Digesting the NanoGPT Code Base by Andrej Karpathy
19:16
Digesting the NanoGPT Code Base by Andrej Karpathy
16 сағат бұрын
[CMU RI] CoFRIDA: Self-Supervised Fine-Tuning for Human-Robot Co-Painting
15:39
[CMU RI] CoFRIDA: Self-Supervised Fine-Tuning for Human-Robot Co-Painting
21 сағат бұрын
[Deepmind]Mobility VLA:Multimodal Instruction Navigation with Long-Context VLMs & Topological Graphs
16:34
[Deepmind]Mobility VLA:Multimodal Instruction Navigation with Long-Context VLMs & Topological Graphs
21 сағат бұрын
AI Evolution: DSPy Transforms Prompting with Self-Improving Pipelines
15:58
AI Evolution: DSPy Transforms Prompting with Self-Improving Pipelines
14 күн бұрын
[LLM 101 Series] (RAG tricks HyDE, CMU) Precise Zero-Shot Dense Retrieval without Relevance Labels
16:53
[LLM 101 Series] (RAG tricks HyDE, CMU) Precise Zero-Shot Dense Retrieval without Relevance Labels
14 күн бұрын
[UW, Allen AI, IBM] Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
14:25
[UW, Allen AI, IBM] Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
14 күн бұрын
[LLM 101 Series] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
15:42
[LLM 101 Series] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
28 күн бұрын
[LLM 101 Series] Neural Machine Translation of Rare Words with Subword Units
16:28
[LLM 101 Series] Neural Machine Translation of Rare Words with Subword Units
28 күн бұрын
[LLM 101 Series] EFFICIENTLY SCALING TRANSFORMER INFERENCE
17:22
[LLM 101 Series] EFFICIENTLY SCALING TRANSFORMER INFERENCE
Ай бұрын
[CVPR 24 Best Paper] Generative Image Dynamics
16:33
[CVPR 24 Best Paper] Generative Image Dynamics
Ай бұрын
[CVPR 24 Best Paper] Rich Human Feedback for Text-to-Image Generation
18:56
[CVPR 24 Best Paper] Rich Human Feedback for Text-to-Image Generation
Ай бұрын
[Meta AI] An Image is Worth More Than 16x16 Patches: Exploring Transformers on Individual Pixels
14:17
[Meta AI] An Image is Worth More Than 16x16 Patches: Exploring Transformers on Individual Pixels
Ай бұрын
[Stanford] Generative Agents: Interactive Simulacra of Human Behavior
16:23
[Stanford] Generative Agents: Interactive Simulacra of Human Behavior
Ай бұрын
[NVIDIA] NVIDIA Nemotron-4 340B Technical Report (Better than GPT-4?)
15:11
[NVIDIA] NVIDIA Nemotron-4 340B Technical Report (Better than GPT-4?)
Ай бұрын
Empowering LLM to use Smartphone for Intelligent Task Automation
15:00
Empowering LLM to use Smartphone for Intelligent Task Automation
Ай бұрын
[Northeastern, MIT, Princeton] Reflexion: Language Agents with Verbal Reinforcement Learning
15:36
[Northeastern, MIT, Princeton] Reflexion: Language Agents with Verbal Reinforcement Learning
Ай бұрын
Can LLM interpret fMRI via embeddings? Crafting Interpretable Embeddings by Asking LLMs Questions
15:17
Can LLM interpret fMRI via embeddings? Crafting Interpretable Embeddings by Asking LLMs Questions
Ай бұрын
[Stanford] Show, Don't Tell: Aligning Language Models with Demonstrated Feedback
16:32
[Stanford] Show, Don't Tell: Aligning Language Models with Demonstrated Feedback
Ай бұрын
[ICRA 2024 Best Paper] NoMaD: Goal Masking Diffusion Policies for Navigation and Exploration
14:14
[ICRA 2024 Best Paper] NoMaD: Goal Masking Diffusion Policies for Navigation and Exploration
Ай бұрын
[Stanford] FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
13:33
[Stanford] FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
Ай бұрын
[Deepmind] Teach LLMs to Phish: Stealing Private Information from Language Models
17:12
[Deepmind] Teach LLMs to Phish: Stealing Private Information from Language Models
Ай бұрын
[Deepmind] Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning
17:43
[Deepmind] Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning
Ай бұрын
[MetaAI] Better & Faster Large Language Models via Multi-token Prediction
11:29
[MetaAI] Better & Faster Large Language Models via Multi-token Prediction
Ай бұрын
[Radboud, U of Amsterdam] Fine Tuning vs. RAG, Which is Better?
10:16
[Radboud, U of Amsterdam] Fine Tuning vs. RAG, Which is Better?
2 ай бұрын
[Stanford, UofT] Observational Scaling Laws and the Predictability of Language Model Performance
9:21
[Stanford, UofT] Observational Scaling Laws and the Predictability of Language Model Performance
2 ай бұрын
[UC Berkeley, UIUC, NYU] Fine-Tuning VLM Agent via RL
11:21
[UC Berkeley, UIUC, NYU] Fine-Tuning VLM Agent via RL
2 ай бұрын
[Google] Demystifying MoE in LLM: What is it?
9:43
[Google] Demystifying MoE in LLM: What is it?
2 ай бұрын
[Nature, Stanford] GenAI for designing structurally novel antibiotics
11:07
[Nature, Stanford] GenAI for designing structurally novel antibiotics
2 ай бұрын
[Nature] Research Shows AI Deciphers Ancient Texts
11:38
[Nature] Research Shows AI Deciphers Ancient Texts
2 ай бұрын
Пікірлер
@Amintbtb
@Amintbtb 5 күн бұрын
Thanks. Would be great if you also share slides
@trendinresearch
@trendinresearch 4 күн бұрын
Hey thank you for your suggestion. I shared it here: drive.google.com/file/d/1d7-phG14p43NiYSqyu9lFWWIRHdl9S4t/view?usp=sharing
@philcui9268
@philcui9268 6 күн бұрын
Is this generated by AI?
@trendinresearch
@trendinresearch 6 күн бұрын
Partially
@uno27heaven
@uno27heaven 18 күн бұрын
Hey man how are u making these videos?
@trendinresearch
@trendinresearch 17 күн бұрын
Thank you for your question. To answer your question, the videos are generated by AI.
@anhtunguyen77
@anhtunguyen77 Ай бұрын
amazing!
@Jshicwhartz
@Jshicwhartz 2 ай бұрын
There is actually a fix for this issue, and Microsoft use it. There is another agent which gets given the most reason message, and you force it to output a function_call saying if it's a jailbreak or not, then you pass this to your main agent and boom. It doesn't work. I won't give all the solutions away, but it's one of many which we use internally. Research stopped getting shared publicly late 2022.
@trendinresearch
@trendinresearch 2 ай бұрын
Thanks, do you have a link of that paper?
@basiclick
@basiclick 2 ай бұрын
wer waiting.. )
@jasonliu1587
@jasonliu1587 2 ай бұрын
good to know
@jasonliu1587
@jasonliu1587 2 ай бұрын
Can I segment svg file?
@trendinsightgpt
@trendinsightgpt 2 ай бұрын
what are some of the energy based models?
@gabrielrainer328
@gabrielrainer328 2 ай бұрын
Very important
@BooleanDisorder
@BooleanDisorder 4 ай бұрын
I definitely think self-reward is the path forward. Like planting a seed that automatically grow with enough resources.
@jamar9205
@jamar9205 6 ай бұрын
👊 *promosm*