Alireza

a guy passionate for R&D, with references from big tech companies and interviewed by tech giants!

[Alireza @Alireza_aca]

@Mahdi4ai خیلی خیلی ممنون عزیز 🫶
همچنین برای خودت دوستم :)

[Alireza @Alireza_aca]

تایملاین #آلمان #برلین
من از اول اکتبر توی آلمان جابجا میشم و میام برلین و دنبال خونه‌ام:
▫️آپارتمان یا WG برای سابلت موقت چند ماهه
▫️یا یه آپارتمان برای اجاره بلندمدت
اگه هر مورد آپارتمانی، wg یا حتی خوابگاه دانشگاهی میشناسین، ممنون میشم بهم اطلاع بدید.

مرسی برای ریتوییت

[Alireza @Alireza_aca]

After 10 months working on my thesis at @MercedesBenz Tech Innovation, the R&D project is wrapped up successfully and our paper is now out.
We are presenting "Speak to the City" this weekend at #ACM #AutoUI26 in Sweden 🇸🇪

Check out our paper: https://t.co/5eYUIj4suo

[Alireza @Alireza_aca]

@MaybeIAmTheDev @dgdgnemetnistm برلینا هیلفه همینه که توی عکس هست؟
یا یه برلینا هیلفه دیکه است اینکه بعضیا میگن؟

[Alireza @Alireza_aca]

@Mahdi4ai دمت گرم، ممنونم از توییتت 🫶🙌

[Alireza @Alireza_aca]

Study about #LLMs

Then expand your knowledge with Vision Transformers and you can easily learn #VLMs

Then read about Discretization or Continuous Flow, and you can learn #VLAs

#transformer

[Neo Kim @systemdesignone]

If I had to start with LLM from scratch, I'd learn these 30 concepts:

1 LLM
2 Token
3 Tokenization
4 Embeddings
5 Latent Space
6 Parameters
7 Pre-training
8 Base Model
9 Instruct Model
10 Fine-Tuning
11 Alignment
12 RLHF
13 Prompt
14 System Prompt
15 User Prompt
16 Context Window
17 Zero-Shot Learning
18 Few-Shot Learning
19 Chain-of-Thought
20 Inference
21 Latency
22 Temperature
23 Hallucination
24 Grounding
25 RAG
26 Workflow
27 Agent
28 Multimodality
29 Benchmarks
30 Guardrails

What else should make this list?

[Turing Post (Ksenia Se) @TheTuringPost]

15+ LoRA (Low-Rank Adaptation) variants you should know

▪️ Original LoRA
▪️ QLoRA
▪️ DoRA
▪️ QDoRA
▪️ rsLoRA (Rank-Stabilized)
▪️ VeRA (Vector-based Random Adaptation)
▪️ SingLoRA (Single-Matrix LoRA)
▪️ Sensitivity-LoRA
▪️ ARD-LoRA (Adaptive Rank Dynamic)
▪️ Mixture-of-LoRA-Experts
▪️ X-LoRA
▪️ AutoLoRA
▪️ LAG (LoRA-Augmented Generation)
▪️ T-LoRA (Timestep-Dependent)
▪️ Text-to-LoRA
▪️ Doc-to-LoRA
▪️ LoRA-Squeeze
▪️ Mixture of Adapters (MoA)

Save the list and check this out for useful sources to explore them more: https://t.co/ZwtpB9nrc3

[Stable Diffusion Tutorials @SD_Tutorial]

LTX2.3 🖌️🎨Fantasy_Painterly_Style_LoRa

#comfyui #stablediffusion

This LoRA creates a fantasy painterly cinematic style where every subject and environment appears rendered as richly detailed painted imagery

Workflow + repo:👇👇
https://t.co/0OYbGnDGPS

[AK @_akhaliq]

PersonaVLM

Long-Term Personalized Multimodal LLMs

paper: https://t.co/X9aI6covnr

[Alireza @Alireza_aca]

@X @X @nikitabier
Why there is no "unsend" or "request revoke" option on X messages? That's indeed necessary for messaging features!!!