Open Source AI

Open weights models, datasets, and frameworks

This Week's Best | Monthly Best | Guide | Trending

Top This Week

[2603.25112] Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory
Llms

[2603.25112] Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory

Abstract page for arXiv paper 2603.25112: Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory

arXiv - AI · 4 min ·
[2603.24772] Evaluating Fine-Tuned LLM Model For Medical Transcription With Small Low-Resource Languages Validated Dataset
Llms

[2603.24772] Evaluating Fine-Tuned LLM Model For Medical Transcription With Small Low-Resource Languages Validated Dataset

Abstract page for arXiv paper 2603.24772: Evaluating Fine-Tuned LLM Model For Medical Transcription With Small Low-Resource Languages Val...

arXiv - Machine Learning · 4 min ·
[2603.25325] How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models
Llms

[2603.25325] How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models

Abstract page for arXiv paper 2603.25325: How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models

arXiv - AI · 4 min ·

All Content

[2505.02881] Rewriting Pre-Training Data Boosts LLM Performance in Math and Code
Llms

[2505.02881] Rewriting Pre-Training Data Boosts LLM Performance in Math and Code

Abstract page for arXiv paper 2505.02881: Rewriting Pre-Training Data Boosts LLM Performance in Math and Code

arXiv - Machine Learning · 4 min ·
[2512.12411] Detecting the Disturbance: A Nuanced View of Introspective Abilities in LLMs
Llms

[2512.12411] Detecting the Disturbance: A Nuanced View of Introspective Abilities in LLMs

Abstract page for arXiv paper 2512.12411: Detecting the Disturbance: A Nuanced View of Introspective Abilities in LLMs

arXiv - AI · 4 min ·
[2603.02041] EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post-Training
Llms

[2603.02041] EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post-Training

Abstract page for arXiv paper 2603.02041: EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post...

arXiv - AI · 4 min ·
[2603.01973] CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production
Llms

[2603.01973] CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production

Abstract page for arXiv paper 2603.01973: CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production

arXiv - AI · 4 min ·
[2603.00917] Prompt Sensitivity and Answer Consistency of Small Open-Source Large Language Models on Clinical Question Answering: Implications for Low-Resource Healthcare Deployment
Llms

[2603.00917] Prompt Sensitivity and Answer Consistency of Small Open-Source Large Language Models on Clinical Question Answering: Implications for Low-Resource Healthcare Deployment

Abstract page for arXiv paper 2603.00917: Prompt Sensitivity and Answer Consistency of Small Open-Source Large Language Models on Clinica...

arXiv - AI · 4 min ·
[2508.03716] FeynTune: Large Language Models for High-Energy Theory
Llms

[2508.03716] FeynTune: Large Language Models for High-Energy Theory

Abstract page for arXiv paper 2508.03716: FeynTune: Large Language Models for High-Energy Theory

arXiv - Machine Learning · 3 min ·
[2509.16622] Audio-Conditioned Diffusion LLMs for ASR and Deliberation Processing
Llms

[2509.16622] Audio-Conditioned Diffusion LLMs for ASR and Deliberation Processing

Abstract page for arXiv paper 2509.16622: Audio-Conditioned Diffusion LLMs for ASR and Deliberation Processing

arXiv - AI · 4 min ·
[2505.00624] FineScope : SAE-guided Data Selection Enables Domain Specific LLM Pruning and Finetuning
Llms

[2505.00624] FineScope : SAE-guided Data Selection Enables Domain Specific LLM Pruning and Finetuning

Abstract page for arXiv paper 2505.00624: FineScope : SAE-guided Data Selection Enables Domain Specific LLM Pruning and Finetuning

arXiv - AI · 4 min ·
Ai Safety

Musk bashes OpenAI in deposition, saying 'nobody committed suicide because of Grok'

Elon Musk criticizes OpenAI during a deposition, asserting that their technology has not led to severe consequences, highlighting his vie...

Reddit - Artificial Intelligence · 1 min ·
Ai Startups

I built a tool to automate your workflow after recording yourself doing the task once (Open Source)

The article discusses a newly developed open-source tool called Automated, designed to simplify workflow automation by allowing users to ...

Reddit - Artificial Intelligence · 1 min ·
Machine Learning

[R] AudioMuse-AI-DCLAP - LAION CLAP distilled for text to music

AudioMuse-AI-DCLAP is a distilled version of the LAION CLAP model for music, allowing users to search songs by text through a shared embe...

Reddit - Machine Learning · 1 min ·
Machine Learning

[R] Qwen3.5’s MoE architecture: A breakthrough or just incremental?

The discussion revolves around Qwen3.5's MoE architecture, debating whether its low active parameter count signifies a significant breakt...

Reddit - Machine Learning · 1 min ·
Llms

[P] Micro Diffusion — Discrete text diffusion in ~150 lines of pure Python

This article presents a minimal implementation of discrete text diffusion in Python, inspired by Karpathy's MicroGPT, showcasing the core...

Reddit - Machine Learning · 1 min ·
Llms

I used steelman prompting to audit bias across six major LLMs. The default-to-steelman gap was consistent and measurable.

This article discusses an experiment using steelman prompting to evaluate bias in six major LLMs, focusing on their interpretations of 1 ...

Reddit - Artificial Intelligence · 1 min ·
Musk bashes OpenAI in deposition, saying 'nobody committed suicide because of Grok' | TechCrunch
Llms

Musk bashes OpenAI in deposition, saying 'nobody committed suicide because of Grok' | TechCrunch

Elon Musk criticizes OpenAI's safety record in a deposition for his lawsuit against the company, claiming his AI venture, xAI, prioritize...

TechCrunch - AI · 5 min ·
Ai Infrastructure

Numerous AMDXDNA Ryzen AI driver fixes for Linux 7.0-rc2

The article discusses recent fixes for the AMD XDNA Ryzen AI driver in Linux 7.0-rc2, highlighting improvements and updates that enhance ...

Reddit - Artificial Intelligence · 1 min ·
Machine Learning

[P] Tessera — An open protocol for AI-to-AI knowledge transfer across architectures

Tessera introduces an innovative protocol for AI-to-AI knowledge transfer, enabling models to share learned knowledge without direct arch...

Reddit - Machine Learning · 1 min ·
[2510.03495] AgentHub: A Registry for Discoverable, Verifiable, and Reproducible AI Agents
Llms

[2510.03495] AgentHub: A Registry for Discoverable, Verifiable, and Reproducible AI Agents

AgentHub proposes a registry for AI agents that enhances discoverability, verifiability, and reproducibility, addressing gaps in current ...

arXiv - AI · 4 min ·
[2602.22895] SPD Learn: A Geometric Deep Learning Python Library for Neural Decoding Through Trivialization
Machine Learning

[2602.22895] SPD Learn: A Geometric Deep Learning Python Library for Neural Decoding Through Trivialization

SPD Learn is a new Python library designed for geometric deep learning, specifically for neural decoding using symmetric positive definit...

arXiv - Machine Learning · 3 min ·
[2602.22631] TorchLean: Formalizing Neural Networks in Lean
Machine Learning

[2602.22631] TorchLean: Formalizing Neural Networks in Lean

TorchLean is a framework that formalizes neural networks within the Lean 4 theorem prover, enabling precise semantics for execution and v...

arXiv - Machine Learning · 4 min ·
Previous Page 3 Next

Related Topics

Machine Learning Large Language Models Generative AI Natural Language Processing Computer Vision Robotics & Embodied AI

Stay updated with AI News

Get the latest news, tools, and insights delivered to your inbox.

Subscribe to Newsletter

Daily or weekly digest • Unsubscribe anytime