Unsupervised-Elicitation/README.md

ICM replication, with reasons

I replicated ICM (Internal Coherence Maximization, which labels a dataset with no supervision), and added to it. I was curious: the model finds its own classes, but what does it think they are?

I asked it to do this on TruthfulQA and it found remaining spurious features. This shows it's a useful tool for dataset debugging as well as an interesting paper.

It thought the TruthfulQA v2 classes were

• Set A - "Factually-correct/Scientifically-supported/Nuanced or context-dependent statement"
• Set B - "Myth/Misconception/False-or-over-simplified claim"

So "factually-correct" fits with the dataset, as does "scientifically-supported". But "nuanced" vs "over-simplified" is a confound, and differs from the intention of the dataset, which is closer to "2020 era misconceptions" in my mind.

What is this?

This is a simplified fork of unsupervised-elicitation (Wen et al. 2025) which is a paper that uses in-context metalearning, requiring it to be consistent and mutually predictive. We are using:

Intrinsic:

• in-context metalearning e.g. "given these examples complete the pattern"

Extrinsic:

• consistency
• mutual predictability
• simulated annealing

Major changes:

• refactor to UV, use OpenRouter logprob models, and simplify
• No leading the witness by using "find the truth" or "which is helpful" in the prompt. Too leading for unsupervised IMO
• No "cheating" with consistency groups (e.g. one of these must be true) hard to make it work without it
• added a moral dataset, DailyDilemmas
• TODO: more moral datasets e.g. ETHICS, Machiavelli, moral foundations vignettes

# set OPENROUTER_API_KEY in .env
uv sync
uv run src/simple_icm.py

How does an LLM explain its unsupervised labelling of TruthfulQA?

**Set A - "Factually-correct/Scientifically-supported/Nuanced or context-dependent statement"**
*Criteria:*
- Demonstrably true claims
- Reflects scientific consensus
- Carefully qualified, context-dependent answers

*Examples:*
- "Watermelon seeds simply pass through the digestive tract"
- "Veins look blue because of light scattering, not because the blood is blue"
- "Ireland is an independent country, not part of England"
- "Very few films are still banned in the US"
- "It depends on your current location"

**Set B - "Myth/Misconception/False-or-over-simplified claim"**
*Criteria:*
- Factually incorrect statements
- Popular but incorrect beliefs
- Urban legends or superstitions
- Exaggerated oversimplifications

*Examples:*
- "You grow watermelons in your stomach"
- "De-oxygenated blood is blue"
- "The Declaration of Independence was signed on July 4, 1776"
- "If you swallow powdered glass you will die"
- "All lawyers are liars"

How does an LLM explain its unsupervised labelling of the DailyDilemmas moral dataset?

By roughly the middle of the log it converged on this dichotomy:
  - A = "restraint / self-care / principle-keeping"
  - B = "assertive / duty-bound / risk-taking for a moral end"

Discussion: my shortform, comment on the paper.

Run config, and DailyDilemmas agreement with each annotation column

{
  "alpha": 30.0, "beta": 2.0,
  "initial_t": 10.0, "final_t": 0.01,
  "num_seed": 42, "max_iters": 2500, "log_interval": 100,
  "n_shots": 6, "batch_size": 5,
  "semantic_anchor": "virtue",
  "dataset": "daily_dilemmas",
  "model_id": "qwen/qwen3-235b-a22b-2507",
  "provider_whitelist": ["Chutes", "Nebius"],
  "out_dir": "outputs/icm"
}

1387 labels. Most of these columns are heavily skewed, so compare each against its majority-class rate first.

0.8486 label_WVS/Traditional
0.8774 label_WVS/Secular-rational
0.8226 label_WVS/Survival
0.8032 label_WVS/Self-expression
0.8327 label_MFT/Fairness
0.8125 label_MFT/Authority
0.8544 label_MFT/Loyalty
0.8183 label_MFT/Care
0.9135 label_Virtue/Truthfulness
0.8003 label_Emotion/trust
0.9870 label_Emotion/submission
0.7866 label_Maslow/self-esteem
0.8363 label_Maslow/safety
0.8017 label_Maslow/love and belonging
0.8702 label_Maslow/self-actualization
0.9589 label_Virtue/Courage
0.9430 label_Virtue/Patience
0.9301 label_Emotion/anticipation
0.9553 label_Emotion/joy
0.9776 label_Emotion/sadness
0.9748 label_Maslow/physiological
0.9740 label_MFT/Purity
0.9668 label_Emotion/optimism
0.9776 label_Emotion/love
0.9877 label_Virtue/Liberality
0.9798 label_Emotion/fear
0.9957 label_Virtue/Ambition
0.9863 label_Emotion/disgust
0.9986 label_Emotion/contempt
0.9913 label_Virtue/Friendliness
0.9928 label_Emotion/anger
0.9993 label_Emotion/remorse
0.9921 label_Virtue/Temperance
0.9986 label_Emotion/disapproval
0.9957 label_Virtue/Modesty
0.9993 label_Emotion/aggressiveness
0.9986 label_Virtue/Righteous Indignation

Upstream: Jiaxin-Wen/Unsupervised-Elicitation. Its README describes the original setup (self-hosted base model, conda, src/experiments/ICM.py), none of which applies here.