Semantic-guided LoRA (SG-LoRA) → https://huggingface.co/papers/2509.10535
Generates task-specific LoRA parameters from semantic preferences, enabling zero-shot and privacy-preserving adaptation to new tasksPHLoRA (Post-hoc LoRA) → https://huggingface.co/papers/2509.10971
Extracts LoRA adapters after fine-rank fine-tuning by low-rank factoring the weight differencesLoRA-Gen → https://huggingface.co/papers/2506.11638
Generates LoRA parameters from a large cloud model for small edge models, merging them for efficient task specialization and faster inference.DP-FedLoRA → https://huggingface.co/papers/2509.09097
Adds DP noise to LoRA matrices in federated on-device fine-tuning for privacy preservation
Ksenia Se
Kseniase
AI & ML interests
None yet
Recent Activity
replied to
their
post
6 days ago
10 awesome advanced LoRA approaches
Low-Rank Adaptation (LoRA) is the go-to method for efficient model fine-tuning that adds small low-rank matrices instead of retraining full models. The field isn’t standing still – new LoRA variants push the limits of efficiency, generalization, and personalization. So we’re sharing 10 of the latest LoRA approaches you should know about:
1. Mixture-of-LoRA-experts → https://huggingface.co/papers/2509.13878
Adds multiple low-rank adapters (LoRA) into a model’s layers, and a routing mechanism activates the most suitable ones for each input. This lets the model adapt better to new unseen conditions
2. Amortized Bayesian Meta-Learning for LoRA (ABMLL) → https://huggingface.co/papers/2508.14285
Balances global and task-specific parameters within a Bayesian framework to improve uncertainty calibration and generalization to new tasks without high memory or compute costs
3. AutoLoRA → https://huggingface.co/papers/2508.02107
Automatically retrieves and dynamically aggregates public LoRAs for stronger T2I generation
4. aLoRA (Activated LoRA) → https://huggingface.co/papers/2504.12397
Only applies LoRA after invocation, letting the model reuse the base model’s KV cache instead of recomputing the full turn’s KV cache. Efficient in multi-turn conversations
5. LiLoRA (LoRA in LoRA) → https://huggingface.co/papers/2508.06202
Shares the LoRA matrix A across tasks and additionally low-rank-decomposes matrix B to cut parameters in continual vision-text MLLMs
6. Sensitivity-LoRA → https://huggingface.co/papers/2509.09119
Dynamically assigns ranks to weight matrices based on their sensitivity, measured using second-order derivatives
Read further below ↓
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
posted
an
update
6 days ago
10 awesome advanced LoRA approaches
Low-Rank Adaptation (LoRA) is the go-to method for efficient model fine-tuning that adds small low-rank matrices instead of retraining full models. The field isn’t standing still – new LoRA variants push the limits of efficiency, generalization, and personalization. So we’re sharing 10 of the latest LoRA approaches you should know about:
1. Mixture-of-LoRA-experts → https://huggingface.co/papers/2509.13878
Adds multiple low-rank adapters (LoRA) into a model’s layers, and a routing mechanism activates the most suitable ones for each input. This lets the model adapt better to new unseen conditions
2. Amortized Bayesian Meta-Learning for LoRA (ABMLL) → https://huggingface.co/papers/2508.14285
Balances global and task-specific parameters within a Bayesian framework to improve uncertainty calibration and generalization to new tasks without high memory or compute costs
3. AutoLoRA → https://huggingface.co/papers/2508.02107
Automatically retrieves and dynamically aggregates public LoRAs for stronger T2I generation
4. aLoRA (Activated LoRA) → https://huggingface.co/papers/2504.12397
Only applies LoRA after invocation, letting the model reuse the base model’s KV cache instead of recomputing the full turn’s KV cache. Efficient in multi-turn conversations
5. LiLoRA (LoRA in LoRA) → https://huggingface.co/papers/2508.06202
Shares the LoRA matrix A across tasks and additionally low-rank-decomposes matrix B to cut parameters in continual vision-text MLLMs
6. Sensitivity-LoRA → https://huggingface.co/papers/2509.09119
Dynamically assigns ranks to weight matrices based on their sensitivity, measured using second-order derivatives
Read further below ↓
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
replied to
their
post
13 days ago
6 Recent & free sources to master Reinforcement Learning
Almost every week new research and resources on RL come out. Knowledge needs to be constantly refreshed and updated with the latest trends. So today, we’re sharing 6 free sources to help you stay on track with RL:
1. A Survey of Continual Reinforcement Learning → https://arxiv.org/abs/2506.21872
Covers continual RL (CRL): how agents can keep learning and adapt to new tasks without forgetting past ones. It analyses methods, benchmarks, evaluation metrics &challenges
2. The Deep Reinforcement Learning course by Hugging Face → https://huggingface.co/learn/deep-rl-course/unit0/introduction
This is a popular free course, regularly updated. Includes community interaction, exercises, leaderboards, etc.
3. Reinforcement Learning Specialization (Coursera, University of Alberta) → https://www.coursera.org/specializations/reinforcement-learning
A 4-course series introducing foundational RL, implementing different algorithms, culminating in a capstone. It's a great structured path
4. A Technical Survey of Reinforcement Learning Techniques for LLMs → https://huggingface.co/papers/2507.04136
Looks at how RL is being used for/with LLMs for alignment, reasoning, preference signals, etc. Covers methods like RLHF, RLAIF, DPO, PPO, GRPO & applications from code gen to tool use
5. A Survey of Reinforcement Learning for Software Engineering → https://arxiv.org/abs/2507.12483
Good if you're interested in RL-applied domains. Examines how RL is used in software engineering tasks: maintenance, development, evaluation. Covering 115 papers since DRL introduction, it summarizes trends, gaps & challenges
6. A Survey of Reinforcement Learning for LRMs → https://arxiv.org/abs/2509.08827
Tracks the way from LLMs to LRMs via RL. Covers reward design, policy optimization, use cases and future approaches like continual, memory, model-based RL and more
If you liked this, subscribe to The Turing Post https://www.turingpost.com/subscribe
Organizations
replied to
their
post
6 days ago
posted
an
update
6 days ago
Post
6007
10 awesome advanced LoRA approaches
Low-Rank Adaptation (LoRA) is the go-to method for efficient model fine-tuning that adds small low-rank matrices instead of retraining full models. The field isn’t standing still – new LoRA variants push the limits of efficiency, generalization, and personalization. So we’re sharing 10 of the latest LoRA approaches you should know about:
1. Mixture-of-LoRA-experts → Mixture of Low-Rank Adapter Experts in Generalizable Audio Deepfake Detection (2509.13878)
Adds multiple low-rank adapters (LoRA) into a model’s layers, and a routing mechanism activates the most suitable ones for each input. This lets the model adapt better to new unseen conditions
2. Amortized Bayesian Meta-Learning for LoRA (ABMLL) → Amortized Bayesian Meta-Learning for Low-Rank Adaptation of Large Language Models (2508.14285)
Balances global and task-specific parameters within a Bayesian framework to improve uncertainty calibration and generalization to new tasks without high memory or compute costs
3. AutoLoRA → AutoLoRA: Automatic LoRA Retrieval and Fine-Grained Gated Fusion for Text-to-Image Generation (2508.02107)
Automatically retrieves and dynamically aggregates public LoRAs for stronger T2I generation
4. aLoRA (Activated LoRA) → Activated LoRA: Fine-tuned LLMs for Intrinsics (2504.12397)
Only applies LoRA after invocation, letting the model reuse the base model’s KV cache instead of recomputing the full turn’s KV cache. Efficient in multi-turn conversations
5. LiLoRA (LoRA in LoRA) → LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning (2508.06202)
Shares the LoRA matrix A across tasks and additionally low-rank-decomposes matrix B to cut parameters in continual vision-text MLLMs
6. Sensitivity-LoRA → Sensitivity-LoRA: Low-Load Sensitivity-Based Fine-Tuning for Large Language Models (2509.09119)
Dynamically assigns ranks to weight matrices based on their sensitivity, measured using second-order derivatives
Read further below ↓
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
Low-Rank Adaptation (LoRA) is the go-to method for efficient model fine-tuning that adds small low-rank matrices instead of retraining full models. The field isn’t standing still – new LoRA variants push the limits of efficiency, generalization, and personalization. So we’re sharing 10 of the latest LoRA approaches you should know about:
1. Mixture-of-LoRA-experts → Mixture of Low-Rank Adapter Experts in Generalizable Audio Deepfake Detection (2509.13878)
Adds multiple low-rank adapters (LoRA) into a model’s layers, and a routing mechanism activates the most suitable ones for each input. This lets the model adapt better to new unseen conditions
2. Amortized Bayesian Meta-Learning for LoRA (ABMLL) → Amortized Bayesian Meta-Learning for Low-Rank Adaptation of Large Language Models (2508.14285)
Balances global and task-specific parameters within a Bayesian framework to improve uncertainty calibration and generalization to new tasks without high memory or compute costs
3. AutoLoRA → AutoLoRA: Automatic LoRA Retrieval and Fine-Grained Gated Fusion for Text-to-Image Generation (2508.02107)
Automatically retrieves and dynamically aggregates public LoRAs for stronger T2I generation
4. aLoRA (Activated LoRA) → Activated LoRA: Fine-tuned LLMs for Intrinsics (2504.12397)
Only applies LoRA after invocation, letting the model reuse the base model’s KV cache instead of recomputing the full turn’s KV cache. Efficient in multi-turn conversations
5. LiLoRA (LoRA in LoRA) → LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning (2508.06202)
Shares the LoRA matrix A across tasks and additionally low-rank-decomposes matrix B to cut parameters in continual vision-text MLLMs
6. Sensitivity-LoRA → Sensitivity-LoRA: Low-Load Sensitivity-Based Fine-Tuning for Large Language Models (2509.09119)
Dynamically assigns ranks to weight matrices based on their sensitivity, measured using second-order derivatives
Read further below ↓
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
-
- 3 replies
replied to
their
post
13 days ago
Also, don't forget about RL classics - find more resources here: https://huggingface.co/posts/Kseniase/484268922176188
posted
an
update
13 days ago
Post
6274
6 Recent & free sources to master Reinforcement Learning
Almost every week new research and resources on RL come out. Knowledge needs to be constantly refreshed and updated with the latest trends. So today, we’re sharing 6 free sources to help you stay on track with RL:
1. A Survey of Continual Reinforcement Learning → https://arxiv.org/abs/2506.21872
Covers continual RL (CRL): how agents can keep learning and adapt to new tasks without forgetting past ones. It analyses methods, benchmarks, evaluation metrics &challenges
2. The Deep Reinforcement Learning course by Hugging Face → https://huggingface.co/learn/deep-rl-course/unit0/introduction
This is a popular free course, regularly updated. Includes community interaction, exercises, leaderboards, etc.
3. Reinforcement Learning Specialization (Coursera, University of Alberta) → https://www.coursera.org/specializations/reinforcement-learning
A 4-course series introducing foundational RL, implementing different algorithms, culminating in a capstone. It's a great structured path
4. A Technical Survey of Reinforcement Learning Techniques for LLMs → A Technical Survey of Reinforcement Learning Techniques for Large Language Models (2507.04136)
Looks at how RL is being used for/with LLMs for alignment, reasoning, preference signals, etc. Covers methods like RLHF, RLAIF, DPO, PPO, GRPO & applications from code gen to tool use
5. A Survey of Reinforcement Learning for Software Engineering → https://arxiv.org/abs/2507.12483
Good if you're interested in RL-applied domains. Examines how RL is used in software engineering tasks: maintenance, development, evaluation. Covering 115 papers since DRL introduction, it summarizes trends, gaps & challenges
6. A Survey of Reinforcement Learning for LRMs → https://arxiv.org/abs/2509.08827
Tracks the way from LLMs to LRMs via RL. Covers reward design, policy optimization, use cases and future approaches like continual, memory, model-based RL and more
If you liked this, subscribe to The Turing Post https://www.turingpost.com/subscribe
Almost every week new research and resources on RL come out. Knowledge needs to be constantly refreshed and updated with the latest trends. So today, we’re sharing 6 free sources to help you stay on track with RL:
1. A Survey of Continual Reinforcement Learning → https://arxiv.org/abs/2506.21872
Covers continual RL (CRL): how agents can keep learning and adapt to new tasks without forgetting past ones. It analyses methods, benchmarks, evaluation metrics &challenges
2. The Deep Reinforcement Learning course by Hugging Face → https://huggingface.co/learn/deep-rl-course/unit0/introduction
This is a popular free course, regularly updated. Includes community interaction, exercises, leaderboards, etc.
3. Reinforcement Learning Specialization (Coursera, University of Alberta) → https://www.coursera.org/specializations/reinforcement-learning
A 4-course series introducing foundational RL, implementing different algorithms, culminating in a capstone. It's a great structured path
4. A Technical Survey of Reinforcement Learning Techniques for LLMs → A Technical Survey of Reinforcement Learning Techniques for Large Language Models (2507.04136)
Looks at how RL is being used for/with LLMs for alignment, reasoning, preference signals, etc. Covers methods like RLHF, RLAIF, DPO, PPO, GRPO & applications from code gen to tool use
5. A Survey of Reinforcement Learning for Software Engineering → https://arxiv.org/abs/2507.12483
Good if you're interested in RL-applied domains. Examines how RL is used in software engineering tasks: maintenance, development, evaluation. Covering 115 papers since DRL introduction, it summarizes trends, gaps & challenges
6. A Survey of Reinforcement Learning for LRMs → https://arxiv.org/abs/2509.08827
Tracks the way from LLMs to LRMs via RL. Covers reward design, policy optimization, use cases and future approaches like continual, memory, model-based RL and more
If you liked this, subscribe to The Turing Post https://www.turingpost.com/subscribe
-
- 1 reply
replied to
their
post
20 days ago
TreePO → https://huggingface.co/papers/2508.17445
Treats sequence generation as a tree search, branching into multiple reasoning paths. It samples, reuses prefixes, and prunes low-value paths, saving computation and enhancing training efficiencyDuPO → https://huggingface.co/papers/2508.14460
Creates a dual task: taking the model’s output and trying to reconstruct hidden or missing parts of the original input. This reconstruction quality serves as a self-supervised reward, helping the model learn tasks like translation and math reasoningTempFlow-GRPO → https://huggingface.co/papers/2508.04324
Leverages the temporal structure of flow-based text-to-image generation. It uses trajectory branching to assign rewards at key decision points and a noise-aware weighting scheme to focus learning on the most impactful timestepsMixGRPO → https://huggingface.co/papers/2507.21802
Combines stochastic (SDE) and deterministic (ODE) sampling to make training more efficient. It uses a sliding window to apply GRPO optimization only where it matters, cutting computation and training time up to 71%MaPPO (Maximum a Posteriori Preference Optimization) → https://huggingface.co/papers/2507.21183
Improves on DPO by adding prior reward knowledge into the training objective. It treats alignment as a Maximum a Posteriori (MAP) problem, leading to more accurate preference learning without extra hyperparameters
posted
an
update
20 days ago
Post
6967
10 Latest Preference Optimization Techniques
Models need feedback on what makes outputs “good” or “bad.” Policy optimization (PO) turns preferences and rewards into actual training signals. This field is evolving quickly, moving far beyond classics like PPO and GRPO. So here is our overview of 10 newest PO methods:
1. Pref-GRPO → Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning (2508.20751)
Stabilizes text-to-image reinforcement learning (RL) with pairwise preference rewards and a unified UNIGENBENCH benchmark
2. PVPO (Policy with Value Preference Optimization) → PVPO: Pre-Estimated Value-Based Policy Optimization for Agentic Reasoning (2508.21104)
This critic-free RL method uses a pre-trained model as a reference anchor to reduce bias and guide learning, selecting high-value examples through data pre-sampling
3. DCPO (Dynamic Clipping Policy Optimization) → DCPO: Dynamic Clipping Policy Optimization (2509.02333)
Uses dynamic clipping, which adjusts probability limits per token for better token exploration, and smooth reward standardization to balance rewards over training steps and prevent wasted updates
4. ARPO (Agentic Reinforced Policy Optimization) → Agentic Reinforced Policy Optimization (2507.19849)
Optimizes multi-turn LLM agents that use external tools. It uses an entropy-based adaptive rollout to explore post-tool use and an advantage attribution method to better assign credit across steps, leading to more efficient tool use with fewer resources
5. GRPO-RoC (Group Relative Policy Optimization with Resampling-on-Correct) → rStar2-Agent: Agentic Reasoning Technical Report (2508.20722)
Oversamples rollouts, then resamples them to keep diverse mistakes and only the highest-quality correct answers. It reduces noises and ends up with stronger reasoning in a code environment
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
Models need feedback on what makes outputs “good” or “bad.” Policy optimization (PO) turns preferences and rewards into actual training signals. This field is evolving quickly, moving far beyond classics like PPO and GRPO. So here is our overview of 10 newest PO methods:
1. Pref-GRPO → Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning (2508.20751)
Stabilizes text-to-image reinforcement learning (RL) with pairwise preference rewards and a unified UNIGENBENCH benchmark
2. PVPO (Policy with Value Preference Optimization) → PVPO: Pre-Estimated Value-Based Policy Optimization for Agentic Reasoning (2508.21104)
This critic-free RL method uses a pre-trained model as a reference anchor to reduce bias and guide learning, selecting high-value examples through data pre-sampling
3. DCPO (Dynamic Clipping Policy Optimization) → DCPO: Dynamic Clipping Policy Optimization (2509.02333)
Uses dynamic clipping, which adjusts probability limits per token for better token exploration, and smooth reward standardization to balance rewards over training steps and prevent wasted updates
4. ARPO (Agentic Reinforced Policy Optimization) → Agentic Reinforced Policy Optimization (2507.19849)
Optimizes multi-turn LLM agents that use external tools. It uses an entropy-based adaptive rollout to explore post-tool use and an advantage attribution method to better assign credit across steps, leading to more efficient tool use with fewer resources
5. GRPO-RoC (Group Relative Policy Optimization with Resampling-on-Correct) → rStar2-Agent: Agentic Reasoning Technical Report (2508.20722)
Oversamples rollouts, then resamples them to keep diverse mistakes and only the highest-quality correct answers. It reduces noises and ends up with stronger reasoning in a code environment
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
-
- 1 reply
replied to
their
post
27 days ago
Adobe Firefly → https://www.adobe.com/products/firefly.html
What is really important, it's trained on ethically sourced data with C2PA provenance. It integrates into Creative Cloud tools like Photoshop Generative Fill, Express, and Firefly Boards. Recent updates add partner AI models (like Google Imagen, OpenAI) and a new Firefly mobile app for iOS and Android. Price: $9.99-29.99/monthRunway Gen-4 (images and videos) → https://runwayml.com/research/introducing-runway-gen-4
A still-image base model tuned for stylistic control and consistency. Its References feature allows users to input up to 3 images, helping preserve visual identity across outputs. Now fully accessible via the Runway API. Price: $12-76/monthIdeogram 3.0 → https://ideogram.ai/features/3.0
The current leader for clean, controllable text in images with Style Reference and strong layout/typography. Great for posters, logos, marketing, etc. Price: ~$0.03-0.09 per output imageLeonardo Phoenix (Leonardo AI) → https://leonardo.ai/phoenix/
Leonardo’s first foundation model emphasizing prompt adherence + readable text. It offers Style Reference for visual control, and Character Reference for consistent characters across shots. Price: $10-48/monthFreepik Mystic → https://www.freepik.com/ai/mystic
Delivers Full‑HD photorealism including lifelike portraits and accurate in‑image text without requiring post-processing. Built in collaboration with Magnific AI, it's integrated into the Freepik AI Image Generator suite. Price: € 5-143.75/monthPixArt-Σ (open-source) → https://pixart-alpha.github.io/PixArt-sigma-project/
A DiT-based T2I model that directly generates up to 4K, showing strong prompt following with a compact footprint. It's a great OSS alternative for researchers/builders. Freely available
posted
an
update
27 days ago
Post
411
11 Powerful Image Models
Everyone is buzzing around image generation this week, or more specifically, Google's Nano-Banana. So today we want to share a list of models that can be your great toolkit for image generation + editing + multi-turn refinement.
1. Gemini 2.5 Flash Image, or Nano-Banana →
https://deepmind.google/models/gemini/image/
Google’s newest image model with conversational editing, character consistency, and multi-image fusion. Available in AI Studio and the Gemini API. Price: $2.50 per 1M tokens
2. FLUX (Black Forest Labs) → https://bfl.ai/
A family of models known for rich detail and, excellent prompt adherence, and fast iterative generation. Offered in several variants, from Pro to open-source, it's accessible via Hugging Face, Replicate, Azure AI Foundry, etc., and used as a base in many pipelines. Price: $0.025-0.08 per image
3. Midjourney v7 → https://www.midjourney.com/
Enhanced image fidelity, prompt comprehension, and anatomical coherence (hands, bodies, objects) + provides a smart lightbox editor. The Omni-reference tool improves character and object consistency in your images. It remains accessible via Discord with a supporting web interface. Price: $10-60/month
4. Stable Diffusion 3.5 (Stability AI) → https://stability.ai/stable-image
Open-weights line with improved text rendering, photorealism, and
prompt adherence compared to earlier versions. It introduces technical innovations through its MMDiT architecture. Price: $0.025-0.065 per image
5. OpenAI GPT-Image-1 →https://platform.openai.com/docs/guides/image-generation?image-generation-model=gpt-image-1
It's the same multimodal model that powers ChatGPT's image capabilities, offering high-fidelity image generation, precise edits, including inpainting, and accurate text rendering. Available via the Images API. Price: $40 per 1M tokens
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
Everyone is buzzing around image generation this week, or more specifically, Google's Nano-Banana. So today we want to share a list of models that can be your great toolkit for image generation + editing + multi-turn refinement.
1. Gemini 2.5 Flash Image, or Nano-Banana →
https://deepmind.google/models/gemini/image/
Google’s newest image model with conversational editing, character consistency, and multi-image fusion. Available in AI Studio and the Gemini API. Price: $2.50 per 1M tokens
2. FLUX (Black Forest Labs) → https://bfl.ai/
A family of models known for rich detail and, excellent prompt adherence, and fast iterative generation. Offered in several variants, from Pro to open-source, it's accessible via Hugging Face, Replicate, Azure AI Foundry, etc., and used as a base in many pipelines. Price: $0.025-0.08 per image
3. Midjourney v7 → https://www.midjourney.com/
Enhanced image fidelity, prompt comprehension, and anatomical coherence (hands, bodies, objects) + provides a smart lightbox editor. The Omni-reference tool improves character and object consistency in your images. It remains accessible via Discord with a supporting web interface. Price: $10-60/month
4. Stable Diffusion 3.5 (Stability AI) → https://stability.ai/stable-image
Open-weights line with improved text rendering, photorealism, and
prompt adherence compared to earlier versions. It introduces technical innovations through its MMDiT architecture. Price: $0.025-0.065 per image
5. OpenAI GPT-Image-1 →https://platform.openai.com/docs/guides/image-generation?image-generation-model=gpt-image-1
It's the same multimodal model that powers ChatGPT's image capabilities, offering high-fidelity image generation, precise edits, including inpainting, and accurate text rendering. Available via the Images API. Price: $40 per 1M tokens
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
-
- 1 reply
posted
an
update
about 2 months ago
Post
3580
6 Must-read books about AI and Machine Learning:
Sharing some free, useful resources for you. In this collection, we’ve gathered the most recent books to give you up-to-date information on key fundamental topics. Hope this helps you master AI and machine learning:
1. Machine Learning Systems by Vijay Janapa Reddi → https://www.mlsysbook.ai/
Provides a framework for building effective ML solutions, covering data engineering, optimization, hardware-aware training, inference acceleration, architecture choice, and other key principles
2. Generative Diffusion Modeling: A Practical Handbook by Zihan Ding, Chi Jin → https://arxiv.org/abs/2412.17162
Offers a unified view of diffusion models: probabilistic, score-based, consistency, rectified flow, pre/post-training. It aligns notations with code to close the “paper-to-code” gap.
3. Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges → https://arxiv.org/abs/2104.13478
Explores unified geometric principles to analyze neural networks' architectures: CNNs, RNNs, GNNs, Transformers, and guide the design of the future ones
4. Mathematical Foundations of Geometric Deep Learning by Haitz Saez de Ocariz Borde and Michael Bronstein → https://arxiv.org/abs/2508.02723
Dives into the the key math concepts behind geometric Deep Learning: geometric and analytical structures, vector calculus, differential geometry, etc.
5. Interpretable Machine Learning by Christoph Molnar → https://github.com/christophM/interpretable-ml-book
Practical guide to simple, transparent models (e.g., decision trees) and model-agnostic methods like LIME, Shapley values, permutation importance, and accumulated local effects.
6. Understanding Deep Learning by Simon J.D. Prince → https://udlbook.github.io/udlbook/
Explores core deep learning concenpts: models, training, evaluation, RL, architectures for images, text, and graphs, addressing open theoretical questions
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
Sharing some free, useful resources for you. In this collection, we’ve gathered the most recent books to give you up-to-date information on key fundamental topics. Hope this helps you master AI and machine learning:
1. Machine Learning Systems by Vijay Janapa Reddi → https://www.mlsysbook.ai/
Provides a framework for building effective ML solutions, covering data engineering, optimization, hardware-aware training, inference acceleration, architecture choice, and other key principles
2. Generative Diffusion Modeling: A Practical Handbook by Zihan Ding, Chi Jin → https://arxiv.org/abs/2412.17162
Offers a unified view of diffusion models: probabilistic, score-based, consistency, rectified flow, pre/post-training. It aligns notations with code to close the “paper-to-code” gap.
3. Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges → https://arxiv.org/abs/2104.13478
Explores unified geometric principles to analyze neural networks' architectures: CNNs, RNNs, GNNs, Transformers, and guide the design of the future ones
4. Mathematical Foundations of Geometric Deep Learning by Haitz Saez de Ocariz Borde and Michael Bronstein → https://arxiv.org/abs/2508.02723
Dives into the the key math concepts behind geometric Deep Learning: geometric and analytical structures, vector calculus, differential geometry, etc.
5. Interpretable Machine Learning by Christoph Molnar → https://github.com/christophM/interpretable-ml-book
Practical guide to simple, transparent models (e.g., decision trees) and model-agnostic methods like LIME, Shapley values, permutation importance, and accumulated local effects.
6. Understanding Deep Learning by Simon J.D. Prince → https://udlbook.github.io/udlbook/
Explores core deep learning concenpts: models, training, evaluation, RL, architectures for images, text, and graphs, addressing open theoretical questions
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
- 2 replies
replied to
their
post
about 2 months ago
iVideoGPT → https://huggingface.co/papers/2405.15223
Unifies visual observations, actions, and rewards into a single token sequence, enabling scalable, interactive world modeling of high-dimensional environmentsMaskGWM → https://huggingface.co/papers/2502.11663
It's used for autonomous driving. It improves long-horizon and multi-view prediction by combining video generation with MAE-style feature-level context learning. Its innovations include: scalable Diffusion Transformers, diffusion-aware mask tokens, and spatial-temporal masking.World-model-augmented (WMA) web agent → https://huggingface.co/papers/2410.13232
This mix of a world model and LLM-based web agents enables agents to simulate future outcomes in natural language and avoid mistakes in long-horizon tasks. The world model's transition-focused abstraction allows for efficient policy improvementNavigation World Models from Meta →
https://huggingface.co/papers/2412.03572
Allows agents to simulate and evaluate navigation trajectories before acting. Powered by a large Conditional Diffusion Transformer, NWM adapts to dynamic constraints and generalizes to unfamiliar environments with a single imageСosmos World Foundation Models by NVIDIA →
https://huggingface.co/papers/2501.03575
Include 3 model families: 1) Cosmos-Predict1 simulates how the visual world evolves over time, learning physical world dynamics from video clips; 2) Cosmos-Transfer1 allows to guide world generation using multiple spatial control signals: segmentation, depth, edge maps, blurred visual inputs, etc.; 3) Cosmos-Reason1 reasons about what is happening, what will happen next, and what actions are feasible.DreamerV3, Google DeepMind → https://arxiv.org/abs/2301.04104
A single, general-purpose world model-based RL algorithm. It demonstrates robust, farsighted planning in complex environments without human data or reward shaping, and excels in tasks like collecting diamonds in Minecraft from scratch.Genie 2, Google DeepMind →
https://deepmind.google/discover/blog/genie-2-a-large-scale-foundation-world-model/
Generates diverse training environments for embodied agents. From a single image prompt, it creates playable virtual worlds controllable via keyboard and mouse usable by both humans and AI systems.
posted
an
update
about 2 months ago
Post
3440
12 Powerful World Models
World models are one of the most challenging areas in AI, pushing the boundaries of reasoning, perception, and planning. They're gen AI systems that help models and agents learn internal representations of real-world environments.
Today, we invite you to take a look at 12 standout examples:
1. WorldVLA → WorldVLA: Towards Autoregressive Action World Model (2506.21539)
This autoregressive world model integrates action prediction and visual world modeling in a single framework, allowing each to enhance the other. It introduces an attention masking strategy to reduce action prediction errors
2. SimuRA → https://arxiv.org/abs/2507.23773
A generalized world model that uses a language-based world model to simulate and plan actions before execution, enabling more general and flexible reasoning
3. PAN (Physical, Agentic, and Nested) world models → Critiques of World Models (2507.05169)
Has a hybrid architecture that combines discrete concept-based reasoning (via LLMs) with continuous perceptual simulation (via diffusion models), enabling rich multi-level, multimodal understanding and prediction
4. MineWorld by Microsoft Research → MineWorld: a Real-Time and Open-Source Interactive World Model on Minecraft (2504.08388)
Enables real-time, interactive world modeling in Minecraft by combining visual and action tokenization within an autoregressive Transformer. It uses parallel decoding for fast scene generation (4–7 FPS)
5. WorldMem → WORLDMEM: Long-term Consistent World Simulation with Memory (2504.12369)
Uses a memory bank with attention over time-stamped frames and states to maintain long-term and 3D spatial consistency in scene generation. So it reconstruct past scenes and simulate dynamic world changes across large temporal gaps
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
Plus explore this article for a comprehensive overview of the history and current evolution of world models: https://www.turingpost.com/p/topic-35-what-are-world-models
World models are one of the most challenging areas in AI, pushing the boundaries of reasoning, perception, and planning. They're gen AI systems that help models and agents learn internal representations of real-world environments.
Today, we invite you to take a look at 12 standout examples:
1. WorldVLA → WorldVLA: Towards Autoregressive Action World Model (2506.21539)
This autoregressive world model integrates action prediction and visual world modeling in a single framework, allowing each to enhance the other. It introduces an attention masking strategy to reduce action prediction errors
2. SimuRA → https://arxiv.org/abs/2507.23773
A generalized world model that uses a language-based world model to simulate and plan actions before execution, enabling more general and flexible reasoning
3. PAN (Physical, Agentic, and Nested) world models → Critiques of World Models (2507.05169)
Has a hybrid architecture that combines discrete concept-based reasoning (via LLMs) with continuous perceptual simulation (via diffusion models), enabling rich multi-level, multimodal understanding and prediction
4. MineWorld by Microsoft Research → MineWorld: a Real-Time and Open-Source Interactive World Model on Minecraft (2504.08388)
Enables real-time, interactive world modeling in Minecraft by combining visual and action tokenization within an autoregressive Transformer. It uses parallel decoding for fast scene generation (4–7 FPS)
5. WorldMem → WORLDMEM: Long-term Consistent World Simulation with Memory (2504.12369)
Uses a memory bank with attention over time-stamped frames and states to maintain long-term and 3D spatial consistency in scene generation. So it reconstruct past scenes and simulate dynamic world changes across large temporal gaps
Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe
Plus explore this article for a comprehensive overview of the history and current evolution of world models: https://www.turingpost.com/p/topic-35-what-are-world-models
-
- 1 reply
replied to
their
post
2 months ago
CISPO: Clipped Importance Sampling Policy Optimization →
https://huggingface.co/papers/2506.13585
This RL algorithm from the MiniMax-M1 project clips importance-sampling weights instead of per-token updates. This lets all tokens (even rare but crucial ones) contribute to learning, avoiding the token-level clipping. CISPO also avoids KL penalties and uses group relative advantage like GRPO.PAPO: Perception-Aware Policy Optimization → https://huggingface.co/papers/2507.06448
Enhances RL in vision-language tasks by adding a KL-based perception loss to the GRPO objective for better visual alignment during training. It boosts accuracy by 4–8% and reduces perception errors by ~30%.OPO: On-Policy RL with Optimal Baseline → https://huggingface.co/papers/2505.23585
A simplified RL algorithm from Microsoft that enforces strict on-policy training by using freshly sampled outputs from the current policy for every update, minimizing off-policy drift. It minimizes gradient variance, avoiding auxiliary models and regularization.EXPO: Expressive Policy Optimization → https://huggingface.co/papers/2507.07986
Trains complex policies by pairing a large base model with a lightweight edit policy that suggests better actions, selecting the best of both without backpropagating through the base.
posted
an
update
2 months ago
Post
5041
9 new policy optimization techniques
Reinforcement Learning (RL) won't stuck in the same old PPO loop - in the last two months alone, researchers have introduced a new wave of techniques, reshaping how we train and fine-tune LLMs, VLMs, and agents.
Here are 9 fresh policy optimization techniques worth knowing:
1. GSPO: Group Sequence Policy Optimization → Group Sequence Policy Optimization (2507.18071)
Shifts from token-level to sequence-level optimization, clipping, and rewarding to capture the full picture and increase stability compared to GRPO. GSPO-token variation also allows token-level fine-tuning.
2. LAPO: Length-Adaptive Policy Optimization → LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimization (2507.15758)
A two-stage RL framework that trains models to adaptively control reasoning length by learning typical solution lengths for shorter and more efficient reasoning.
3. HBPO: Hierarchical Budget Policy Optimization → Hierarchical Budget Policy Optimization for Adaptive Reasoning (2507.15844)
This one trains model to adapt reasoning depth based on problem complexity. It divides training samples into subgroups with different token budgets, using budget-aware rewards to align reasoning effort with task difficulty.
4. SOPHIA: Semi-off-policy reinforcement learning → Semi-off-Policy Reinforcement Learning for Vision-Language Slow-thinking Reasoning (2507.16814)
Combines on-policy visual understanding from the Vision Language Models (VLMs) with off-policy reasoning from an LM, assigning outcome-based rewards and propagating visual rewards backward through the reasoning steps.
5. RePO: Replay-Enhanced Policy Optimization → RePO: Replay-Enhanced Policy Optimization (2506.09340)
Introduces a replay buffer into on-policy RL for LLMs, retrieving diverse off-policy samples for each prompt to broaden the training data per prompt
Read further below ⬇️
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
Reinforcement Learning (RL) won't stuck in the same old PPO loop - in the last two months alone, researchers have introduced a new wave of techniques, reshaping how we train and fine-tune LLMs, VLMs, and agents.
Here are 9 fresh policy optimization techniques worth knowing:
1. GSPO: Group Sequence Policy Optimization → Group Sequence Policy Optimization (2507.18071)
Shifts from token-level to sequence-level optimization, clipping, and rewarding to capture the full picture and increase stability compared to GRPO. GSPO-token variation also allows token-level fine-tuning.
2. LAPO: Length-Adaptive Policy Optimization → LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimization (2507.15758)
A two-stage RL framework that trains models to adaptively control reasoning length by learning typical solution lengths for shorter and more efficient reasoning.
3. HBPO: Hierarchical Budget Policy Optimization → Hierarchical Budget Policy Optimization for Adaptive Reasoning (2507.15844)
This one trains model to adapt reasoning depth based on problem complexity. It divides training samples into subgroups with different token budgets, using budget-aware rewards to align reasoning effort with task difficulty.
4. SOPHIA: Semi-off-policy reinforcement learning → Semi-off-Policy Reinforcement Learning for Vision-Language Slow-thinking Reasoning (2507.16814)
Combines on-policy visual understanding from the Vision Language Models (VLMs) with off-policy reasoning from an LM, assigning outcome-based rewards and propagating visual rewards backward through the reasoning steps.
5. RePO: Replay-Enhanced Policy Optimization → RePO: Replay-Enhanced Policy Optimization (2506.09340)
Introduces a replay buffer into on-policy RL for LLMs, retrieving diverse off-policy samples for each prompt to broaden the training data per prompt
Read further below ⬇️
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
-
- 1 reply
posted
an
update
2 months ago
Post
6241
6 Essential Reads on core AI/ML topics:
Time to look at some free useful resources that can help you upgrade your knowledge of AI and machine learning!
Today we offer you these 6 must-read surveys that can be your perfect guides to the major fields and techniques:
1. Foundations of Large Language Models by Tong Xiao and Jingbo Zhu → https://arxiv.org/abs/2501.09223
Many recommend this 270-page book as a good resource to focus on fundamental concepts, such as pre-training, generative models, prompting, alignment, and inference
2. Large Language Models Post-Training: Surveying Techniques from Alignment to Reasoning -> A Survey on Post-training of Large Language Models (2503.06072)
Read this to master policy optimization (RLHF, DPO, GRPO), supervised and parameter-efficient fine-tuning, reasoning, integration, and adaptation techniques
3. Agentic Large Language Models, a survey by Leiden University → https://arxiv.org/abs/2503.23037
Surveys agentic LLMs across reasoning, tools, and multi-agent collaboration, highlighting their synergy. It also explores their promise, risks and applications in medicine, finance, science.
4. A Survey of Context Engineering for Large Language Models → A Survey of Context Engineering for Large Language Models (2507.13334)
Defines Context Engineering as systematic info design for LLMs beyond prompting, covering retrieval, processing, management, and architectures like RAG and multi-agent systems
5. A Survey of Generative Categories and Techniques in Multimodal Large Language Models → https://arxiv.org/abs/2506.10016
Covers multimodal models, exploring six generative modalities, key techniques (SSL, RLHF, CoT), architectural trends, and challenges
6. Large Language models for Time Series Analysis: Techniques, Applications, and Challenges → https://arxiv.org/abs/2506.11040
Explains how LLMs transform time series analysis by enhancing pattern recognition and long-term dependency handling + shows how to build them
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
Time to look at some free useful resources that can help you upgrade your knowledge of AI and machine learning!
Today we offer you these 6 must-read surveys that can be your perfect guides to the major fields and techniques:
1. Foundations of Large Language Models by Tong Xiao and Jingbo Zhu → https://arxiv.org/abs/2501.09223
Many recommend this 270-page book as a good resource to focus on fundamental concepts, such as pre-training, generative models, prompting, alignment, and inference
2. Large Language Models Post-Training: Surveying Techniques from Alignment to Reasoning -> A Survey on Post-training of Large Language Models (2503.06072)
Read this to master policy optimization (RLHF, DPO, GRPO), supervised and parameter-efficient fine-tuning, reasoning, integration, and adaptation techniques
3. Agentic Large Language Models, a survey by Leiden University → https://arxiv.org/abs/2503.23037
Surveys agentic LLMs across reasoning, tools, and multi-agent collaboration, highlighting their synergy. It also explores their promise, risks and applications in medicine, finance, science.
4. A Survey of Context Engineering for Large Language Models → A Survey of Context Engineering for Large Language Models (2507.13334)
Defines Context Engineering as systematic info design for LLMs beyond prompting, covering retrieval, processing, management, and architectures like RAG and multi-agent systems
5. A Survey of Generative Categories and Techniques in Multimodal Large Language Models → https://arxiv.org/abs/2506.10016
Covers multimodal models, exploring six generative modalities, key techniques (SSL, RLHF, CoT), architectural trends, and challenges
6. Large Language models for Time Series Analysis: Techniques, Applications, and Challenges → https://arxiv.org/abs/2506.11040
Explains how LLMs transform time series analysis by enhancing pattern recognition and long-term dependency handling + shows how to build them
Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
- 1 reply
replied to
their
post
3 months ago
FreeLoRA → https://huggingface.co/papers/2507.01792
Enables training-free image generation with multiple subjects by fine-tuning each LoRA module on one subject. During inference, subject-aware activation applies modules only to their target tokens, ensuring clean, interference-free fusion.LoRA-Augmented Generation (LAG) → https://huggingface.co/papers/2507.05346
Uses large collections of task-specific LoRA adapters without needing extra training or data. It selects and applies the most relevant adapters at each layer and token, exceling in knowledge-intensive tasks.ARD-LoRA (Adaptive Rank Dynamic LoRA) → https://huggingface.co/papers/2506.18267
Adjusts the rank of LoRA adapters dynamically across transformer layers and heads by learning per-head scaling factors through a meta-objective. It balances performance, efficiency, using fewer parameters and reducing memory use.WaRA → https://huggingface.co/papers/2506.24092
Designed for vision tasks, it uses wavelet transforms and decomposes weight updates into multiple resolutions, capturing both coarse and detailed patterns.BayesLoRA → https://huggingface.co/papers/2506.22809
Adds uncertainty estimation to LoRA adapters using MC-Dropout, helping models gauge confidence in unfamiliar situations. It detects variance outside fine-tuned distributions, supporting more cautious and adaptive behavior of models.Dual LoRA Learning (DLoRAL) → https://huggingface.co/papers/2506.15591
Trains two LoRA branches: C-LoRA captures temporal coherence from degraded input, while D-LoRA improves visual detail. It's used for video super-resolution that enhances both spatial detail and temporal consistency.Safe Pruning LoRA (SPLoRA) → https://huggingface.co/papers/2506.18931
Improves the safety of LoRA-tuned LMs by selectively removing LoRA layers that reduce alignment, using a new E-DIEM metric to detect safety-related shifts without relying on data labels.PLoP (Precise LoRA Placement) → https://huggingface.co/papers/2506.20629
A lightweight method that automatically selects optimal LoRA adapter placement during fine-tuning based on the model and task
posted
an
update
3 months ago
Post
5172
13 New types of LoRA
LoRA (Low-Rank Adaptation) is a popular lightweight method for fine-tuning AI models. It doesn't update the full model, it adds small trainable components, low-rank matrices, while keeping the original weights frozen. Only these adapters are trained.
Recently, many interesting new LoRA variations came out, so it’s a great time to take a look at these 13 clever approaches:
1. T-LoRA → T-LoRA: Single Image Diffusion Model Customization Without Overfitting (2507.05964)
A timestep-dependent LoRA method for adapting diffusion models with a single image. It dynamically adjusts updates and uses orthogonal initialization to reduce overlap, achieving better fidelity–alignment balance than standard LoRA
2. SingLoRA → SingLoRA: Low Rank Adaptation Using a Single Matrix (2507.05566)
Simplifies LoRA by using only one small matrix instead of usual two, and multiplying it by its own transpose (like A × Aᵀ). It uses half the parameters of LoRA and avoids scale mismatch between different matrices
3. LiON-LoRA → LiON-LoRA: Rethinking LoRA Fusion to Unify Controllable Spatial and Temporal Generation for Video Diffusion (2507.05678)
Improves control and precision in video diffusion models when training data is limited. It builds on LoRA, adding 3 key principles: linear scalability, orthogonality, and norm consistency. A controllable token and modified self-attention enables smooth adjustment of motion
4. LoRA-Mixer → LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routing (2507.00029)
Combines LoRA and mixture-of-experts (MoE) to adapt LLMs for multiple tasks. It dynamically routes task-specific LoRA experts into linear projections of attention modules, supporting both joint training and frozen expert reuse
5. QR-LoRA → QR-LoRA: Efficient and Disentangled Fine-tuning via QR Decomposition for Customized Generation (2507.04599)
Separates content and style when combining multiple LoRA adapters. It implements QR decomposition to structure parameter updates, where the orthogonal Q matrix reduces interference between features, and the R matrix captures specific transformations
Read further in the comments 👇
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
LoRA (Low-Rank Adaptation) is a popular lightweight method for fine-tuning AI models. It doesn't update the full model, it adds small trainable components, low-rank matrices, while keeping the original weights frozen. Only these adapters are trained.
Recently, many interesting new LoRA variations came out, so it’s a great time to take a look at these 13 clever approaches:
1. T-LoRA → T-LoRA: Single Image Diffusion Model Customization Without Overfitting (2507.05964)
A timestep-dependent LoRA method for adapting diffusion models with a single image. It dynamically adjusts updates and uses orthogonal initialization to reduce overlap, achieving better fidelity–alignment balance than standard LoRA
2. SingLoRA → SingLoRA: Low Rank Adaptation Using a Single Matrix (2507.05566)
Simplifies LoRA by using only one small matrix instead of usual two, and multiplying it by its own transpose (like A × Aᵀ). It uses half the parameters of LoRA and avoids scale mismatch between different matrices
3. LiON-LoRA → LiON-LoRA: Rethinking LoRA Fusion to Unify Controllable Spatial and Temporal Generation for Video Diffusion (2507.05678)
Improves control and precision in video diffusion models when training data is limited. It builds on LoRA, adding 3 key principles: linear scalability, orthogonality, and norm consistency. A controllable token and modified self-attention enables smooth adjustment of motion
4. LoRA-Mixer → LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routing (2507.00029)
Combines LoRA and mixture-of-experts (MoE) to adapt LLMs for multiple tasks. It dynamically routes task-specific LoRA experts into linear projections of attention modules, supporting both joint training and frozen expert reuse
5. QR-LoRA → QR-LoRA: Efficient and Disentangled Fine-tuning via QR Decomposition for Customized Generation (2507.04599)
Separates content and style when combining multiple LoRA adapters. It implements QR decomposition to structure parameter updates, where the orthogonal Q matrix reduces interference between features, and the R matrix captures specific transformations
Read further in the comments 👇
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
-
- 1 reply
replied to
their
post
3 months ago
AllVoiceLab MCP Server -> https://github.com/allvoicelab/AllVoiceLab-MCP
Enables AI agents to access advanced text-to-speech, voice conversion, and video translation APIs, powering use cases like global content localization, AI audiobooks, and voice-driven media production.MCP Email Server -> https://github.com/Shy2593666979/mcp-server-email
For email functionality: write and send emails with multiple recipients, add and search files within specified directories.Google Admin MCP Server -> https://github.com/securityfortech/google-admin-mcp
Manage Google Workspace users through the Admin Directory API (list, create, get info about users, etc.)Android MCP Server -> https://github.com/minhalvp/android-mcp-server
Provides programmatic control over Android devices through ADB (Android Debug Bridge).DeepView MCP -> https://github.com/ai-1st/deepview-mcp
Enables IDEs (Cursor, Windsurf, etc.) to analyze large codebases using Gemini's extensive context window.Calculator MCP Server -> https://github.com/githejie/mcp-server-calculator
May sound easy, but it's essential for precise numerical calculations within LLMsMCP Aggregator -> https://github.com/nazar256/combine-mcp
Combines multiple MCP servers into a single interface for more convenient use
posted
an
update
3 months ago
Post
6595
13 Outstanding MCP Servers
MCP is redefining how AI assistants connect to the world of data and tools, so no wonder MCP servers are in high demand now. That’s why we’ve curated 13 cool MCP servers to upgrade your workflow:
1. Hugging Face Official MCP Server -> https://github.com/evalstate/hf-mcp-server
Provides an access and interaction with Hugging Face models, datasets, and Gradio Spaces for dynamic tool integration and configuration across environments.
2. Browser MCP -> https://browsermcp.io/
An MCP server +Chrome extension. It allows to automate your browser with AI apps like VS Code, Claude, Cursor, and Windsurf.
3. Bright Data MCP -> https://github.com/brightdata/brightdata-mcp
This one is for working with data in real-time: searching the web, navigating websites, taking action and retrieving data.
4. JSON MCP -> https://github.com/VadimNastoyashchy/json-mcp
Interact with JSON files: split, merge, find specific data, and validate content within them.
5. Octagon Deep Research MCP -> https://github.com/OctagonAI/octagon-deep-research-mcp
Allows for deep research via AI agents, integrating seamlessly with MCP clients like Claude Desktop and Cursor for powerful, unlimited research capabilities.
6. VLM Run MCP Server -> https://docs.vlm.run/mcp/introduction
Provides an agent the ability to see, understand and process visual content.
Read further in the comments 👇
P.S.:
Our most read explanation of MCP on Hugging Face https://huggingface.co/blog/Kseniase/mcp
Our first list of 13 awesome MCP servers: https://huggingface.co/posts/Kseniase/204958200717570
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
MCP is redefining how AI assistants connect to the world of data and tools, so no wonder MCP servers are in high demand now. That’s why we’ve curated 13 cool MCP servers to upgrade your workflow:
1. Hugging Face Official MCP Server -> https://github.com/evalstate/hf-mcp-server
Provides an access and interaction with Hugging Face models, datasets, and Gradio Spaces for dynamic tool integration and configuration across environments.
2. Browser MCP -> https://browsermcp.io/
An MCP server +Chrome extension. It allows to automate your browser with AI apps like VS Code, Claude, Cursor, and Windsurf.
3. Bright Data MCP -> https://github.com/brightdata/brightdata-mcp
This one is for working with data in real-time: searching the web, navigating websites, taking action and retrieving data.
4. JSON MCP -> https://github.com/VadimNastoyashchy/json-mcp
Interact with JSON files: split, merge, find specific data, and validate content within them.
5. Octagon Deep Research MCP -> https://github.com/OctagonAI/octagon-deep-research-mcp
Allows for deep research via AI agents, integrating seamlessly with MCP clients like Claude Desktop and Cursor for powerful, unlimited research capabilities.
6. VLM Run MCP Server -> https://docs.vlm.run/mcp/introduction
Provides an agent the ability to see, understand and process visual content.
Read further in the comments 👇
P.S.:
Our most read explanation of MCP on Hugging Face https://huggingface.co/blog/Kseniase/mcp
Our first list of 13 awesome MCP servers: https://huggingface.co/posts/Kseniase/204958200717570
If you like it, also subscribe to the Turing Post: https://www.turingpost.com/subscribe
-
- 1 reply
replied to
their
post
3 months ago
DeepResearcher -> https://github.com/GAIR-NLP/DeepResearcher
An RL framework for training deep research agents end-to-end in real-world environments with web search, exhibiting emergent behaviour like planning, multi-source validation, self-reflection, and honest defining when the agent doesn't know the answerSearch-R1 -> https://github.com/PeterGriffinJin/Search-R1
Features interleaved search access and an open-source RL training pipeline supporting various algorithms (PPO, GRPO, etc.), LLMs (LLaMA3, Qwen2.5, etc.), and search engines (online, local, retrievers)ReCall -> https://github.com/Agent-RL/ReCall
Trains LLMs to reason with tools via RL, no supervised tool-use data needed. It enables agentic use of tools like OpenAI o3 and supports synthetic data generation across diverse environments and multi-step tasksOWL -> https://github.com/camel-ai/owl
A framework built on CAMEL-AI framework enabling dynamic multi-agent collaboration for task automation across diverse domains
Here's an awesome study exploring the entire roadmap of Deep Research assistants. Don't forget to check it out -> https://huggingface.co/papers/2506.18096
posted
an
update
3 months ago
Post
3711
10 Open-source Deep Research assistants
Deep Research agents are quickly becoming our daily co-workers — built for complex investigations, not just chat. With modular architecture, advanced tool use and real web access, they go far beyond typical AI. While big-name agents get the spotlight, we want to highlight some powerful recent open-source alternatives:
1. DeerFlow -> https://github.com/bytedance/deer-flow
A modular multi-agent system combining LMs and tools for automated research and code analysis. It links a coordinator, planner, team of specialized agent, and reporter, and converts reports to speech via Text-to-Speech (TTS)
2. Alita -> https://github.com/CharlesQ9/Alita
Uses a single problem-solving module for scalable reasoning through simplicity. It self-evolves by generating and reusing Model Context Protocols (MCPs) from open-source tools to build external capabilities for diverse tasks
3. WebThinker -> https://github.com/RUC-NLPIR/WebThinker
Lets reasoning models autonomously search the web and navigate pages. Deep Web Explorer allows interaction with links and follow-up searches. Through a Think-Search-and-Draft process models generate and refine reports in real time. RL training with preference pairs improves the workflow
4. SimpleDeepSearcher -> https://github.com/RUCAIBox/SimpleDeepSearcher
A lightweight framework showing that supervised fine-tuning is a real alternative to complex RL, using simulated web interactions and multi-criteria curation to generate high-quality training data
5. AgenticSeek -> https://github.com/Fosowl/agenticSeek
A private, on-device assistant that picks the best agent expert for browsing, coding, or planning—no cloud needed. Includes voice input via speech-to-text
6. Suna -> https://github.com/kortix-ai/suna
Offers web browsing, file and doc handling, CLI execution, site deployment, and API/service integration—all in one assistant
Subscribe to the Turing Post:https://www.turingpost.com/subscribe
Read further ⬇️
Deep Research agents are quickly becoming our daily co-workers — built for complex investigations, not just chat. With modular architecture, advanced tool use and real web access, they go far beyond typical AI. While big-name agents get the spotlight, we want to highlight some powerful recent open-source alternatives:
1. DeerFlow -> https://github.com/bytedance/deer-flow
A modular multi-agent system combining LMs and tools for automated research and code analysis. It links a coordinator, planner, team of specialized agent, and reporter, and converts reports to speech via Text-to-Speech (TTS)
2. Alita -> https://github.com/CharlesQ9/Alita
Uses a single problem-solving module for scalable reasoning through simplicity. It self-evolves by generating and reusing Model Context Protocols (MCPs) from open-source tools to build external capabilities for diverse tasks
3. WebThinker -> https://github.com/RUC-NLPIR/WebThinker
Lets reasoning models autonomously search the web and navigate pages. Deep Web Explorer allows interaction with links and follow-up searches. Through a Think-Search-and-Draft process models generate and refine reports in real time. RL training with preference pairs improves the workflow
4. SimpleDeepSearcher -> https://github.com/RUCAIBox/SimpleDeepSearcher
A lightweight framework showing that supervised fine-tuning is a real alternative to complex RL, using simulated web interactions and multi-criteria curation to generate high-quality training data
5. AgenticSeek -> https://github.com/Fosowl/agenticSeek
A private, on-device assistant that picks the best agent expert for browsing, coding, or planning—no cloud needed. Includes voice input via speech-to-text
6. Suna -> https://github.com/kortix-ai/suna
Offers web browsing, file and doc handling, CLI execution, site deployment, and API/service integration—all in one assistant
Subscribe to the Turing Post:https://www.turingpost.com/subscribe
Read further ⬇️
-
- 2 replies