02:03

Training LLMs as a Startup

02:52

What Led to the State of Foundation Model Training Report

06:37

Key Challenges and Success Factors in Foundation Model Training

02:40

Dense Backpropagation Improves Routing for Sparsely-Gated Mixture-of-Experts

04:41

LLM Training Decisions: Why Mindset is Everything

06:27

What Drives Companies to Build Their Own Foundation Models

02:41

What are Foundation Models?

02:44

Self-Stitching: Efficient Transfer Learning Using Stitching Layer

21:36

Fireside Chat on the State of Foundation Model Training Report

03:06

XLAND-100B: Large-scale Multi-task Dataset for In-context Reinforcement Learning

04:35

STaRK: Benchmarking LLM Retrieval on Textual and Relational Knowledge Bases

18:21

Demo | neptune.ai | Experiment Tracker for Foundation Model Training

01:51

MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs

01:07

What’s on Horizon for Cleareye.ai

02:53

ALPS: Improved Optimization for Highly Sparse One-Shot Pruning for LLMs

04:07

NeurIPS 2024: Impossible GenAI Questions With Karol Lasocki

06:56

Efficient Multi-Prompt Evaluation of LLMs

01:16

Growing AI Research Team at DeepL

03:13

NeurIPS 2024: Impossible GenAI Questions With Christianah Oyewale

03:08

NeurIPS 2024: Impossible GenAI Questions With Finn Schmidt

04:05

Minimising Outlier Features in Neural Network Training

04:47

Lessons From LLM Training: Insights From Genentech

02:35

NeurIPS 2024: Impossible GenAI Questions With Yinan Wang

04:43

NeurIPS 2024: Impossible GenAI Questions With Rulin Shao

02:24

SUGARCREPE++ Dataset: Evaluating Vision-Language Model Sensitivity

02:27

Lessons from the Frontlines of Foundation Model Training

03:33

CPU-Only LLM Architectures: Are We There Yet?

02:43

NeurIPS 2024: Impossible GenAI Questions With Ashkan Mirzaei

02:32

Towards Calibrated Robust Fine-Tuning of Vision-Language Models

01:24

UQ-Guided Hyperparameter Optimization for Iterative Learners

01:56

LLM Training Infrastructure: Scaling Up vs. New Architectures

04:17

How is LLM Fine-Tuning Making Big Money?

04:07

NeurIPS 2024: Impossible GenAI Questions With Borna Barahimi

05:25

UniTox: Automating Drug Toxicity Detection

03:27

Memory-Efficient LLM Training and Fine-Tuning via Gradient Subspace Tracking

03:44

Monitoring and Debugging When Training LLMs

03:48

Secret Transformers: Could They Be Reverse-Engineered?

01:54

NeurIPS 2024: Impossible GenAI Questions With Yash Semlani

02:40

ConStat: Performance-Based Contamination Detection in LLMs

02:35

Fine-Grained Complexity of Gradient Computation for Training LLMs

02:49

Optimizing LLM Compute Resources Based on Task Complexity

03:11

How Did DeepSeek R1 Achieve Massive Training Cost Reductions?

04:36

NeurIPS 2024: Impossible GenAI Questions With Xiaowen Zhang

03:40

Deconstructing What Makes a Good Optimizer for Language Models

03:32

Evaluating Language Models as Risk Scores

05:34

Dealing With Infrastructure and GPU Challenges

05:20

NeurIPS 2024: Impossible GenAI Questions With Zirui Wang

02:21

NeurIPS 2024: Impossible GenAI Questions With Tristan Engst

03:13

CoVoMix: Multi-Speaker Dialogue Generation

03:57

Observational Scaling Laws and Predictability of Language Model Performance

04:08

Lessons Learned From LLM Training

02:53

NeurIPS 2024: Impossible GenAI Questions With Pratik Kunapuli

02:07

NeurIPS 2024: Impossible GenAI Questions With Sri Harsha Dumpala

06:07

Attack Atlas: Challenges and Pitfalls in Red Teaming GenAI

05:16

Optimizing Small Language Models for In-Vehicle Function-Calling

02:10

Challenges in Training Biomedical LLMs

03:54

NeurIPS 2024: Impossible GenAI Questions With Lukas Klein

02:35

NeurIPS 2024: Impossible GenAI Questions With Amaury Gouverneur

02:23

JailbreakBench: An Open Robustness Benchmark for Jailbreaking LLMs

30:59

Learnings From Teams Training Large-Scale Models: Monitoring at Hyperscale