Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT GPT-5.6, Codex GPT-5.6, GPT-5.5. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xA…

A Multimodal Reasoning Agent with Stateful Experiences

Sparking "Thinking with Videos" via Reinforcement Learning

High-performance In-browser LLM Inference Engine

Let us control diffusion models!

[NeurIPS 2024] Official implementation for paper "Can Graph Learning Improve Planning in LLM-based Agents?"

Vector (and Scalar) Quantization, in Pytorch

Development repository for the Triton language and compiler

🦁 Lion, new optimizer discovered by Google Brain using genetic algorithms that is purportedly better than Adam(w), in Pytorch

Grounded SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & Recognize Anything - Automatically Detect , Segment and Generate Anything

Train Models Contrastively in Pytorch

Repository hosting code for "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152).

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

A place to save my Oh-My-Zsh configs and customized files

Code for CIKM'23 "Replace Scoring with Arrangment: A Contextual Set-to-Arrangement Framework for Learning-to-Rank"

GUNDAM is a data management system that prioritizes data using language models.

CORE is a plug-and-play conversational agent for any recommender system.

The released code of paper Debiased Recommendation with User Feature Balancing

Code for ICLR'22 "Inductive Relation Prediction Using Analogy Subgraph Embeddings"

A Toolkit for Mining Data in a Structural Fashion

Code for WWW'22 "Who to Watch Next: Two-side Interactive Networks for Live Broadcast Recommendation"

Code for WWW'22 "Learn over Past, Evolve for Future: Search-based Time-aware Recommendation with Sequential Behavior Data"

Code for SIGIR'20 “A Deep Recurrent Survival Model for Unbiased Ranking”

Code for CIKM'19 "CoRide: Joint Order Dispatching and Fleet Management for Multi-Scale Ride-Hailing Platforms"

A latent text-to-image diffusion model