7 projects
honestcode
The deterministic verification layer for AI coding agents. Catch invented APIs, undefined symbols, and wrong calls before your agent moves on.
cotstruct
Chain-of-thought reasoning topology analysis toolkit — deterministic, LLM-free.
physicausal
Causal World Models for Physical Reasoning
sigx
Implicit Feedback Signal Extraction for LLM Alignment
lookagain
LookAgain: VLM Reliability Auditor -- test if your VLM truly relies on visual evidence.
rtsa
Reasoning Trace Structure Analysis Toolkit — extract, analyze, prune, fingerprint, and benchmark CoT reasoning graphs
ib-components
Independent building blocks for Information Bottleneck research: VIB compression, Riemannian optimization, and architectural improvements
