Samuel Albanie
21.4K subscribers
12:39
The Agent Company: Benchmarking LLM Agents on Consequential Real World Tasks
Samuel Albanie
2.1K views • 5 months ago
26:00
Prover-Verifier Games improve legibility of LLM outputs
Samuel Albanie
1.4K views • 6 months ago
24:02
Deliberative Alignment: Reasoning Enables Safer Language Models
Samuel Albanie
1.6K views • 6 months ago
24:04
Alignment Faking in Large Language Models
Samuel Albanie
8.7K views • 6 months ago
16:00
RE-Bench: measuring AI agents at AI R&D vs human experts
Samuel Albanie
7.5K views • 6 months ago
5:17
NeurIPS 2024 Poster - On scalable oversight
Samuel Albanie
769 views • 6 months ago
1:20
NeurIPS 2024 Poster - No "Zero-Shot" Without Exponential Data
Samuel Albanie
2.2K views • 6 months ago
3:31
Still a long way to go for Computer Vision? The GRAB Benchmark
Samuel Albanie
3K views • 6 months ago
11:29
Gemini 1.5 Pro has a massive context window
Samuel Albanie
2.8K views • 1 year ago
16:02
Challenges with unsupervised LLM knowledge discovery
Samuel Albanie
1.4K views • 1 year ago
19:35
Anthropic - AI sleeper agents?
Samuel Albanie
2.4K views • 1 year ago
16:01
Mamba - a replacement for Transformers?
Samuel Albanie
257K views • 1 year ago
15:24
How does Gemini compare to GPT-4?
Samuel Albanie
3.1K views • 1 year ago
28:48
Self-supervised vision
Samuel Albanie
7.9K views • 1 year ago
30:49
Vision Transformer Basics
Samuel Albanie
45K views • 1 year ago
25:31
Is Chain of Thought faithful?
Samuel Albanie
2.3K views • 1 year ago
12:49
How strong is Claude 2?
Samuel Albanie
5.3K views • 1 year ago
15:49
What does AI believe is true?
Samuel Albanie
1.9K views • 1 year ago
17:10
Can we verify training data?
Samuel Albanie
1.1K views • 1 year ago
8:54
What is Superalignment?
Samuel Albanie
5.4K views • 1 year ago
8:53
What is SDXL 0.9?
Samuel Albanie
1.6K views • 1 year ago
25:52
Eliciting Latent Knowledge
Samuel Albanie
2.1K views • 1 year ago
14:47
What is KOSMOS-2?
Samuel Albanie
4.6K views • 2 years ago
37:27
Possible catastrophic AI risks?
Samuel Albanie
2.6K views • 2 years ago
13:40
Textbooks Are All You Need
Samuel Albanie
226K views • 2 years ago
12:14
AI Safety & Capabilities News (19th June 2023)
Samuel Albanie
1.2K views • 2 years ago
1:38
Fact-check ChatGPT - Filtir ChatGPT Plugin Demo
Samuel Albanie
1K views • 2 years ago
9:31
AI News (12th June 2023)
Samuel Albanie
678 views • 2 years ago
12:15
AI News (5th June 2023)
Samuel Albanie
476 views • 2 years ago
11:48
AI News (31st May 2023)
Samuel Albanie
401 views • 2 years ago
Load More