community-projects
AI health condition recognition - anyone wanna collaborate
Visualizing Neural Network Optimizer Trajectories
Interpretability
Agentic Research inspired by Simon Willison
Other Modalities
Sharing Two Open-Source Projects for LLM Research
Dataset server Neurvance
Learning to backtrack
Other Modalities
Alignment
Embedding-space semantic pollution: measurement and alternative tokeniz
NLP
Understanding How LLMs Understand Humans: Training Data Attribution for Social Reasoning
NLP
Platinum-CoT: High-Value Technical Reasoning
K-Splanifolds: A new ML algo that outperforms MLPs
Affective Regulation Layer for Language Models
Other Modalities
NLP
Alignment
Interpretability
Dynamically teleporting between layers in transformer based models
NLP
MAS for Enterprise Research
Anyone interested in Computer Vision and Solar Forecasting?
Alignment
Polyphonic MusicVAE
Other Modalities
Weird repeated pattern in 100 MB of frontier-model conversation logs
Project Dragon: 94% RAG Vector Compression with 0.92 Fidelity (Hexagonal Resonance Arch)
The AI HEART Project
Open-source Sesame-level conversational speech model via fine-tuning Kokoro
Exploring and improving evolution strategies for LLMs
[🔥Active] Call for Annotating/Dev/Review Process Rewards of General Reasoning Problems in LLMs
NLP
Structural Affordances for LLM Reasoning + NTP applications
NLP
Information Theory for Distributed Thinking (and its Empirics)
collaborator needed for NeurIPS Paper on AI & Mathematical Creativity (6-Day Sprint)
measuring creative math reasoning in LLMs (NeurIPS Sprint!)
Open-source LLM Adversarial Evaluation Platform
common pile, but audio
RIFT (Resonant Interference Framework for Transformations)
Scalable oversight to mitigate sycophancy
NLP
Adding an Output Latent to MLA
NLP
Interpretability
NanoGPT for Diffusion Language Models (Speedrun, MLSys)
NLP
Scaling Laws for LLM Based Data Compression
Biologically Plausible Alternative to Backpropagation
Other Modalities
New Decoder/Encoder architectures
NLP
RANDUMB: MOSTLY RANDOM NEURAL NETWORKS
Summer of Open Science / Open Science job board
Common Pile Edu
NLP
From Bits to Thoughts
Personal Finance advisor- Evaluation
Symbolic Compression Schemas for semantic anchoring.
AI for math
Hi @community-projects,
Ai Mastery Holy grail Document
UQLM: Uncertainty Quantification for Language Models
Expressive RWKV Variant
NLP
Custom RNN architecture with potentially really long memory – need help with optimization
AI Co-Scientist
LLM architecture universality laws
Ranked list of projects in AI interpretability based on recent ML conference (ICLR)
NLP
Alignment
Interpretability
Predicted Environmental Impacts of Large-Language-Models