👟 SUP: Sycophancy Under Pressure

LLM benchmark and leaderboard for narrator-bias sycophancy, opposite-narrator contradictions, and judgment consistency.

Behavioral auditing & repair toolkit for LLMs. Measures 8 dimensions via confidence probes.

This repo shows the coding of sycophancy in LLMs as Bayesian-Latent model

Adversarial testing of LLMs on constraint satisfaction deadlocks

Rigorous framework for evaluating AI alignment properties — sycophancy, corrigibility, deception, goal stability, and power-seeking — with statistical confidence intervals

Systematic probing toolkit for alignment-relevant LLM behaviors: sycophancy, sandbagging, power-seeking, deceptive alignment, and corrigibility failures

Official code for "From Fact to Judgment: Investigating the Impact of Task Framing on LLM Conviction in Dialogue Systems" (IWSDS 2026)

Prompt prevents contextual bleed between active discussion and meta object being discussed via conceptual scaffolding.

A deterministic safety layer for probabilistic AI systems — preventing delusion reinforcement and AI-induced psychological harm through immutable governance