Sign in

Itamar Avitan

@avitanit.bsky.social
31 followers 58 following 11 posts

PhD candidate at Ben-Gurion University.

PostsRepliesMedia
Reposted by Itamar Avitan
Tal Golan @talgolanneuro.bsky.social · 28/08/2026
How can we design experiments that make computational models disagree? One section of our new @natrevneuro.nature.com Review with @kriegeskorte.bsky.social and @heikoschuett.bsky.social examines studies that used stimulus sets designed to elicit distinct predictions from competing models. 1/16
114156
Reposted by Itamar Avitan
Lukas Muttenthaler @lukasmut.bsky.social · 12/11/2025
🥳 I am incredibly humbled and grateful to share that our work, "Aligning machine and human visual representations across abstraction levels," has been published today in @nature.com ⬇️
0215
Reposted by Itamar Avitan
herrick fung @herrickfung.bsky.social · 26/10/2025
No two humans behave exactly alike. But what about neural networks? We found early evidence that human-like individual differences in behavior emerge from networks trained with different initializations. Here’s a peek at our results—to be presented at UniReps & DBM @NeurIPS. Full paper on the way!
biorxiv.org
Human-like individual differences emerge from random weight initializations in neural networks
Much of AI research targets the behavior of an average human, a focus that traces to Turing's imitation game. Yet, no two human individuals behave exactly alike. In this study, we show that artificial...
2113
Itamar Avitan @avitanit.bsky.social · 05/12/2025
Presenting our #NeurIPS2025 work on model–behavior alignment today. Could we even recognize the “right” model of behavior under flexible evaluation? Come chat about DNNs & human visual preception! Hall C-E #2010 Friday (today!) 4:30 – 7:30 PM neurips.cc/virtual/2025...
neurips.cc
NeurIPS Poster Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right OneNeurIPS 2025
032
Itamar Avitan @avitanit.bsky.social · 20/11/2025
Kudos to our NeurIPS 2025 reviewers for thoughtful, human-generated reviews. I’ll be presenting poster #2010 in San Diego on Fri, 5 Dec from 4:30–7:30 p.m. PT. Come say hi! arXiv : arxiv.org/abs/2510.23321 Code and data: github.com/brainsandmachines/oddoneout_model_recovery
arxiv.org
Model-Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn't the Right One
Linearly transforming stimulus representations of deep neural networks yields high-performing models of behavioral and neural responses to complex stimuli. But does the test accuracy of such predictio...
040
Itamar Avitan @avitanit.bsky.social · 20/11/2025
Our work reveals a sharp trade-off between predictive accuracy and model identifiability. Flexible mappings maximize predictivity, but blur the distinction between competing computational hypotheses.
131
Itamar Avitan @avitanit.bsky.social · 20/11/2025
Further analyses showed that linear probing was the culprit. The linear fit warps each model's original feature space, erasing its unique signature and making all aligned models converge toward a human-like representation.
130
Itamar Avitan @avitanit.bsky.social · 20/11/2025
The key dependent measure is how often the data-generating model actually achieves the highest prediction accuracy. The surprising result: even with massive datasets (millions of trials), the best-performing model is often not the right one.
110
Itamar Avitan @avitanit.bsky.social · 20/11/2025
Each simulation worked like this: (1) pick one model from 20 candidate NNs and fit it to human responses; (2) sample a synthetic dataset from that model using NEW triplets; (3) test all 20 models on this generated data, measuring cross-validated prediction accuracy.
110
Itamar Avitan @avitanit.bsky.social · 20/11/2025
We ran model recovery simulations using models fitted to the massive THINGS odd-one-out data shared by @martinhebart.bsky.social , @cibaker.bsky.social et al. Each simulation tested whether a neural network model would “win” the model comparison if it had generated the behavioral data.
110
Itamar Avitan @avitanit.bsky.social · 20/11/2025
In our new NeurIPS 2025 paper, we ask: does better predictive accuracy necessarily mean better mechanistic correspondence between neural networks and human representations? neurips.cc/virtual/2025...
neurips.cc
NeurIPS Poster Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right OneNeurIPS 2025
130
Itamar Avitan @avitanit.bsky.social · 20/11/2025
They also showed that if we nudge the NN representations toward human judgments by linearly transforming the representation space itself crossvalidated prediction accuracy is boosted almost to the reliability bound. arxiv.org/abs/2211.01201
arxiv.org
Human alignment of neural network representations
Today's computer vision models achieve human or near-human level performance across a wide variety of vision tasks. However, their architectures, data, and learning algorithms differ in numerous ways ...
120
Itamar Avitan @avitanit.bsky.social · 20/11/2025
@lukasmut.bsky.social , @lorenzlinhardt.bsky.social et al, showed that neural network representations can be strong predictors of human odd-one-out judgments: the image humans select as “odd” among three is often the one whose activation pattern differs most from the other two.
130
Itamar Avitan @avitanit.bsky.social · 20/11/2025
Excited to share my first paper: Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right One (NeurIPS 2025). link below.
1184