@SquareZollo

girl/boy dad & adventures @zemelgroup. formerly barstool and music stuff.

 Tarrytown, NY
Joined February 2011
  • Tweets177
  • Following698
  • Followers1.5K
18Media
DeltaNet-PC is out! It turns out that predictive coding offered a native mechanism for test-time memorization of sequential data all along. The crucial insight was to treat updates to the fast associative memory as an inferred latent variable, not a long term learned weight.
2
14
2
83
7,015
tom zollo retweeted
Iain Dunning
@iaindunningSep 23
Announcing the 2nd iteration of our PhD Winter Internship, for Jan 2027. 1st iteration this year was a big success - PhDs from variety of disciplines got to have a taste of working at HRT, then return later for summer internships and/or full time work. hudsonrivertrading.com/hrt-j…

PhD Winter Internship - 2027

​​The application deadline for this position is Oct 30, 2026 HRT is seeking qualified PhD candidates for our Winter Internship Program. Our PhD Winter Internship program is a 3 week introduction to...

hudsonrivertrading.com
3
6
1
140
20,253
tom zollo retweeted
will brown
@willcbAug 29
calibration is such a deeply under-studied problem and its importance keeps becoming more apparent every day
20
15
7
244
31,866
Our new preprint on parallelizing training of temporally precise spiking neural networks is out! We show up to 44x speedups over a conventional sequential baseline. 1/N
1
4
2
16
1,656
tom zollo retweeted
Zemel Group@zemelgroupMar 5
New work on continual learning and controllable memory from the Zgroup!
Max S. Bennett
@maxsbennettFeb 27
New paper: Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language. We built a neural memory system that lets you tell an AI — in natural language — what to remember and what to ignore over time. 🧠👇
1
4
521
Love this direction! @SquareZollo looks again like great work - congrats!
tom zollo
@SquareZolloFeb 27
Super psyched to finally share our new continual learning paper “Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language”
1
1
3
589
New paper: Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language. We built a neural memory system that lets you tell an AI — in natural language — what to remember and what to ignore over time. 🧠👇
1
4
1
17
3,928
Deadlines? Get it done, yesterday. Agents for Excel Workflows. 👇
10
12
2
46
3,880
Max Fried's 1st Regular Season in Pinstripes: - 19 Wins (leads @MLB) - 2.86 ERA - 189 Ks - All-Star
Max Fried pregame at Yankee Stadium.

ALT Max Fried pregame at Yankee Stadium.

125
1,342
131
12,737
369,841
tom zollo retweeted
MeekMill
@MeekMill31 Aug 2025
Working on a ai tool that can change the world lol
1,032
1,634
1,671
12,271
8,410,648
New work on fair decision-making with LLMs from @Zhall333 et al!
Zara Hall@Zhall33311 Aug 2025
Increasing test-time compute can lead to more accurate LLM decisions, but also more unfair. How can we reap the benefits of modern inference techniques while also ensuring unbiased decision making? We explore this question in our new paper! 🧵
2
2
8,431
tom zollo retweeted
Zara Hall@Zhall33311 Aug 2025
Increasing test-time compute can lead to more accurate LLM decisions, but also more unfair. How can we reap the benefits of modern inference techniques while also ensuring unbiased decision making? We explore this question in our new paper! 🧵
1
1
2
4
14,308
tom zollo retweeted
Jimmy Wang@jwang7717 Aug 2025
Super excited to share that my paper with @SquareZollo, @zemelgroup, @hsnamkoong won a best paper award at the ICLR 2025 Workshop on Quantifying Uncertainty in Foundation Models!
tom zollo
@SquareZollo14 Jul 2025
How can we train LLMs to: 1) Accurately quantify uncertainty 2) Dynamically ask the right questions to reduce it? Excited to share our new #ICML2025 work on this topic!
1
2
5
3,488
tom zollo retweeted
MLB
@MLB6 Aug 2025
104 MPH IN 107.1 MPH OUT Lourdes Gurriel Jr. ties it!
136
253
185
3,247
557,240
So excited to hear that Kathy won the ACL Lifetime Achievement Award! I feel incredibly fortunate and honored to have had her as one of my PhD advisors, and I’ve learned so much from her over the years. Big congrats!
ACL 2027@aclmeeting30 Jul 2025
🕊️ Lifetime Achievement Award at #ACL2025NLP A standing ovation for Prof. Kathy McKeown, recipient of the ACL 2025 Lifetime Achievement Award! 🌟
1
13
2,449
Vision-Language-Action models are becoming the go-to architecture for robot generalization. But we know that to be reliable in high-stakes and open world environments, robots must “know what they don’t know,” so that they can recognize tasks where they are likely to fail.
1
3
3
1,566
tom zollo retweeted
Zhun Deng
@zhun_deng15 Jul 2025
Excited to introduce our new work at ICML 2025: 1. Conformal Risk Control for LLM Alignment, arxiv.org/pdf/2502.20285 with @lihua_lei_stat 2. Auto-Eval for Quantile-Based Risk Measures arxiv.org/pdf/2507.05220 with @zemelgroup @SquareZollo Please take a look if interested!
3
16
1,641
great fun sharing PersonalLLM at ICLR with @SquareZollo! will be presenting data mixtures optimization with bayes opt tomorrow at Data-FM
1
4
458
.@JimmyRandazzo breaks down the Yankees' #OpeningDay win over Milwaukee, Devin Williams' rocky debut in pinstripes and how the Bombers can have a successful 2025 campaign #RepBx
3
9
22
5,001
Think your AI passed all fairness tests? Think again 🤔 A must-read paper on discrimination testing for GenAI Shows critical gaps between current fairness testing methods and regulatory needs, with compelling case studies on how these gaps enable discriminatory outcomes 👇👇👇
Talia Gillis@TaliaGillis2 Jan 2025
🚨New Working Paper! 🚨 What does "fair and non-discriminatory" mean for GenAI? How can it even be achieved? We explore how discrimination law and AI regulation often misalign with current bias evaluation methods
1
2
4
1,434
Load more