Richard Antonello
@NeuroRJ
Followers
359
Following
441
Media
10
Statuses
224
Postdoc in the Mesgarani Lab at Columbia University. Studying how the brain processes language by using LLMs. (Formerly @HuthLab at UT Austin)
Joined May 2020
RT @katie_kang_: LLMs excel at fitting finetuning data, but are they learning to reason or just parrotingš¦?. We found a way to probe a modeā¦.
0
122
0
RT @sparse_emcheng: We'll be presenting this at #ACL2025 ! Come find me and @tomjiralerspong in Vienna :).
0
2
0
RT @yufan_zhuang: š¤ÆYour LLM just threw away 99.9 % of what it knows. Standard decoding samples one token at a time and discards the rest oā¦.
0
7
0
For those attending NAACL, today I'll be presenting recent work on how we can use language encoding models to identify functional specialization throughout cortex. Stop by my talk at 10:30 at the CMCL workshop!
1
1
15
RT @karansdalal: Today, we're releasing a new paper āĀ One-Minute Video Generation with Test-Time Training. We add TTT layers to a pre-traiā¦.
0
938
0
RT @mariannearr: šØAnnouncing our #ICLR2025 Oral!. š„Diffusion LMs are on the rise for parallel text generation! But unlike autoregressive LMā¦.
0
133
0
RT @DanielCohenOr1: Vectorization into a neat SVG!šØāØ .Instead of generating a messy SVG (left), we produce a structured, compact representaā¦.
0
126
0
RT @GretaTuckute: Our @CogCompNeuro GAC paper is out! We focus on two main questions: . 1ā£ How should we use neuroscientific data in modelā¦.
0
9
0
RT @TomerUllman: "The Illusion Illusion" . vision language models recognize images of illusions. but they also say non-illusions are illuā¦.
0
16
0
RT @GeelingC: Catch me and @czlwang presenting our poster today (12/14) 3:30-5:30pm at the #NeurIPS2024 NeuroAI Workshop! š§ .
0
1
0
RT @patrickmineault: New post! What do brain scores teach us about brains? Does accounting for variance in the brain mean that an ANN is brā¦.
0
67
0
Come by our poster (#3801), exploring how we can use the question-answering abilities of LLMs to build more #interpretable models of language processing in the š§ , starting in one hour at #NeurIPS !.
LLM embeddings are opaque, hurting them in contexts where we really want to understand whatās in them (like neuroscience). Our new work asks whether we can craft *interpretable embeddings* just by asking yes/no questions to black-box LLMs. š§µ
0
1
9