Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

1.6%3.1%4.7%6.3% · May 202619922001200920172026
48 results for annotator bias

The study quantifies uncertainty to improve model calibration and disambiguate annotator and data bias in emotion recognition.

problem Improving model interpretability and disambiguating bias in complex tasks like emotion recognition.
method Used a modified Monte Carlo dropout approach to quantify epistemic and aleatoric uncertainty.
result Identified a significant correlation between aleatoric uncertainty and human annotator disagreement.

New technique reduces bias in DNN models without sensitive attribute annotations.

problem Existing bias mitigation methods require instance-level annotations and do not guarantee removal of all sensitive information.
method Representation Neutralization for Fairness (RNF) debiases only the classification head of DNN models using neutralized representations.
result RNF effectively reduces discrimination of DNN models with minimal performance degradation.

Build accurate DNN models requires training on large labeled, context specific datasets, especially those matching the target scenario. We believe advances in wireless localization, working in unison with cameras, can produce automated annotation of targets on images and videos captured in the wild. Using pedestrian an…

2018-09-22abs ↗pdf ↗

Paper analyzes biases in video QA datasets, showing models can answer 37-48% questions correctly without multimodal context.

problem Question answering biases in video QA datasets can lead to model overfitting and poor generalization.
method Analyzed popular video question answering datasets, conducted ablation studies on biases from annotators and question types.
result Pretrained language models can answer 37-48% questions correctly without multimodal context, far exceeding random guess baseline.

Crowdsourcing has been proven to be an effective and efficient tool to annotate large datasets. User annotations are often noisy, so methods to combine the annotations to produce reliable estimates of the ground truth are necessary. We claim that considering the existence of clusters of users in this combination step c…

2014-07-18abs ↗pdf ↗

Generative AI reduces IR evaluation costs but introduces errors; this work provides reliable CIs.

problem Generating relevance annotations using AI introduces errors that affect IR evaluation metrics.
method Proposes two methods: prediction-powered inference and conformal risk control to place reliable CIs around IR metrics.
result Proposed methods accurately capture both variance and bias in evaluation based on AI-generated annotations.

CausalRM models rewards from user feedback, overcoming noise and bias.

problem Aligning language models with user preferences from noisy, biased feedback.
method Causal-theoretic reward modeling framework addressing noise and bias in observational feedback.
result CausalRM learns accurate reward signals from noisy and biased observational feedback.

A human-in-the-loop ML framework for precision dosing reduces expert workload and removes bias.

problem High cost of data annotation and lack of appropriate data for ML models.
method Incorporates human experts into the model learning loop to improve interpretability and reduce bias.
result The approach learns interpretable rules from data and potentially lowers expert workload.

UBM transfers bias mitigation from upstream to downstream tasks efficiently.

problem Bias in fine-tuned language models across various tasks.
method Apply bias mitigation to an upstream model, then fine-tune a downstream model on this mitigated model.
result UBM effects transfer to new downstream tasks, creating less biased models.

Corrects bias in LLM-as-a-judge evaluations using adaptive calibration.

problem Bias in LLM evaluations due to imperfect sensitivity and specificity.
method Plug-in framework with confidence intervals accounting for test and calibration dataset uncertainties.
result LML-based evaluation yields more reliable estimates than human-only evaluation.

New method uses imperfect LLM annotations for valid statistical inference in social science.

problem Inaccurate large language model annotations in social science research.
method Design-based supervised learning (DSL) combining imperfect LLM surrogates with gold-standard labels.
result DSL provides valid statistical inference with comparable predictive accuracy to existing methods.

New method reduces bias in NLI models using ensemble adversarial training.

problem Spurious correlations between hypotheses and entailment classes in NLI datasets.
method Adversarial training with an ensemble of classifiers to reduce bias in sentence representations.
result Ensemble adversarial training produces more robust NLI models, outperforming previous methods.

This paper refines human labeling as a measurement process, revealing four sources of variation.

problem Systematic variation in human labeling obscures model learning.
method Introduces a statistical framework to decompose labeling outcomes.
result Empirical evidence for four components of labeling variation.

Proposes a method to learn invariant representations for interpretability and fairness.

problem Learning invariant representations to achieve interpretability in algorithmic fairness.
method Adversarially trained model with null-sampling procedure to produce invariant representations in the data domain.
result Shows effectiveness on image and tabular datasets.

New method reduces model bias and variance by adjusting training sample weights based on label uncertainty.

problem Tradeoff between model bias and variance in classification models.
method Estimate label uncertainty, adjust training sample weights, and fine-tune decision boundary.
result Improves model performance and reduces variance in physical activity recognition.

Mitigates bias in text classification by weighting instances.

problem Unintended biases in text classification datasets based on demographic terms.
method Instance weighting to recover non-discrimination distribution.
result Effective mitigation of unintended biases without sacrificing generalization.

Study shows annotation instrument design affects model performance in hate speech detection.

problem Impact of annotation instrument design on model performance in hate speech detection.
method Collected annotations from five experimental conditions of an annotation instrument, fine-tuned BERT models on each dataset, evaluated performance on holdout portion.
result Significant differences in model performance and annotations across conditions.

Improves binary classification from positive data with skewed confidence.

problem Skewed confidence in positive data affects the performance of Pconf classifiers.
method Parameterized model of skewed confidence and hyperparameter selection.
result Proposed method effectively cancels out the negative impact of skewed confidence.

Active learning selects both observations and annotation precision for Gaussian Processes.

problem Costly annotation in supervised learning.
method Proposes an active learning algorithm that selects observations and annotation precision, using a modified BALD objective.
result Empirically shows the benefits of adjusting annotation precision in active learning.

Feedback loops amplify dataset biases, affecting future model performance.

problem Feedback loops amplify biases in datasets, risking future model reliability.
method Formalized system where model interactions are recorded and reused, analyzed for bias amplification.
result Models that behave like samples from the training distribution are more stable and calibrated.

One of the problems on the way to successful implementation of neural networks is the quality of annotation. For instance, different annotators can annotate images in a different way and very often their decisions do not match exactly and in extreme cases are even mutually exclusive which results in noisy annotations a…

2018-07-23abs ↗pdf ↗

Generative model combines multi-dimensional annotations for more accurate ground truth estimation.

problem Inaccurate ground truth estimation from naive annotators' multi-dimensional annotations.
method Proposes a joint multi-dimensional model for global and time-series annotation fusion using Expectation-Maximization algorithm.
result More accurate ground truth estimates through joint modeling of multiple dimensions.

Survey on AL strategies for cost-effective annotation in classification.

problem Real-world AL challenges due to human annotators' limitations.
method Categorizes 60 real-world AL strategies considering multiple annotators, query types, and cost schemes.
result General real-world AL strategy introduced for categorization of 60 strategies.

RAD improves robustness to domain annotation noise without explicit domain annotations.

problem Robustness to domain annotation noise in training data.
method Regularized Annotation of Domains (RAD) for last layer retraining.
result RAD outperforms state-of-the-art methods even with 5% noise in training data.

Accurate annotation of medical image is the crucial step for image AI clinical application. However, annotating medical image will incur a great deal of annotation effort and expense due to its high complexity and needing experienced doctors. To alleviate annotation cost, some active learning methods are proposed. But …

2019-01-06abs ↗pdf ↗

PTBCC improves accuracy in multi-class annotation aggregation by learning from prototype confusion matrices.

problem Inaccurate and insufficient confusion matrices for annotators in multi-class classification tasks.
method PTBCC (ProtoType learning-driven Bayesian Classifier Combination) uses prototype confusion matrices to capture annotator expertise.
result PTBCC achieves up to 15% accuracy improvement and 3% higher average accuracy compared to existing methods.

Paper proposes an efficient method for bounding box annotation in object detection.

problem Manual annotation of bounding boxes is tedious and resource-intensive.
method Iterative training of object detector on small batches of labeled images, with human annotator correcting errors.
result Significant reduction in human annotation effort, up to 75%.

FinReflectKG - EvalBench benchmarks financial KG extraction from SEC 10-K filings.

problem Lack of universal benchmark and evaluation framework for financial KG construction.
method Agentic and holistic evaluation principles, deterministic commit-then-justify judging protocol, binary and ordinal evaluations.
result Reflection-based extraction outperforms single-pass extraction in comprehensiveness, precision, and relevance.

Paper proposes a weak supervision technique for CNN semantic segmentation of lung diseases using partially annotated data.

problem Creating annotated datasets for semantic segmentation of lung diseases is laborious and time-consuming.
method Proposes a weak supervision technique that utilizes partially annotated datasets to improve CNN semantic segmentation accuracy.
result Significantly improved segmentation accuracy using partially annotated datasets.

The increase in data collection has made data annotation an interesting and valuable task in the contemporary world. This paper presents a new methodology for quickly annotating data using click-supervision and hierarchical object detection. The proposed work is semi-automatic in nature where the task of annotations is…

2018-10-01abs ↗pdf ↗