A multi-agent framework for simulating emergent social structures and relational order using LLMs.

An LLM-powered Verilog debugging framework leveraging golden reference models and CDFGs.

Structured reasoning framework and benchmark for cinematographic video captioning.

A neuro-symbolic driving framework that supervises driving VLAs with rule-grounded reasoning traces.

Self-hosted media server streaming from Stremio, local files, or WebDAV via Jellyfin API.

Policy-Aware Value-field Equalization for stabilizing action-gradient fields in reinforcement learning.

Adversarially Anchored Preference Alignment framework for LLM post-training.

A flow-matching model for generating human-like robot grasps from RGB-D images.

A hierarchical graph RAG framework for context-aware and relation-aware retrieval.

A multimodal CAD dataset with executable programs and kernel-validated feature histories.

A hub between different proof systems like Agda, Lean, and Rocq for symbolic informalization.

A lightweight multi-platform CPU emulator framework based on QEMU

A framework for certifying faithfulness between natural-language and formal mathematical statements.

A unified and comprehensive codebase for LLM-based Multi-Agent Systems.

A proof-carrying substrate that rewrites software into canonical behavior profiles to reduce LLM training noise.

A terminal AI coding assistant with cross-session memory and an MIT license.

An action controllable multi-subject world model for generative video games.

A harness designed to assist LLM agents in understanding and fixing compiler bugs in LLVM.

Implementation and evaluation results for iterative retrieval-reasoning loops in scientific QA.

A message-passing network architecture for geometric graphs with theoretical guarantees for non-isomorphic graph separation.