NeurIPS, every mention
72 scenes · ← back to NeurIPS
tap a year for its mentions
every year anyone Shawn Wang 39Yi Tay 2Pranav Reddy 2Mikhail Parakhin 2Loubna Ben Allal 2Kevin Ben-Smith 2Ari Morcos 2Anjney Midha 2Andy Konwinski 2Andy Beam 2
Verbatim, from the transcripts: the passages where NeurIPS comes up
🔬They Thought the Model Was Broken — Matt McPartlon & Neil Patil, Chai Discovery
- ▶ 1:15:37 unnamed speaker I was at maybe NeurIPS three, four years ago, right?
🔬 RL with Verifiable Rewards, but the Verifier is a Lab — Lila Sciences
🔬 "The Most Innovative Diffusion Research Is Happening in Drug Discovery, Not Image Generation"
- ▶ 57:50 Evan Feinberg I would say there was an eval crisis in our field that is now in transition right now as our field, which was previously, let's say a lot more quiet, but if you go to NeurIPS and ICML every year, year after year, their workshops get a lot…
Why AI Labs With Unlimited GPUs Still Fail — Anjney Midha, AMP
- ▶ 6:41 Anjney Midha They know how to, sure, have they, are they sponsoring happy hours at NeurIps?
- ▶ 39:45 Anjney Midha Like, I remember the first time we met was at NeurIps in New Orleans.
AI-Native Engineering: 100% adoption, 5x search throughput, unlimited tokens — Mikhail Parakhin
- ▶ 50:48 Mikhail Parakhin We talked about it briefly once at NeurIPS. 2 times in the scene
Mistral: Voxtral TTS, Forge, Leanstral, & Mistral 4 — w/ Pavan Kumar Reddy & Guillaume Lample
Agent Inference at the "Speed of Light" — How NVIDIA moves like a $4.3 Trillion Startup
- ▶ 19:21 Shawn Wang Whenever I meet someone who's done work in tabular stuff, graph neural networks, time series, these are basically, when I go to new reps, I go to ICML, I walk the back halls. 2 times in the scene
Claude Code for Finance + The Global Memory Shortage: Doug O'Laughlin, SemiAnalysis
- ▶ 1:18 Shawn Wang I think I first met you at, like, New Orleans or, like, one of the, one of the Neuripses. 2 times in the scene
🔬Generating Molecules, Not Just Models
- ▶ 6:50 unnamed speaker I was at, uh, NeurIPS when I guess the results of this famous competition came out.
Goodfire AI’s Bet: Interpretability as the Next Frontier of Model Design — Myra Deng & Mark Bissell
- ▶ 2:27 unnamed speaker Uh, when you and I talked at NeurIps, you were talking a little bit about your production use cases and your customers, and then not to bury the lead today, we're also announcing the fundraise, uh, your series B, a hundred and fifty…
- ▶ 41:53 unnamed speaker And I mean, like, just, you know, being in Europe's, you see the Interp room, uh, one, like, old school conferences, like, I think they had a very tiny room till they got lucky, and they got it doubled, but there's definitely a lot of…
Captaining IMO Gold, Deep Think, On-Policy RL, Feeling the AGI in Singapore — Yi Tay
Artificial Analysis: The Independent LLM Analysis House — with George Cameron and Micah Hill-Smith
- ▶ 39:25 Shawn Wang At the same time, I do feel like, you know, especially coming back from Europe's people do feel like Ilya is probably right that the paradigm is, doesn't have
[State of Research Funding] Beyond NSF, Slingshots, Open Frontiers — Andy Konwinski, Laude Institute
- ▶ 6:58 Andy Konwinski Yeah, I've been asking people what's been buzzing as I've been talking to people at NeurIPS this year, and this idea that the
- ▶ 17:06 Shawn Wang And you're also doing the LOD Lounge here in NeurIbs. 4 times in the scene
[State of Code Evals] After SWE-bench, Code Clash & SOTA Coding Benchmarks recap — John Yang
- ▶ 0:12 unnamed speaker We're here at NeurIPS with John Yang of SweetBench and many other things, but welcome.
[State of MechInterp] SAEs in Production, Circuit Tracing, AI4Science, "Pragmatic" Interp — Goodfire
[State of AI Papers 2025] Fixing Research with Social Signals, OCR & Implementation — Team AlphaXiv
[NeurIPS Best Paper] 1000 Layer Networks for Self-Supervised RL — Kevin Wang et al, Princeton
- ▶ 0:13 unnamed speaker Uh, we are basically trying to provide the best optimal sort of podcast experience of Europe for people who are not here.
[State of Context Engineering] Agentic RAG, Context Rot, MCP, Subagents — Nina Lopatina, Contextual
- ▶ 0:17 unnamed speaker We're going to talk about the state of context engineering in general, but, uh, one thing I wanted to also give people a sense, we're not here, of the NeurIps flavor discussion and all that. 3 times in the scene
- ▶ 24:47 unnamed speaker This is your fifth NeurIPS. 3 times in the scene
[State of Post-Training] From GPT-4.1 to 5.1: RLVR, Agent & Token Efficiency — Josh McGrath, OpenAI
- ▶ 8:32 unnamed speaker You know, what, what do we know today in, uh, in Neuro's 25 that we didn't know in Neuro's 24? 3 times in the scene
[State of RL/Reasoning] IMO/IOI Gold, OpenAI o3/GPT-5, and Cursor Composer — Ashvin Nair, Cursor
- ▶ 0:00 unnamed speaker Okay, we're here at NeurIps. 4 times in the scene
SAM 3: The Eyes for AI — Nikhila & Pengchuan (Meta Superintelligence), ft. Joseph Nelson (Roboflow)
- ▶ 4:29 Joseph Nelson Like one of the, for example, debtor detection transformers, which was born out of NeurIPS last year.
⚡ Inside Google Labs: Building The Gemini Coding Agent — Jed Borovik, Jules + AIE CODE Preview
- ▶ 16:05 unnamed speaker And then I also, the, the, the concurrent thread to this was, I went to a bunch of the academic ML conferences, NeurIPS, ICML, ICLR. 2 times in the scene
Terminal-Bench 2.0: the most impt coding agent benchmark of 2025 gets a v2! Launch + Q&A w/ founders
- ▶ 24:17 Andy Konwinski Then I met up with Ludwig and Mike and other Chris and others at NeurIPS, I think, is where we first connected about this idea, and Ludwig said, I have this
⚡ Open Model Pretraining Masterclass — Elie Bakouch, HuggingFace SmolLM 3, FineWeb, FinePDF
- ▶ 49:32 Shawn Wang At NeurIPS. 2 times in the scene
Terminal-Bench: Pushing Claude Code, OpenAI Codex, Factory Droid, et al to the limits
- ▶ 34:54 unnamed speaker I think I talked to him at one of the Neuripses, but, uh, yeah.
Why RL Won — Kyle Corbitt, OpenPipe (acq. CoreWeave)
- ▶ 18:56 Shawn Wang I think, I think they were at NeurIPS for two, 23 and people were very excited about it. 2 times in the scene
A Technical History of Generative Media
- ▶ 55:39 unnamed speaker I was part of a group at NeurIPS that we meet at NeurIPS every year. 2 times in the scene
Better Data is All You Need — Ari Morcos, Datology
- ▶ 27:27 Ari Morcos And in many ways, actually the paper that was the foundational paper for datology, it's called Beyond Neural Scaling Laws, and was fortunate to get a best paper at NeurIPS.
- ▶ 39:47 Ari Morcos Um, and I actually, that paper I mentioned that was at NeurIPS, that one actually has data pruning, uh, in the title.
The RLVR Revolution — with Nathan Lambert (AI2, Interconnects.ai)
- ▶ 0:35 Shawn Wang Uh, when you just joined AI two, then you release all the almost, you joined us again at NeurIps where you did the, uh, open models.
⚡️Math Olympiad gold medalist explains OpenAI and Google DeepMind IMO Gold Performances
- ▶ 29:16 Dr. Jasper Zhang Uh, and, um, but then I, I would say none of this will probably worth, like, the best paper in add new ribs, but it's more about, like, you just be the best in terms of each technique and just get more data, get more compute.
Information Theory for Language Models: Jack Morris
- ▶ 10:52 Shawn Wang I like met them at Europe and like they had nobody at their, like there was presenting the speech three.
The Utility of Interpretability — Emmanuel Amiesen
- ▶ 44:39 unnamed speaker Like, they mailed out pieces of the Golden Gate for people at NeurIps, I think, or ICML.
GPT 4.1: The New OpenAI Workhorse
- ▶ 14:21 Shawn Wang Um, I, for listeners who, who've been listening to us for a while, we actually covered this paper in Europe's last year, uh, two years ago, cog eval, where they actually modeled graphs for graph traversals for agent planning.
The #1 SWE-Bench Verified Agent
- ▶ 28:34 Guy Gur-Ari I think it was NeurIps.
Snipd: The AI Podcast App for Learning — with CEO Kevin Ben-Smith
- ▶ 1:08:32 Kevin Ben-Smith I remember, I was at the, I think it was, uh, Nips Conference, back when in San Diego? 2 times in the scene
Gemini 2.0 Flash and Flash Thinking: the new SOTA models for the agentic era
- ▶ 16:57 unnamed speaker Um, that Paige, like on the day of Gemini to Flash's launch, came over to our New Rips event and demoed it.
The Agent Reasoning Interface: Claude, ChatGPT Canvas, Tasks, Operator — with Karina Nguyen, OpenAI
- ▶ 14:43 Shawn Wang Oh, because it was like relatively new after new RIPS?
The Unreasonable Effectiveness of Reasoning Distillation: using DeepSeek R1 to beat OpenAI o1
- ▶ 1:05 Shawn Wang We, uh, I actually interviewed both at the original Datacomp and the new Datacomp LM team, uh, at NeurIps.
DeepSeek V3, SGLang, and the state of Open Model Inference in 2025 (Quantization, MoEs, Pricing)
- ▶ 14:41 unnamed speaker I went to Jeff Dean's session in Europe's, uh, and he also mentioned as well that I think one of the Gemini, I think Gemini F-Five Pro is an MOE, which I don't think we knew before that.
The State of Reasoning — from Nathan Lambert, Interconnects/AI2 [LS Live @ NeurIPS 2024]
- ▶ 0:09 Nathan Lambert This is a quick talk that I gave at NeurIPS, uh, the latent space unofficial industry event.
2024 Year in Review: The Big Scaling Debate, the Four Wars of AI, Top Themes and the Rise of Agents
- ▶ 3:17 Shawn Wang View the conference as like NeurIpsis, I think like 16, 17,000 people. 4 times in the scene
- ▶ 13:53 Shawn Wang Researchers that I talked with in Europe's were kind of positive on this because basically you need private test sets to prevent contamination. 2 times in the scene
- ▶ 27:10 Shawn Wang I tried to go to Neurobs in order to find these kinds of papers because 6 times in the scene
- ▶ 1:06:58 Shawn Wang There's literally a benchmark called frontier math, which I spent a bit of time talking about at NeurIPS. 2 times in the scene
- ▶ 1:15:24 Shawn Wang Um, I do think the XLSTM paper, which we will cover when we cover the sort of NeurIPS papers, um, is worth a look.
- ▶ 1:41:54 Shawn Wang Every new reps that I go, I have some chats with researchers and I'll just highlight the best prediction from that group.
- ▶ 1:47:44 Shawn Wang And, you know, with some amount of conclusion on, in, in Europe's that we were also kind of pointing to in the winter, yeah, winter episode, but like, it's not a winter by any means.
Best of 2024: Synthetic Data / Smol Models, Loubna Ben Allal, HuggingFace [LS Live! @ NeurIPS 2024]
- ▶ 18:38 Loubna Ben Allal For example, here, I asked it to give me a joke about NeurIPS. 2 times in the scene
The State of AI Startups in 2024 [LS Live @ NeurIPS]
- ▶ 0:33 Shawn Wang It's very nice that Europe is always the end of the year.
- ▶ 3:15 Pranav Reddy So the, um, I, I thought it'd be useful to start from, like, what was happening at NeurIPS last year in December, twenty-twenty-three. 2 times in the scene
- ▶ 26:06 unnamed speaker Thanks for joining our first talk from Latent Space Live at NeurIPS, in Vancouver.
In the Arena: How LMSys changed LLM Benchmarking Forever
- ▶ 0:24 Shawn Wang Anastasius, I actually saw you, I think, at last year's NeurIPS. 2 times in the scene
The Ultimate Guide to Prompting - with Sander Schulhoff from LearnPrompting.org
- ▶ 37:32 Shawn Wang Like, I think there was a golden period where you published, like, an X of thought paper, and you could get into, like, new reps or something.
The Winds of AI Winter (Q2 Four Wars of the AI Stack Recap)
- ▶ 32:15 unnamed speaker I actually interviewed Datacomp at NeurIPS last year, and they've branched out from just vision and images to language models.
- ▶ 46:59 unnamed speaker I was kind of bearish on conformers because I looked at the state of existing conformer research in ICM, Eclair, and, and NeurIPS, and they were far, far, far behind Whisper, mostly because of scale, like the, the sheer amount of resources…
The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
- ▶ 20:44 unnamed speaker Yeah, I covered him on, on my New Rips podcast. 4 times in the scene
- ▶ 35:16 Yi Tay So I think Hiohwa scares me because it's like, I think that was at New Rips, like, we were doing some work at the New Orleans,
- ▶ 1:44:51 unnamed speaker But more generally, efficiency, in my mind, ah, when I go to ICLEAR, I go to NeurIPS, I see efficiency paper, 90% of the chance, like, I'm just gonna ignore it.
Open Source AI is AI we can Trust — with Soumith Chintala of Meta AI
- ▶ 51:49 unnamed speaker It seems like, I think in Europe's I walked through a lot of the posters and whatnot, there seems to be mold collapse in a way in the research, a lot of people working on, on the same things.
The Four Wars of the AI Stack - Dec 2023 Recap
- ▶ 18:06 unnamed speaker Uh, so I, I would say, like, that, that is, when I went to NeurIPS, that was the number one discussion out of every single researcher's mouth. 2 times in the scene
The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
- ▶ 1:15:28 unnamed speaker Uh, so as of time of recording, we actually about to publish our NeurIps recap where we talk to the authors.
The State of Silicon and the GPU Poors - with Dylan Patel of SemiAnalysis
- ▶ 1:01:12 Dylan Patel Like, I think there's even stories of, like, like, you know, not so long ago, NeurIps, when it used to have the more unfortunate name, he would go there and just give away GPUs to people, right?
The End of Finetuning — with Jeremy Howard of Fast.ai
- ▶ 28:15 Jeremy Howard Yeah, actually, Andre Capathy grabbed me when I saw him at NeurIpes a few years ago, and he's like, I have to tell you, thanks to the fast AI courses, when people come to Tesla, and they need to know more about deep learning, we always…