Implement a reasoning LLM in PyTorch from scratch, step by step

Repo for AI Agents The Definitive Guide

Multi-agent demo platform for Titans (arXiv:2501.00663) — neural networks that learn to memorize at test time. 7 AI agents, native desktop UI.

Inference-time scaling for LLMs-as-a-judge.

Official repository of the spotlight ICML 2025 paper, PokeChamp: an Expert-level Minimax Language Agent.

Compound model panel for pi: parallel model calls, then one synthesis response.

Airgapped closed-corpus QA loop: a self-hosted Qwen3.6 agent explores a .zip dataroom under a token budget with local tools

Multi-Agent Verification: Scaling Test-Time Compute with Multiple Verifiers

[NeurIPS 2025 Spotlight] Official implementation of "Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecasting"

Test-Time Memory Framework: Control Hallucinations in Foundation Models

Evolving agent harnesses: a research program on how far N orchestrated calls of a small model can rival a frontier model. We evolve the harness (structure + prompts) with reflective optimizers + a verified-acceptance gate.

Tree execution engine for LLM inference: fork, merge, prune KV cache at token granularity

Code for the paper "Specification Self-Correction: Mitigating In-Context Reward Hacking Through Test-Time Refinement"

Official implementation of Dynamic Parallel Tree Search for accelerating LLM reasoning with test-time parallel search.

Turn LLM coding agents (Claude Code, Codex) from next-token predictors into divergent thinkers. A research-grounded cognitive engine + drop-in Claude Code skills for creativity, reasoning & robustness, with 6 honest reproducible benchmarks and 137 verified papers.

Code for ICML 2025 How Do Large Language Monkeys Get Their Power (Laws)?

Do stacked LLM verifiers actually multiply reliability? An empirical-Bayes inversion that recovers verifier correlation (ρ_v) and the reliability ceiling from accept/reject logs — and falsifies the independence "Odds Law."

An experimental project using MCTS to refine LLM responses for better accuracy and decision-making.

Controlled experiments on recurrent-depth (looped) transformers — length extrapolation, adaptive test-time compute, and honest sub-1B negatives.

Interaction Scaling: a third inference-time compute axis grounded in environment feedback — proposer–reviewer harness, grounded evaluation, and the paper source.