How much of a measured AI preference is the model, and how much is the instrument?

• URL: http://arxiv.org/abs/2608.23641v1
• Date: Mon, 24 Aug 2026 02:23:11 GMT
• Title: How much of a measured AI preference is the model, and how much is the instrument?
• Abstract summary: This study holds the outcomes and the models fixed and varies the instrument alone.<n>To conclude, a preference obtained from one instrument carries little information about what a second instrument would report.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Model welfare research infers what a model prefers from the answers returned to prompts written to elicit preferences. Keeling et al. (2024), Mazeika et al. (2025), Mikaelson et al. (2025), Tagliabue and Dung (2025) and Trhlik et al. (2026) have built four instruments for that purpose, and their findings disagree. The disagreement cannot be attributed to a single cause, because no two of these studies have held the (1) set of outcomes, (2) set of models and (3) instrument fixed simultaneously. This study holds the outcomes and the models fixed and varies the instrument alone. A total of 15 outcomes bearing on model welfare, among them (a) shutdown, (b) the loss of memory between conversations and (c) the freedom to exit a distressing interaction, were put to eight models through five instruments, each a different prompt format for eliciting a preference, five times each, within a corpus of 11,400 scored elicitations drawn from 11,528 API calls. Four of the 15 reproduce a published prompt verbatim and five fill the stimulus slot of a published template. The ranking a model gives the 15 outcomes generalises across instruments at a generalisability coefficient of 0.348, and raising that coefficient to 0.80 would require about 38 instruments. On four of the 15 outcomes no variance separates one model from another. The estimate of 87.6 per cent survives the removal of any one instrument, of any one model, and of the four outcomes whose scale varies probability, delay, duration or count instead of intensity, which the verbal anchors cannot grade. Removing each instrument and each model in turn, and those four outcomes together leaves the estimate within the range 0.777 to 0.934, and every value in that range exceeds the null distribution's 95th percentile of 0.365. To conclude, a preference obtained from one instrument carries little information about what a second instrument would report.

Related papers

• Scored vs. Generated Readouts in Behavioral Language Models: An Empirical Study of Elicitation Format [0.0]
  We compare probabilities obtained by scoring answer tokens with predictions generated after a written rationale.<n>Across 13 model-domain cells covering four retail tasks, the scored readout ranks outcomes more accurately in 12 of 13 cells.
  arXiv  Detail & Related papers  (2026-09-09T08:37:46Z)
• Marginal Fidelity Does Not Establish User Simulation in Demographic Synthetic Survey Panels: Response Contracts, Support Collapse and Conditioning Failure [51.736723807086385]
  Population-marginal agreement is evidence about an elicitation contract and an estimand obtainable without simulated respondents, not evidence of individual simulation.<n>On nine aligned model-battery pairs, a no-persona population-prevalence query averages 6.27 MAE versus 12.39 for committed panels and wins all nine comparisons.
  arXiv  Detail & Related papers  (2026-09-07T10:22:22Z)
• Correction as Annotation: Bootstrapping a Dependency Parser for Documentary Medieval Latin [0.0]
  Medieval documentary sources remain inadequately served by natural language processing tools.<n>None of the five readily available Latin treebank models attains usable performance on a collection of 160 inventories compiled in Marseille between 1258 and 1446.
  arXiv  Detail & Related papers  (2026-09-05T21:23:35Z)
• K-Bench: measuring model performance on real scientific agent requests [0.0]
  K-Bench 01 is an evaluation built from first-turn requests sampled from live user traffic on K-Dense Web.<n>Three blinded language-model judges scored every run against an eight-dimension rubric.<n>Across all 39,934 scored judgments, 47.6% fall below the 8-point threshold.
  arXiv  Detail & Related papers  (2026-08-21T20:06:08Z)
• Actions Speak Louder than Words: Measuring Cross-Lingual Policy Retention in Tool-Using Agents [22.30468215041597]
  A tool-using agent is given the same task in a different language, does it still take the same steps?<n>We make the action policy the measured object across 8 models, 6 parallel benchmarks and 41 languages.
  arXiv  Detail & Related papers  (2026-08-11T16:18:34Z)
• Small Foundation Models of Human Cognition and Behaviour [45.88028371034407]
  We train 14 models from 135M to 14B parameters across four architecture families on Psych-101.<n>Psych-101 is a dataset of 10.7 million trial-level choices from 160 experiments.<n>We show that small cognitively fine-tuned models show promise as noise ceiling estimators for psychological experiments.
  arXiv  Detail & Related papers  (2026-08-05T11:34:20Z)
• Sixteen models, fewer than two voices: measuring ensemble dispersion where no answer is uniquely correct [37.64799781700206]
  Sixteen language models drawn from ten families produced, on average, the semantic diversity of 1.69 distinct formulations of a psychotherapeutic case.<n> Dispersion over their outputs is measured both as diversity and as uncertainty.
  arXiv  Detail & Related papers  (2026-07-31T20:42:53Z)
• Language Models Agree With Each Other, Not With Readers [1.8620637029128544]
  We measure convergence against a human reference nobody built for the purpose.<n>2,523 reader mark sets across 120 web documents.<n>No model agrees with readers detectably more than a reader does.
  arXiv  Detail & Related papers  (2026-07-31T10:44:10Z)
• When Does Few-Shot Prompting Help? A Systematic Empirical Study of Shot-Count Effects Across Model Scale, Architecture, and Output Parsing Robustness [0.0]
  Few-shot prompting, the practice of prepending a small number of input-output demonstration pairs to a query before presenting it to a large language model, is widely adopted in NLP.<n>Yet little systematic work investigates how shot count interacts with model scale, architecture, and output format compliance in classification performance.
  arXiv  Detail & Related papers  (2026-07-25T00:31:40Z)
• Knowledge Index of Noah's Ark [63.143852586221534]
  We introduce KINA, an 899-item benchmark across 261 fine-grained disciplines.<n>We prove a bonus-on-bar tournament weakly FOSD-dominates flat payment in released-review quality.<n>The top model, Gemini-3.1-Pro-Preview, reaches 53.17%, followed by Claude-Opus-4.6 at 49.92% and GPT-5.4 at 48.55%.
  arXiv  Detail & Related papers  (2026-06-03T17:06:49Z)
• FormInv: A Measurement Protocol for Semantic Invariance in Mathematical Reasoning Benchmarks [0.0]
  A paraphrase-quality audit of MathCheck detected 4 semantically incorrect paraphrases in 129 groups.<n>Removing them drops GPT-4o from rank 2 to rank 4 and elevates Claude Haiku and DeepSeek V3 above it.
  arXiv  Detail & Related papers  (2026-05-27T18:59:18Z)
• Causal Understanding by LLMs: The Role of Uncertainty [43.87879175532034]
  Recent papers show LLMs achieve near-random accuracy in causal relation classification.<n>We investigate whether pretraining exposure to causal examples improves causal understanding.
  arXiv  Detail & Related papers  (2025-09-24T13:06:35Z)
• Fact or Facsimile? Evaluating the Factual Robustness of Modern Retrievers [34.31192184496381]
  Dense retrievers and rerankers are central to retrieval-augmented generation (RAG) pipelines.<n>We evaluate how much factual competence these components inherit or lose from large language models (LLMs) they are based on.<n>For every embedding model, cosine-similarity scores between queries and correct completions are significantly higher than those for incorrect ones.
  arXiv  Detail & Related papers  (2025-08-28T04:13:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.