@fededagos
fededagos.me
Joined December 2011
  • Tweets264
  • Following499
  • Followers86
16Media
Can’t wait to finally share what we’ve been up to! This has been both the most work and the most satisfying collaboration of my PhD yet, and we want to put it out there in the cleanest way possible. Paper and tokenizer dashboard will be out soon!
Jai Bhagat
@jkbhagatioOct 5
A little late, but very happy to announce that our work on building and evaluating tokenizers for neural (spiking) data, led by @fededagos , Alex Gilbert, and Susi Keller as a joint collaboration between @metamorphiclabs and @uni_tue , has been accepted as an Oral at Neurips 2026 Evaluations and Datasets track! 🧠 🔤 Was quite a lot of work but equally a lot of fun putting this together, incredibly rewarding to be regarded as a top 15 / ~4000 submissions! We're excited to share a website and dashboard soon 🤓
1
2
8
183
openretina is now out on eLife!
ThomasZenkel@ThomasZenkelJul 23
📣 Our paper introducing openretina, a collaborative framework for retina modelling across datasets and species, is now published as an eLife Reviewed Preprint. 🧵
2
51
Federico D'Agostino retweeted
Sergio Hernández@SergioHzGzJul 2
Long-horizon tasks + sparse rewards = no supervision. So, we need dense supervision methods… but how do we evaluate them? Meet QVal, our latest release👇
1
5
1
18
2,902
Federico D'Agostino retweeted
Vishaal Udandarao@vishaal_uraoJul 2
🚀New Paper arxiv.org/abs/2606.28551 Everyone obsesses over VLM architectures & training recipes. But what about the data? Presenting the latest work in the DataComp-series: a testbed for VLM data curation with 1,000+ controlled experiments and some surprising lessons 👀 🧵👇
11
67
14
258
37,664
Federico D'Agostino retweeted
Vedang Lad
@vedangladJun 16
How well can you describe the feature selectivity of a vision neuron … with words? Interpretability has long borrowed from neuroscience — and maybe it can give back too! 🧵
5
40
12
161
107,243
Federico D'Agostino retweeted
Lukas Thede@lukas_thedeMar 11
🚨 New paper! arxiv.org/abs/2603.06610 Are you sure your post-trained LLM isn’t forgetting something? Adapting LLMs is known to cause forgetting. We usually measure it via general knowledge benchmarks. But if MMLU doesn’t drop…. are you really fine?
1
9
1
28
4,106
Federico D'Agostino retweeted
Ilze Amanda Auzina@AmandaIlzeFeb 13
How can agents learn in long, open-ended tasks where success is rare and rewards are sparse? 👀 🚨 Enter ∆Belief-RL: we show how to use agent’s own belief updates as a dense reward for turn-level credit assignment. the result? Surprisingly strong generalization. (1/8) 🧵⬇️
5
45
10
333
44,033
Federico D'Agostino retweeted
Patrik Reizinger@rpatrik96Jan 22
When I discover a paper on @arxiv, I save it to @zotero, then I want to forget about it. But it gets published, so when I cite it, it still says "arXiv preprint". Multiply by 50. Nobody has time to track all of these. So I built a tool with @claudeai to free up mental space.
3
22
3
211
33,113
🚨 New paper at #NeurIPS2025! A systematic fixation-level comparison of a performance-optimized DNN scanpath model and a mechanistic cognitive model reveals behaviourally relevant mechanisms that can be added to the mechanistic model to substantially improve performance. 🧵👇
1
4
2
8
2,042
more replies
💬 Conceptually: Deep neural networks should be viewed as scientific instruments. They tell us what is predictable in human behavior. We then use that information to ask why, building fully interpretable models that approach the performance of their black-box counterparts.
1
77
This is a concrete step toward bridging the performance/understanding gap in vision science. ⚙️ Code: github.com/bethgelab/what-mo… 📄 Paper: openreview.net/forum?id=cnrw… 🙏 A joint effort with Matthias Kümmerer, Lisa Schwetlick, @bethgelab #NeurIPS #CognitiveModeling
3
137
🚨 New paper alert! 🚨 We’ve just launched openretina, an open-source framework for collaborative retina modeling across datasets and species. A 🧵👇 (1/9)
1
9
1
17
1,247
more replies
This is just the beginning. We see openretina as more than a Python package—it aims to be the start of an initiative to foster open collaboration in computational retina research. We’d love your feedback! (8/9)
1
156
Try it out and help us improve the accessibility of retinal datasets and models together. A team effort with: @ThomasZenkel @dgonschorek.bsky.social @LaraHoefling @teuler.bsky.social @bethgelab #openscience #computationalneuroscience (9/9)
65
Load more