhishekDedhe]

Had a great time presenting at #Analogy2024 and #CogSci2024 about the algorithms underlying hierarchical patterning!

Thanks to @ClaireS012345, @margipavlova, and @IchienNick for organizing!

[Dr Laura Speed @laurajspeed]

Excited for @asifa_majid 's Elman symposium this afternoon @cogsci_soc Please RT because our talks aren't listed in the program #CogSci2024

[nick ichien @IchienNick]

@MaryCWhatley1 @UCLA @WCU Congrats Mary!

[nick ichien @IchienNick]

@notnaughtknot Come visit anytime!

[nick ichien @IchienNick]

@notnaughtknot Congrats!!

[Leor Hackel @leorhackel]

Dave Kalkstein presents work on person centered cognition at #SPSP2024:

The presence of a person changes how we process a visual scene, leading us to understand the world in terms of how it relates to that person.

See our paper here: https://t.co/RSU8yDv0WA

[Ben Rottman @BenRottman]

Congrats to my grad student Yiwen Zhang on two papers published on causal learning with experiences spaced out over many days.

[I.A.B @IbanDlank]

Are Large Language Models good at modeling how humans process language? A new Forum piece in @TrendsCognSci is a reminder that the question itself has different interpretations, each with its own consequences. https://t.co/nBi5hbkde8

[Melanie Mitchell @MelMitchell1]

Excellent piece by @strwbilly on evaluation issues for LLMs.

Includes comments from @TaylorWWebb , @NatalieShapira, @weidingerlaura, @LucyCheke, and me.

https://t.co/QzXmUeR0mg

[nick ichien @IchienNick]

So despite model-generated interpretations being judged as of higher quality than human-generated interpretations, both reflect similar sensitivity to pragmatic constraints on communication

[nick ichien @IchienNick]

New preprint with Keith Holyoak and Dusan Stamenkovic: GPT-4-generates interpretations of novel English-translated Serbian metaphors that are judged as being of higher quality than human-generated interpretations (1/n)
https://t.co/mXRPc2GcTC

[nick ichien @IchienNick]

For example, interpretations of “A man is a butterfly” and “A butterfly is a man” both imply a characterization of "man" (e.g., “A man goes through many phases of growth in order to reach his full potential.”) (3/n)

[Anna Leshinskaya @AnnaLeshinskaya]

This work is now published! We show complementary roles for lateral entorhinal cortex and middle temporal gyrus in building integrative temporal relational knowledge from experiences of events. EC integrates rapidly, MTG builds slowly.

https://t.co/RzmYJthjTb

[nick ichien @IchienNick]

@having__trouble hot

[LLM Security @llm_sec]

* People ask LLMs to write code
* LLMs recommend imports that don't actually exist
* Attackers work out what these imports' names are, and create & upload them with malicious payloads
* People using LLM-written code then auto-add malware themselves

https://t.co/Va9w18RpWu