deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
Text Generation
โข
Updated
โข
166k
โข
โข
706
Ruslan
uzvisa
AI & ML interests
None yet
Recent Activity
liked
a model
2 days ago
deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
reacted
to
seawolf2357's
post
with ๐
3 days ago
๐ Just Found an Interesting New Leaderboard for Medical AI Evaluation!
I recently stumbled upon a medical domain-specific FACTS Grounding leaderboard on Hugging Face, and the approach to evaluating AI accuracy in medical contexts is quite impressive, so I thought I'd share.
๐ What is FACTS Grounding?
It's originally a benchmark developed by Google DeepMind that measures how well LLMs generate answers based solely on provided documents. What's cool about this medical-focused version is that it's designed to test even small open-source models.
๐ฅ Medical Domain Version Features
236 medical examples: Extracted from the original 860 examples
Tests small models like Qwen 3 1.7B: Great for resource-constrained environments
Uses Gemini 1.5 Flash for evaluation: Simplified to a single judge model
๐ The Evaluation Method is Pretty Neat
Grounding Score: Are all claims in the response supported by the provided document?
Quality Score: Does it properly answer the user's question?
Combined Score: Did it pass both checks?
Since medical information requires extreme accuracy, this thorough verification approach makes a lot of sense.
๐ Check It Out Yourself
The actual leaderboard: https://huggingface.co/spaces/MaziyarPanahi/FACTS-Leaderboard
๐ญ My thoughts: As medical AI continues to evolve, evaluation tools like this are becoming increasingly important. The fact that it can test smaller models is particularly helpful for the open-source community!
Organizations
None yet
uzvisa's activity
liked
a
model
2 days ago
reacted to
seawolf2357's
post with ๐
3 days ago
Post
1222
๐ Just Found an Interesting New Leaderboard for Medical AI Evaluation!
I recently stumbled upon a medical domain-specific FACTS Grounding leaderboard on Hugging Face, and the approach to evaluating AI accuracy in medical contexts is quite impressive, so I thought I'd share.
๐ What is FACTS Grounding?
It's originally a benchmark developed by Google DeepMind that measures how well LLMs generate answers based solely on provided documents. What's cool about this medical-focused version is that it's designed to test even small open-source models.
๐ฅ Medical Domain Version Features
236 medical examples: Extracted from the original 860 examples
Tests small models like Qwen 3 1.7B: Great for resource-constrained environments
Uses Gemini 1.5 Flash for evaluation: Simplified to a single judge model
๐ The Evaluation Method is Pretty Neat
Grounding Score: Are all claims in the response supported by the provided document?
Quality Score: Does it properly answer the user's question?
Combined Score: Did it pass both checks?
Since medical information requires extreme accuracy, this thorough verification approach makes a lot of sense.
๐ Check It Out Yourself
The actual leaderboard: MaziyarPanahi/FACTS-Leaderboard
๐ญ My thoughts: As medical AI continues to evolve, evaluation tools like this are becoming increasingly important. The fact that it can test smaller models is particularly helpful for the open-source community!
I recently stumbled upon a medical domain-specific FACTS Grounding leaderboard on Hugging Face, and the approach to evaluating AI accuracy in medical contexts is quite impressive, so I thought I'd share.
๐ What is FACTS Grounding?
It's originally a benchmark developed by Google DeepMind that measures how well LLMs generate answers based solely on provided documents. What's cool about this medical-focused version is that it's designed to test even small open-source models.
๐ฅ Medical Domain Version Features
236 medical examples: Extracted from the original 860 examples
Tests small models like Qwen 3 1.7B: Great for resource-constrained environments
Uses Gemini 1.5 Flash for evaluation: Simplified to a single judge model
๐ The Evaluation Method is Pretty Neat
Grounding Score: Are all claims in the response supported by the provided document?
Quality Score: Does it properly answer the user's question?
Combined Score: Did it pass both checks?
Since medical information requires extreme accuracy, this thorough verification approach makes a lot of sense.
๐ Check It Out Yourself
The actual leaderboard: MaziyarPanahi/FACTS-Leaderboard
๐ญ My thoughts: As medical AI continues to evolve, evaluation tools like this are becoming increasingly important. The fact that it can test smaller models is particularly helpful for the open-source community!
reacted to
openfree's
post with ๐ฅ
10 days ago
Post
2570
๐ง AI Brand Naming with 15 Specialized Theories
๐ฏ Core Features
15 Expert Theories for professional brand naming
Bilingual Support Korean/English for global brands
Unified Evaluation System creativity/memorability/relevance scores
Real-time Visualization theory-specific custom designs
openfree/Naming
๐ฌ Applied Theories
Cognitive Theories (4)
๐ฆ Square Theory - Semantic square structure with 4-word relationships
๐ Sound Symbolism - Psychological connections between phonemes and meaning
๐ง Cognitive Load - Minimized processing for instant recognition
๐๏ธ Gestalt Theory - Perceptual principles where whole exceeds parts
Creative Theories (3)
๐ Conceptual Blending - Merging concepts to create new meanings
๐ง SCAMPER Method - 7 creative transformation techniques
๐ฟ Biomimicry - Nature-inspired wisdom from 3.8 billion years of evolution
Strategic Theories (2)
โ Jobs-to-be-Done - Customer-centric problem-solving focus
๐ญ Design Thinking - Human-centered innovation methodology
Cultural Theories (3)
๐ญ Jung's Archetype - 12 universal archetypes for emotional connection
๐ Linguistic Relativity - Cross-cultural thinking patterns consideration
๐งฌ Memetics - Cultural transmission and evolutionary potential
Differentiation Theories (3)
โก Von Restorff Effect - Uniqueness for 30x better recall
๐จ Color Psychology - Emotional associations and color meanings
๐ Network Effects - Value maximization through network structures
๐ซ Special Features
Each theory provides unique visualizations and customized analysis:
Square Theory โ 4-corner relationship diagram
Blending โ Concept fusion flowchart
Color โ Interactive color palette display
Theory-specific insights for each approach
๐จ Output Information
Core: Brand name, slogan, values, emotions, personality
Visual: Colors, concepts, typography styles
Linguistic: Pronunciation, etymology, global adaptability
Strategic: Differentiation, positioning, growth potential
Theory-specific...
๐ฏ Core Features
15 Expert Theories for professional brand naming
Bilingual Support Korean/English for global brands
Unified Evaluation System creativity/memorability/relevance scores
Real-time Visualization theory-specific custom designs
openfree/Naming
๐ฌ Applied Theories
Cognitive Theories (4)
๐ฆ Square Theory - Semantic square structure with 4-word relationships
๐ Sound Symbolism - Psychological connections between phonemes and meaning
๐ง Cognitive Load - Minimized processing for instant recognition
๐๏ธ Gestalt Theory - Perceptual principles where whole exceeds parts
Creative Theories (3)
๐ Conceptual Blending - Merging concepts to create new meanings
๐ง SCAMPER Method - 7 creative transformation techniques
๐ฟ Biomimicry - Nature-inspired wisdom from 3.8 billion years of evolution
Strategic Theories (2)
โ Jobs-to-be-Done - Customer-centric problem-solving focus
๐ญ Design Thinking - Human-centered innovation methodology
Cultural Theories (3)
๐ญ Jung's Archetype - 12 universal archetypes for emotional connection
๐ Linguistic Relativity - Cross-cultural thinking patterns consideration
๐งฌ Memetics - Cultural transmission and evolutionary potential
Differentiation Theories (3)
โก Von Restorff Effect - Uniqueness for 30x better recall
๐จ Color Psychology - Emotional associations and color meanings
๐ Network Effects - Value maximization through network structures
๐ซ Special Features
Each theory provides unique visualizations and customized analysis:
Square Theory โ 4-corner relationship diagram
Blending โ Concept fusion flowchart
Color โ Interactive color palette display
Theory-specific insights for each approach
๐จ Output Information
Core: Brand name, slogan, values, emotions, personality
Visual: Colors, concepts, typography styles
Linguistic: Pronunciation, etymology, global adaptability
Strategic: Differentiation, positioning, growth potential
Theory-specific...
liked
a
model
23 days ago
ะัะปะธัะฝะฐั ะผะพะดะตะปั!
#2 opened 25 days ago
by
uzvisa
liked
a
model
25 days ago
liked
2
models
27 days ago
Improved Jinja Chat Template for Apriel-Nemotron-15b-Thinker GGUF (e.g., for LM Studio)
๐
1
10
#1 opened about 1 month ago
by
debeast6
liked
a
model
27 days ago
liked
a
model
about 1 month ago
Iโm excited to hear any updates from the DeepCogito Team!
1
#1 opened about 1 month ago
by
uzvisa
reacted to
nyuuzyou's
post with ๐
about 1 month ago
Post
2223
๐ผ๏ธ OpenClipart SVG Dataset -
nyuuzyou/openclipart
Collection of 178,604 Public Domain Scalable Vector Graphics (SVG) clipart images featuring:
- Comprehensive metadata: title, description, artist name, tags, original page URL, and more.
- Contains complete SVG XML content (minified) for direct use or processing.
- All images explicitly released into the public domain under the CC0 license.
- Organized in a single
Collection of 178,604 Public Domain Scalable Vector Graphics (SVG) clipart images featuring:
- Comprehensive metadata: title, description, artist name, tags, original page URL, and more.
- Contains complete SVG XML content (minified) for direct use or processing.
- All images explicitly released into the public domain under the CC0 license.
- Organized in a single
train split with 178,604 entries.
reacted to
merterbak's
post with ๐ฅ
about 1 month ago
Post
4866
Qwen 3 models released๐ฅ
It offers 2 MoE and 6 dense models with following parameter sizes: 0.6B, 1.7B, 4B, 8B, 14B, 30B(MoE), 32B, and 235B(MoE).
Models: Qwen/qwen3-67dd247413f0e2e4f653967f
Blog: https://qwenlm.github.io/blog/qwen3/
Demo: Qwen/Qwen3-Demo
GitHub: https://github.com/QwenLM/Qwen3
โ Pre-trained 119 languages(36 trillion tokens) and dialects with strong translation and instruction following abilities. (Qwen2.5 was pre-trained on 18 trillion tokens.)
โ Qwen3 dense models match the performance of larger Qwen2.5 models. For example, Qwen3-1.7B/4B/8B/14B/32B perform like Qwen2.5-3B/7B/14B/32B/72B.
โ Three stage done while pretraining:
โข Stage 1: General language learning and knowledge building.
โข Stage 2: Reasoning boost with STEM, coding, and logic skills.
โข Stage 3: Long context training
โ It supports MCP in the model
โ Strong agent skills
โ Supports seamless between thinking mode (for hard tasks like math and coding) and non-thinking mode (for fast chatting) inside chat template.
โ Better human alignment for creative writing, roleplay, multi-turn conversations, and following detailed instructions.
It offers 2 MoE and 6 dense models with following parameter sizes: 0.6B, 1.7B, 4B, 8B, 14B, 30B(MoE), 32B, and 235B(MoE).
Models: Qwen/qwen3-67dd247413f0e2e4f653967f
Blog: https://qwenlm.github.io/blog/qwen3/
Demo: Qwen/Qwen3-Demo
GitHub: https://github.com/QwenLM/Qwen3
โ Pre-trained 119 languages(36 trillion tokens) and dialects with strong translation and instruction following abilities. (Qwen2.5 was pre-trained on 18 trillion tokens.)
โ Qwen3 dense models match the performance of larger Qwen2.5 models. For example, Qwen3-1.7B/4B/8B/14B/32B perform like Qwen2.5-3B/7B/14B/32B/72B.
โ Three stage done while pretraining:
โข Stage 1: General language learning and knowledge building.
โข Stage 2: Reasoning boost with STEM, coding, and logic skills.
โข Stage 3: Long context training
โ It supports MCP in the model
โ Strong agent skills
โ Supports seamless between thinking mode (for hard tasks like math and coding) and non-thinking mode (for fast chatting) inside chat template.
โ Better human alignment for creative writing, roleplay, multi-turn conversations, and following detailed instructions.
reacted to
Kseniase's
post with ๐
about 2 months ago
Post
7244
11 new types of RAG
RAG is evolving fast, keeping pace with cutting-edge AI trends. Today it becomes more agentic and smarter at navigating complex structures like hypergraphs.
Here are 11 latest RAG types:
1. InstructRAG -> InstructRAG: Leveraging Retrieval-Augmented Generation on Instruction Graphs for LLM-Based Task Planning (2504.13032)
Combines RAG with a multi-agent framework, using a graph-based structure, an RL agent to expand task coverage, and a meta-learning agent for better generalization
2. CoRAG (Collaborative RAG) -> CoRAG: Collaborative Retrieval-Augmented Generation (2504.01883)
A collaborative framework that extends RAG to settings where clients train a shared model using a joint passage store
3. ReaRAG -> ReaRAG: Knowledge-guided Reasoning Enhances Factuality of Large Reasoning Models with Iterative Retrieval Augmented Generation (2503.21729)
It uses a Thought-Action-Observation loop to decide at each step whether to retrieve information or finalize an answer, reducing unnecessary reasoning and errors
4. MCTS-RAG -> MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search (2503.20757)
Combines RAG with Monte Carlo Tree Search (MCTS) to help small LMs handle complex, knowledge-heavy tasks
5. Typed-RAG - > Typed-RAG: Type-aware Multi-Aspect Decomposition for Non-Factoid Question Answering (2503.15879)
Improves answers on open-ended questions by identifying question types (a debate, personal experience, or comparison) and breaking it down into simpler parts
6. MADAM-RAG -> Retrieval-Augmented Generation with Conflicting Evidence (2504.13079)
A multi-agent system where models debate answers over multiple rounds and an aggregator filters noise and misinformation
7. HM-RAG -> HM-RAG: Hierarchical Multi-Agent Multimodal Retrieval Augmented Generation (2504.12330)
A hierarchical multi-agent RAG framework that uses 3 agents: one to split queries, one to retrieve across multiple data types (text, graphs and web), and one to merge and refine answers
8. CDF-RAG -> CDF-RAG: Causal Dynamic Feedback for Adaptive Retrieval-Augmented Generation (2504.12560)
Works with causal graphs and enables multi-hop causal reasoning, refining queries. It validates responses against causal pathways
To explore what is Causal AI, read our article: https://www.turingpost.com/p/causalai
Subscribe to the Turing Post: https://www.turingpost.com/subscribe
Read further ๐
RAG is evolving fast, keeping pace with cutting-edge AI trends. Today it becomes more agentic and smarter at navigating complex structures like hypergraphs.
Here are 11 latest RAG types:
1. InstructRAG -> InstructRAG: Leveraging Retrieval-Augmented Generation on Instruction Graphs for LLM-Based Task Planning (2504.13032)
Combines RAG with a multi-agent framework, using a graph-based structure, an RL agent to expand task coverage, and a meta-learning agent for better generalization
2. CoRAG (Collaborative RAG) -> CoRAG: Collaborative Retrieval-Augmented Generation (2504.01883)
A collaborative framework that extends RAG to settings where clients train a shared model using a joint passage store
3. ReaRAG -> ReaRAG: Knowledge-guided Reasoning Enhances Factuality of Large Reasoning Models with Iterative Retrieval Augmented Generation (2503.21729)
It uses a Thought-Action-Observation loop to decide at each step whether to retrieve information or finalize an answer, reducing unnecessary reasoning and errors
4. MCTS-RAG -> MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search (2503.20757)
Combines RAG with Monte Carlo Tree Search (MCTS) to help small LMs handle complex, knowledge-heavy tasks
5. Typed-RAG - > Typed-RAG: Type-aware Multi-Aspect Decomposition for Non-Factoid Question Answering (2503.15879)
Improves answers on open-ended questions by identifying question types (a debate, personal experience, or comparison) and breaking it down into simpler parts
6. MADAM-RAG -> Retrieval-Augmented Generation with Conflicting Evidence (2504.13079)
A multi-agent system where models debate answers over multiple rounds and an aggregator filters noise and misinformation
7. HM-RAG -> HM-RAG: Hierarchical Multi-Agent Multimodal Retrieval Augmented Generation (2504.12330)
A hierarchical multi-agent RAG framework that uses 3 agents: one to split queries, one to retrieve across multiple data types (text, graphs and web), and one to merge and refine answers
8. CDF-RAG -> CDF-RAG: Causal Dynamic Feedback for Adaptive Retrieval-Augmented Generation (2504.12560)
Works with causal graphs and enables multi-hop causal reasoning, refining queries. It validates responses against causal pathways
To explore what is Causal AI, read our article: https://www.turingpost.com/p/causalai
Subscribe to the Turing Post: https://www.turingpost.com/subscribe
Read further ๐
- 1 reply
ะะฐะบะธะต ะดะฐะฝะฝัะต ะฒ ััะพะน LLM
#1 opened about 2 months ago
by
uzvisa
liked
a
model
3 months ago
New activity in
RichardErkhov/micks99_-_gemma-2b-instruct-ft-Data-Analytics-Digital-Marketing-Project-Management-QAv2-gguf
3 months ago
ะะปั ัะตะณะพ ะฝัะถะฝะฐ ััะฐ ะผะพะดะตะปั?
8
#1 opened 3 months ago
by
uzvisa
liked
a
model
3 months ago
New activity in
RichardErkhov/micks99_-_gemma-2b-instruct-ft-Data-Analytics-Digital-Marketing-Project-Management-QAv2-gguf
3 months ago
ะะปั ัะตะณะพ ะฝัะถะฝะฐ ััะฐ ะผะพะดะตะปั?
8
#1 opened 3 months ago
by
uzvisa