REALM: Reliable Expertise-Aware Language Model Fine-Tuning from Noisy Annotations

• URL: http://arxiv.org/abs/2604.17289v1
• Date: Sun, 19 Apr 2026 07:06:35 GMT
• Title: REALM: Reliable Expertise-Aware Language Model Fine-Tuning from Noisy Annotations
• Abstract summary: We propose REALM, a method that jointly learns the model parameters and a scalar expertise value for each annotator entirely unsupervised.<n>We evaluate on five question answering benchmarks, fine-tuning three sizes of Flan-T5 under simulated noisy annotations.<n>The proposed algorithm consistently outperforms the naive noisy SFT in the large majority of single- and multi-task settings.
• Score: 13.49289449502791
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Supervised fine-tuning of large language models relies on human-annotated data, yet annotation pipelines routinely involve multiple crowdworkers of heterogeneous expertise. Standard practice aggregates labels via majority vote or simple averaging, discarding annotator identity and causing the model to absorb the errors of unreliable annotators directly into its parameters. We propose REALM, a method that jointly learns the model parameters and a scalar expertise value for each annotator entirely unsupervised, requiring no supervision beyond annotator identity. The key idea is to model each observed label as a mixture between the model's prediction and a uniform random guess, weighted by the annotator's learned expertise. We extend REALM to a multi-task setting via a learned expertise matrix that captures per-annotator reliability across tasks. We evaluate on five question answering benchmarks, fine-tuning three sizes of Flan-T5 under simulated noisy annotations. The proposed algorithm consistently outperforms the naive noisy SFT in the large majority of single- and multi-task settings, across datasets, model sizes, and noise types, with accuracy improvements of up to $50\%$ in the most adversarial regime and gains that grow with model capacity.

Related papers

• The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Prediction [39.10305114526259]
  We introduce a framework combining conformal prediction with Collaborative Filtering-style annotators' representation.<n>Using Non-Conformity Scores, we introduce the Ghost Prediction metric and the Ghost Annotator representation to quantify cases in which model predictions diverge from all available human annotations.<n>Our finding shows that while all models uncertainty increases with annotator disagreement, larger models tend to be more confident in the classification of texts that are not aligned with any human annotation.
  arXiv  Detail & Related papers  (2026-06-01T21:32:37Z)
• More Human, More Efficient: Aligning Annotations with Quantized SLMs [7.699569505613514]
  Large Language Model (LLM) capabilities have outpaced human capacity, leading to the widespread adoption of LLMs in automatic evaluation and annotation.<n>Our work examines the viability of finetuning a quantized Small Language Model of 1.7B parameter size on limited human-annotated data to serve as a highly aligned, deterministic evaluator and annotator.
  arXiv  Detail & Related papers  (2026-04-01T07:46:34Z)
• QuMAB: Query-based Multi-Annotator Behavior Modeling with Reliability under Sparse Labels [23.555446749682467]
  Multi-annotator learning traditionally aggregates diverse annotations to approximate a single ground truth, treating disagreements as noise.<n>We introduce a paradigm shift from sample-wise aggregation to annotator-wise behavior modeling.<n>By treating annotator disagreements as valuable information rather than noise, modeling annotator-specific behavior patterns can reconstruct unlabeled data to reduce annotation cost, enhance aggregation reliability, and explain annotator decision behavior.
  arXiv  Detail & Related papers  (2025-07-23T16:17:43Z)
• Model Utility Law: Evaluating LLMs beyond Performance through Mechanism Interpretable Metric [99.56567010306807]
  Large Language Models (LLMs) have become indispensable across academia, industry, and daily applications.<n>One core challenge of evaluation in the large language model (LLM) era is the generalization issue.<n>We propose Model Utilization Index (MUI), a mechanism interpretability enhanced metric that complements traditional performance scores.
  arXiv  Detail & Related papers  (2025-04-10T04:09:47Z)
• MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Models [70.92847554971065]
  We introduce MT-Eval, a comprehensive benchmark designed to evaluate multi-turn conversational abilities. By analyzing human-LLM conversations, we categorize interaction patterns into four types: recollection, expansion, refinement, and follow-up. Our evaluation of 11 well-known LLMs shows that while closed-source models generally surpass open-source ones, certain open-source models exceed GPT-3.5-Turbo in specific tasks.
  arXiv  Detail & Related papers  (2024-01-30T04:50:28Z)
• Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowds [88.06545572893455]
  In real-world crowd-sourcing scenarios, noise transition matrices are both annotator- and instance-dependent. We first model the mixture of noise patterns by all annotators, and then transfer this modeling to individual annotators. Experiments confirm the superiority of the proposed approach on synthetic and real-world crowd-sourcing data.
  arXiv  Detail & Related papers  (2023-06-05T13:43:29Z)
• Model ensemble instead of prompt fusion: a sample-specific knowledge transfer method for few-shot prompt tuning [85.55727213502402]
  We focus on improving the few-shot performance of prompt tuning by transferring knowledge from soft prompts of source tasks. We propose Sample-specific Ensemble of Source Models (SESoM) SESoM learns to adjust the contribution of each source model for each target sample separately when ensembling source model outputs.
  arXiv  Detail & Related papers  (2022-10-23T01:33:16Z)
• Meta-Learned Confidence for Few-shot Learning [60.6086305523402]
  A popular transductive inference technique for few-shot metric-based approaches, is to update the prototype of each class with the mean of the most confident query examples. We propose to meta-learn the confidence for each query sample, to assign optimal weights to unlabeled queries. We validate our few-shot learning model with meta-learned confidence on four benchmark datasets.
  arXiv  Detail & Related papers  (2020-02-27T10:22:17Z)
• AvgOut: A Simple Output-Probability Measure to Eliminate Dull Responses [97.50616524350123]
  We build dialogue models that are dynamically aware of what utterances or tokens are dull without any feature-engineering. The first model, MinAvgOut, directly maximizes the diversity score through the output distributions of each batch. The second model, Label Fine-Tuning (LFT), prepends to the source sequence a label continuously scaled by the diversity score to control the diversity level. The third model, RL, adopts Reinforcement Learning and treats the diversity score as a reward signal.
  arXiv  Detail & Related papers  (2020-01-15T18:32:06Z)

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.