Pith. sign in

REVIEW 4 cited by

Eliciting the Priors of Large Language Models using Iterated In-Context Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.01860 v1 pith:7UXFCSKD submitted 2024-06-04 cs.CL

Eliciting the Priors of Large Language Models using Iterated In-Context Learning

classification cs.CL
keywords priorslearningiteratedmethodpriorsettingsdistributionseliciting
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

As Large Language Models (LLMs) are increasingly deployed in real-world settings, understanding the knowledge they implicitly use when making decisions is critical. One way to capture this knowledge is in the form of Bayesian prior distributions. We develop a prompt-based workflow for eliciting prior distributions from LLMs. Our approach is based on iterated learning, a Markov chain Monte Carlo method in which successive inferences are chained in a way that supports sampling from the prior distribution. We validated our method in settings where iterated learning has previously been used to estimate the priors of human participants -- causal learning, proportion estimation, and predicting everyday quantities. We found that priors elicited from GPT-4 qualitatively align with human priors in these settings. We then used the same method to elicit priors from GPT-4 for a variety of speculative events, such as the timing of the development of superhuman AI.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior

    cs.AI 2026-06 conditional novelty 7.0

    Behavioral profiles from prediction-market traders are partially stable and identifiable but cannot be transmitted via prompts to reduce LLM forecast correlations or improve Brier scores.

  2. Reconstruction of Personally Identifiable Information from Supervised Finetuned Models

    cs.CR 2026-05 unverdicted novelty 7.0

    PII can be reconstructed from SFT models via prefix attacks, with the new COVA algorithm improving success rates and leakage varying by attacker knowledge and PII type.

  3. From `May' to `Is': Certainty Distortion in Language Model Rewriting

    cs.CL 2026-06 unverdicted novelty 6.0

    LMs systematically inflate expressed certainty during rewriting, affecting up to 75% of outputs with a 1.5-2x bias toward increasing rather than decreasing certainty, and the effect compounds over iterations.

  4. LILO: Bayesian Optimization with Natural Language Feedback

    cs.LG 2025-10 unverdicted novelty 6.0

    LILO integrates LLMs to translate natural language feedback into preference signals for Gaussian process-based Bayesian optimization, outperforming standard preference BO and LLM-only methods on benchmarks.