
International Conference on Learning Representations
13
MENTIONS
13
EPISODES
10
PODCASTS
Search complete. 13 mentions across 13 episodes found for "International Conference on Learning Representations".
Nov 22, 2026
Building Developer Assistants that Think Fast and Slow
M
39:17Mayur NaikGUEST
We have deduplicated it.
M
39:19Mayur NaikGUEST
Is it now safe to use it? In our own work this year, which appeared at ICLR or Code Trek, we started out with this synthetic data set.
M
39:35Mayur NaikGUEST
So again, we are looking at variable misuse detection in Python.
M
39:39Mayur NaikGUEST
We started out with real programs that are written by humans.
Small Language Models, Not Giant LLMs, Should Power the Next Generation of AI Agents
H
45:55HackerNoon AINARRATOR
Low-rank adaptation of large language models.
H
45:58HackerNoon AINARRATOR
ICLR, 1, 2, 3, 2022.
H
46:01HackerNoon AINARRATOR
Xiaohan Huang, Furu Wei, Lei Kui, Xingxing Zhang, and Ming Zhou.
H
46:07HackerNoon AINARRATOR
Unsupervised fine-tuning for text clustering.
How to Design Better Antibodies From a Single 96-Well Plate of Data with Cradle
S
19:50Stef van GriekenGUEST
I won't go into too much detail there.
S
19:52Stef van GriekenGUEST
We presented it at ICLR, I think last summer, and it was one of the 12, I think, papers that were presented at that conference in the Foundation Models for Biology.
S
20:05Stef van GriekenGUEST
In terms of the training procedure, so there is three steps.
S
20:10Stef van GriekenGUEST
So the first step that we go through is we want to generate variants.
Beyond the Demo: ATOMS ACCELERATED at a16z Tech Week
G
0:57GustavoHOST
He went on to become a Kaggle Grand Master, serve as CTO of Solaris Therapeutics, and co-founded Olden Labs, which is a hardware robotics spin-off focused on automating animal research.
G
1:10GustavoHOST
His research has been published in Neuro IPS, ICLR, and in Nature Genetics.
G
1:16GustavoHOST
Today, he's the co-founder and CTO of Cosmon, building AI agents that understand engineering intent and operate CAD and simulation tools autonomously.
G
1:25GustavoHOST
Welcome, Noah.
Challenges in Augmenting Large Language Models with Private Data
A
46:20Ashwinee PandaGUEST
So this is privacy-preserving in-context learning for large language models.
A
46:26Ashwinee PandaGUEST
Another paper that I'll be presenting next week at ICLR.
A
46:29Ashwinee PandaGUEST
So let's say that we have some sensitive database of records, and we want to use data from this database in order to answer questions.
A
46:38Ashwinee PandaGUEST
So for example, you have a document question answering question.
Persistent Pre-Training Poisoning of LLMs
S
0:14speaker_0HOST
Today we have Javier and Yiming for PhD students at CMU and ETH.
S
0:20speaker_0HOST
And they're going to be telling us about a paper that just got accepted to ICLR on persistent pre-training poisoning of LLMs.
J
0:31Javier RandoGUEST
Good.
J
0:32Javier RandoGUEST
Yeah, thank you so much for the invite.
Threat Models for Memorization: Privacy, Copyright, and Everything In-Between
M
30:46Michael AerniGUEST
And yet we seem to have sort of we seem to sort of generate text that exists verbatim on the Internet, which is something is going on.
M
30:55Michael AerniGUEST
And what is going on, we call in some other recent work to be presented at ICLR this year, we call this non-adversarial reproduction.
M
31:03Michael AerniGUEST
Just to make the terminology a bit clear, reproduction means an LLM or you also use it for humans generates or generates verbatim training data.
M
31:15Michael AerniGUEST
That is just a reproduced text snippet is a snippet output by an LLM that exists verbatim in its training data.
What makes a scientific result interesting
T
24:24Tal KachmanGUEST
You already see it.
T
24:25Tal KachmanGUEST
ICLR had 62K papers submitted.
T
24:32Tal KachmanGUEST
I'm now an area chair on a conference called AMAS, which is the multi-agent system conference.
T
24:38Tal KachmanGUEST
It was one of my favorite conferences because it was very small and high quality.
LeRobot: Open-Source End-to-End Robot Learning with Steven & Pepijn from Hugging Face
Y
1:18Yacine MahdidHOST
Stuff like why planning might already be mostly solved and where the real bottleneck in robotics might be, why almost no data is bad data anymore, as long as it's labeled right, how frontier models actually are controlling these robots without training on them, whether VLA are dead now that blackboard models are actually something that works, and why robotics is just harder than language.
Y
1:41Yacine MahdidHOST
I'll link the Le Robo repo and their ICLR paper down below.
Y
1:45Yacine MahdidHOST
It's a great companion to this conversation.
Y
1:47Yacine MahdidHOST
Enjoy! Before we jump into this interview, let's talk about Arcee, the openweight lab sponsoring this interview.
Purple Teaming AI Agents & Non-Deterministic Threat Modeling | ft. Dhivya Chandramouleeswaran (AI Safety Research Fellow) | ScaleToZero Podcast | Powered by Cloudanix | Ep.115
D
64:18Dhivya ChandramouleeswaranGUEST
I'm sure there'll be a revision in like November 26.
D
64:21Dhivya ChandramouleeswaranGUEST
But at least for now, based on the papers that we're seeing in like ICLR and NeurIPS, what does it look like? And how I would encourage folks to read that paper and draw to their own conclusions on What does Eval look like in their own organizations or in their own personal projects based on the trends that we are seeing right now? So, yeah, things like bubble teaming, things like what's the granularity to the level we go into for observation and things like that.
D
64:49Dhivya ChandramouleeswaranGUEST
Like all of that is laid out on the paper.
D
64:51Dhivya ChandramouleeswaranGUEST
I think it's a fantastic paper.
3 more episodes mention International Conference on Learning Representations.
Create an account to see the whole feed, search across every transcript, and follow the entities you care about.