mlsanityData to 14 Oct 2026

Aligned Textual Scoring Rules

arXiv 2507.06221 ↗First posted 8 Jul 2025cs.AI cs.GTcited by 2

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

Scoring rules elicit probabilistic predictions from a strategic agent by scoring the prediction against a ground truth state. A scoring rule is proper if, from the agent's perspective, reporting the true belief maximizes the expected score. With the development of language models, Wu and Hartline (2024) proposes a reduction from textual information elicitation to the numerical (i.e. probabilistic) information elicitation problem, which achieves provable properness for textual elicitation. However, not all proper scoring rules are well aligned with human preference over text. Our paper designs the Aligned Scoring rule (ASR) for text by optimizing and minimizing the mean squared error between a proper scoring rule and a reference score (e.g. human score). Our experiments show that our ASR outperforms previous methods in aligning with human preference while maintaining properness.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 26 of its 47 references that are in the map. First posted on arXiv 8 Jul 2025.

Builds on

26 of 47 references.

References · 26

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  3. 2019BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questionscited by 2,183
  4. 2023GPTScore: Evaluate as You Desirecited by 361
  5. 2020From Arguments to Key Points: Towards Automatic Argument Summarizationcited by 61
  6. 2017Optimal Auctions through Deep Learning: Advances in Differentiable Economicscited by 38
  7. 2017Dispersion for Data-Driven Algorithm Design, Online Learning, and Private Optimizationcited by 37
  8. 2023Is GPT-4 a reliable rater? Evaluating Consistency in GPT-4 Text Ratingscited by 34
  9. 2018Automated Mechanism Design via Neural Networkscited by 32
  10. 2020Optimization of Scoring Rulescited by 28
  11. 2016Sample Complexity of Automated Mechanism Designcited by 24
  12. 2023A Scalable Neural Network for DSIC Affine Maximizer Auction Designcited by 24
14 more
  1. 2016Learning Simple Auctionscited by 23
  2. 2022Differentiable Economics for Randomized Affine Maximizer Auctionscited by 22
  3. 2023Towards LLM-based Autograding for Short Textual Answerscited by 20
  4. 2024Eliciting Informative Text Evaluations with Large Language Modelscited by 15
  5. 2024GemNet: Menu-Based, Strategy-Proof Multi-Bidder Auctions Through Deep Learningcited by 14
  6. 2015Ironing in the Darkcited by 13
  7. 2020Binary Scoring Rules that Incentivize Precisioncited by 12
  8. 2022Optimal Scoring Rules for Multi-dimensional Effortcited by 12
  9. 2024ElicitationGPT: Text Elicitation Mechanisms via Language Modelscited by 9
  10. 2022Contracts with Information Acquisition, via Scoring Rulescited by 8
  11. 2014Combinatorial Auctions via Posted Pricescited by 7
  12. 2023Learning to Incentivize Information Acquisition: Proper Scoring Rules Meet Principal-Agent Modelcited by 7
  13. 2024Automated Design of Affine Maximizer Mechanisms in Dynamic Settingscited by 7
  14. 2022Proper losses for discrete generative modelscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025Incentive-Aligned Multi-Source LLM Summariescited by 1
  2. 2026Scoring Rules! Statistical and Strategic Alignment for Text Evaluation Metricscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Alignment Makes Language Models Normative, Not Descriptive
  2. 2024A Probability--Quality Trade-off in Aligned Language Models and its Relation to Sampling Adaptors
  3. 2024Doing Experiments and Revising Rules with Natural Language and Probabilistic Reasoning
  4. 2026From Constitutions to Control: Interpretable Rewards for Aligning Language Models
  5. 2025LLMs Judge Themselves: A Game-Theoretic Framework for Human-Aligned Evaluation
  6. 2023Human Choice Prediction in Language-based Persuasion Games: Simulation-based Off-Policy Evaluation
  7. 2025Improving LLM-as-a-Judge Inference with the Judgment Distribution
  8. 2026PADMÉ: Preference Alignment Data Synthesis for Meta-Evaluation of LM Agent Evaluators