Lists (3)
Starred repositories
PyTorch Code for Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation
Real-time MCP server that tracks context budget, detects agent loops, and alerts before token overflow — for Claude Code, Cursor, and any MCP-compatible AI tool.
A recording proxy for MCP servers. Captures full request/response traces, pins anything around a failure permanently, and prunes the rest.
Real-time Next.js dashboard for ContextPulse — live context budget bar, tool call waterfall, and agent run history.
Real-time uptime monitoring for MCP servers. Pings registered servers on a schedule, tracks latency, and alerts the moment one goes down, degrades, or recovers.
A PyTorch framework for DABSN—a recurrent architecture with admitted, permanent, and long-term memory—featuring native CPU/CUDA kernels and distributed training.
Launch Claude Code with Hugging Face Inference Providers
[ICML 2026] effGen: Enabling Small Language Models as Capable Autonomous Agents
Python code snippets from Discrete Mathematics for Computer Science specialization at Coursera
Unofficial implementation of the Dreamer 4 world model in PyTorch.
Code for paper OpenWebRL: Online Multi-Turn Reinforcement Learning for Visual Web Agents
Offical Implementation for "Recursive Multi-Agent Systems"
[NeurIPS 2025] RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning
Official code for "SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization"
HY-Embodied: Embodied Foundation Models for Real-World Agents
Repo for paper "Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability"
[ACL 2026 oral] SeLaR: Selective Latent Reasoning in Large Language Models
Jointly Optimizing Large Language Models for Reasoning and Self-Refinement
Continuous Thought Machines, because thought takes time and reasoning is a process.
Awesome Jailbreak, red teaming arxiv papers (Automatically Update Every 12th hours)