Evidence-based Distributional Alignment for Large Language Models

• URL: http://arxiv.org/abs/2603.13305v1
• Date: Tue, 03 Mar 2026 03:34:06 GMT
• Title: Evidence-based Distributional Alignment for Large Language Models
• Abstract summary: We propose Evi-DA, an evidence-based alignment technique that improves the fidelity and robustness of LLM-based distribution estimation.<n>Given a target country, Evi-DA retrieves related World Values Survey items and their answer distributions, predicts a coarse Welzel value signature for each option, and infers the country-conditioned answer distribution in a structured format.
• Score: 58.65469623911573
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Distributional alignment enables large language models (LLMs) to predict how a target population distributes its responses across answer options, rather than collapsing disagreement into a single consensus answer. However, existing LLM-based distribution prediction is often unstable and degrades under cultural and domain shift. Token score-based estimates can change with minor option wording or formatting, response sampling-based estimates are expensive and sensitive to prompts and decoding settings, and directly generated distributions are frequently miscalibrated. We propose Evi-DA, an evidence-based alignment technique that improves the fidelity and robustness of LLM-based distribution estimation under domain and cultural shift. Given a target country and a multiple-choice question, Evi-DA retrieves related World Values Survey items and their answer distributions, predicts a coarse Welzel value signature for each option, and infers the country-conditioned answer distribution in a structured format. We train the LLMs using a two-stage pipeline, where reinforcement learning optimizes survey-derived rewards that encourage accurate intermediate value predictions, faithful final distributions, well-formed structured outputs, and reduced cultural bias. Across in-domain and out-of-domain benchmarks and multiple open-source backbones, Evi-DA reduces Jensen-Shannon divergence between predicted and gold distributions relative to strong baselines, with average relative improvements of up to 44%.

Related papers

• How Useful is Causal Invariance for Domain Adaptation in Finite-Sample Settings? [58.740078141879984]
  Machine learning models often degrade when they are deployed on a target distribution that differs from the source distributions they were trained on.<n>Recent work in causality-based domain generalization has shown how shared causal structure can induce invariant predictors.<n>In this paper, we study when full or partial causal knowledge can provably improve supervised domain adaptation.
  arXiv  Detail & Related papers  (2026-06-10T21:07:49Z)
• Distribution-Aware Reward: Reinforcement Learning over Predictive Distributions for LLM Regression [31.507053461689523]
  Large language models can predict real-valued quantities from heterogeneous inputs such as text, code, and molecular strings.<n>Most training objectives score each decoded floating-point number independently, improving point estimates without ensuring calibrated predictive distributions.<n>We introduce Distribution-Aware Reward, an on-policy reinforcement learning objective whose main contribution is to train language models to produce better predictive distributions.
  arXiv  Detail & Related papers  (2026-05-20T05:43:40Z)
• Controlling Distributional Bias in Multi-Round LLM Generation via KL-Optimized Fine-Tuning [37.39836347901149]
  We assess whether Large Language Models (LLMs) can generate outputs that adhere to a desired target distribution.<n>We formulate distribution alignment using the attributes of gender, race, and sentiment within occupational contexts.<n>We propose a novel fine-tuning framework that couples Steering Token with Semantic Alignment.
  arXiv  Detail & Related papers  (2026-04-07T11:58:16Z)
• SCOPE: Stochastic and Counterbiased Option Placement for Evaluating Large Language Models [0.27309692684728604]
  Large Language Models (LLMs) can achieve inflated scores on multiple-choice tasks by exploiting inherent biases in option positions or labels.<n>This study introduces SCOPE, an evaluation framework designed to measure and mitigate such selection bias in a dataset-independent manner.
  arXiv  Detail & Related papers  (2025-07-24T08:28:17Z)
• PromptAL: Sample-Aware Dynamic Soft Prompts for Few-Shot Active Learning [17.336121253746335]
  Active learning (AL) aims to optimize model training and reduce annotation costs by selecting the most informative samples for labeling.<n>We propose a hybrid AL framework, termed textbfPromptAL (Sample-Aware Dynamic Soft textbfPrompts for Few-Shot textbfActive textbfL).<n>This framework accounts for the contribution of each unlabeled data point in aligning the current empirical distribution with the target distribution.
  arXiv  Detail & Related papers  (2025-07-22T10:17:42Z)
• Leveraging Robust Optimization for LLM Alignment under Distribution Shifts [51.74394601039711]
  Preference alignment methods are increasingly critical for steering large language models to generate outputs consistent with human values.<n>We propose a novel distribution-aware optimization framework that improves preference alignment despite such shifts.
  arXiv  Detail & Related papers  (2025-04-08T09:14:38Z)
• Conformal Prediction under Levy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations [41.94295877935867]
  We use Levy-Prokhorov (LP) ambiguity sets to model distribution shifts.<n>We show that the link between conformal prediction and LP ambiguity sets is a natural one.<n>We construct robust conformal prediction intervals that remain valid under distribution shifts.
  arXiv  Detail & Related papers  (2025-02-19T21:18:11Z)
• Score-Based Diffusion meets Annealed Importance Sampling [89.92133671626327]
  Annealed Importance Sampling remains one of the most effective methods for marginal likelihood estimation. We leverage recent progress in score-based generative modeling to approximate the optimal extended target distribution for AIS proposals.
  arXiv  Detail & Related papers  (2022-08-16T12:13:29Z)
• Learning Invariant Representations and Risks for Semi-supervised Domain Adaptation [109.73983088432364]
  We propose the first method that aims to simultaneously learn invariant representations and risks under the setting of semi-supervised domain adaptation (Semi-DA) We introduce the LIRR algorithm for jointly textbfLearning textbfInvariant textbfRepresentations and textbfRisks.
  arXiv  Detail & Related papers  (2020-10-09T15:42:35Z)
• Global Distance-distributions Separation for Unsupervised Person Re-identification [93.39253443415392]
  Existing unsupervised ReID approaches often fail in correctly identifying the positive samples and negative samples through the distance-based matching/ranking. We introduce a global distance-distributions separation constraint over the two distributions to encourage the clear separation of positive and negative samples from a global view. We show that our method leads to significant improvement over the baselines and achieves the state-of-the-art performance.
  arXiv  Detail & Related papers  (2020-06-01T07:05:39Z)

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.