#synthetic-data

20 curated events

papersTODAY 04:00 UTC

arXiv paper analyzes how AI-generated data affects dataset decomposition

A new preprint examines what happens to batch decomposition and downstream model performance when training sets mix human data with text or images produced by existing large language models. Using random datasets containing anomalies, the authors study criticality in dissimilar decomposition and undersampling techniques. The work aims to clarify the statistical behavior of datasets that are increasingly populated with synthetic samples.

research/training-data/synthetic-data/llm

papersTODAY 04:00 UTC

Auditing User-Level Privacy in Private Evolution Synthetic Data

A new arXiv paper examines how to audit user-level privacy guarantees in Private Evolution, a method for generating synthetic data in federated settings. The approach collects clipped user votes over a shared candidate bank and turns them into a differentially private histogram with calibrated noise. The work focuses on verifying that individual users' raw data remains protected under this mechanism.

privacy/differential-privacy/synthetic-data/federated-learning

papersTODAY 04:00 UTC

Paper Proposes Diagnostics for LLM-Based Synthetic Consumer Panels

A new arXiv preprint examines how large language models are used as stand-ins for survey respondents, a practice that can cut costs dramatically compared with traditional polling. The authors argue that aggregate validation scores hide systematic problems such as compressed variance and flipped coefficient signs, and they offer diagnostic and correction methods to address them.

synthetic-data/llm-evaluation/surveys/benchmarks

papersTODAY 04:00 UTC

Synthetic Data Method Targets Few-Shot Cryo-ET Subtomogram Classification

A new arXiv paper addresses the limited availability of labeled data for subtomogram classification in cryo-electron tomography. The authors propose a method to close the gap between simulated cryo-ET data and real experimental images, aiming to improve classification when only a few labeled examples exist. The approach is categorized under machine learning research.

cryo-et/few-shot-learning/synthetic-data/research

papersSEP 10 04:00 UTC

Paper examines distilling synthetic data for time series foundation models

A new arXiv preprint looks at how time series foundation models are pretrained on artificially generated trajectories where the underlying data-generating process is known. The work focuses on distillation methods rather than the conventional loss-based pretraining objectives that compare model outputs against targets. It aims to improve how these models learn from synthetic time series data.

time-series/foundation-models/synthetic-data/distillation

papersTODAY 04:00 UTC

Paper Proposes Framework for Judging When Synthetic Survey Data Is Trustworthy

A new arXiv paper argues that the debate over synthetic data in marketing research has been stuck between two extremes: treating large language models as a replacement for human survey respondents, or rejecting them outright. The authors say the more useful question is when synthetic respondents can be trusted, and they outline how that reliability should be evaluated. The work focuses on marketing research but touches on broader issues of validating model-generated data.

synthetic-data/research/llm/evaluation

papersTODAY 04:00 UTC

CodeTS Generates Time Series from Text via Executable Code

A new arXiv paper introduces CodeTS, a method that turns natural-language descriptions into time series by generating and running executable code rather than sampling outputs directly. This design makes the resulting synthetic data verifiable and suited to scenarios where real observations are scarce or expensive to collect. The work appears in the cs.LG and cs.AI listings.

research/time-series/code-generation/synthetic-data

papersTODAY 04:00 UTC

arXiv paper revisits disparate impact fairness metric for synthetic data generation

A revised arXiv preprint examines disparate impact as a fairness criterion for synthetic data generation, asking whether generated records deliver equal utility across sensitive demographic groups. The authors position their work as a departure from prior fair synthetic-data research, which they say addresses related but distinct fairness goals. The paper is a research contribution and does not announce any released model or tool.

fairness/synthetic-data/research/bias

papersSEP 11 04:00 UTC

arXiv Paper Proposes Classifier Reconstruction to Predict Synthetic Data Utility

A new arXiv preprint examines how well synthetic images help binary classification tasks where positive examples are scarce, as in medical imaging and industrial inspection. The authors propose measuring a "discriminative span" and reconstructing a classifier to predict how useful generated samples will be. The work aims to guide synthetic data selection in severely imbalanced settings.

synthetic-data/computer-vision/research/class-imbalance

papersSEP 11 04:00 UTC

PEARL Framework Evaluates Differentially Private Synthetic Educational Data

A new arXiv paper introduces PEARL, a task-aware framework for assessing differentially private synthetic data generated from learner records. The work targets personalized learning systems, where performance, behavioral, and demographic data are highly sensitive. PEARL aims to measure how well such synthetic data supports downstream educational tasks while preserving privacy.

differential-privacy/synthetic-data/education/evaluation

papersSEP 10 04:00 UTC

Verified Code World Models Proposed to Cheaply Scale LLM Domain Generalization

A new paper examines how large language models can generalize in domains that lack abundant real, labeled examples. By expressing a domain's dynamics as code, the authors show a single template can instantiate many simulated world models whose executions yield verified training data. The goal is to manufacture generalization examples cheaply where real-world annotation is scarce.

llm/synthetic-data/world-models/verification

papersSEP 10 04:00 UTC

Study Audits Subgroup Privacy Risks in Differentially Private Synthetic Text

A new paper introduces an auditing framework that runs membership inference attacks at the subgroup level against synthetic text produced under differential privacy. It explores whether formal worst-case privacy guarantees hold up in practice for smaller groups represented in the underlying data. The work offers data publishers a way to gauge real-world leakage before sharing synthetic text in place of sensitive datasets.

privacy/synthetic-data/differential-privacy/auditing

papersSEP 10 04:00 UTC

MultiSynt/MT: open synthetic corpus offers 4.8T tokens in 36 languages for multilingual pretraining

Researchers have introduced MultiSynt/MT, an open synthetic parallel dataset totaling roughly 4.8 trillion target-language tokens spanning 36 languages. Because web-scale training data is heavily skewed toward English, the corpus is intended to give model builders far more multilingual material for pretraining large language models. The work is described in a paper posted on arXiv.

multilingual/datasets/open-source/synthetic-data

papersSEP 10 04:00 UTC

Study analyzes SGD-based learning with synthetic data in high-dimensional linear regression

A newly cross-listed arXiv paper investigates how stochastic gradient descent behaves when training combines human-generated and synthetic data in a high-dimensional linear regression setting. It engages with prior work on model collapse, a phenomenon where keeping even a fixed share of synthetic samples stops model performance from improving as training scales. The findings aim to clarify the conditions under which synthetic data can genuinely extend training beyond limited human datasets.

synthetic-data/model-collapse/training/theory

papersSEP 10 04:00 UTC

MADS framework generates persuasion dialogue data via multi-agent self-play

Researchers introduced MADS, a scalable framework that produces multi-turn persuasive conversations through agent self-play. The setup uses three coordinated agents, including user agents that role-play varied persona-driven behaviors, to generate diverse dialogue datasets. The paper appears on arXiv with cross-listings in artificial intelligence and computational linguistics.

agents/synthetic-data/persuasion/research

papersSEP 10 04:00 UTC

EvolveScaler paper generates evolving-context data with executable state machines

A new arXiv preprint, EvolveScaler, addresses situations where newer events in a long interaction can override or invalidate statements made earlier. The authors build synthetic datasets of such shifting information by pairing executable state machines with natural-language rendering, yielding material that tests how well models track what remains valid over time. The approach is aimed at benchmarking and training systems that must reason over dynamically changing contexts rather than static records.

long-context/synthetic-data/reasoning/evaluation

papersSEP 10 04:00 UTC

Paper Revisits Whether Neural Networks Can Match Statistical Models for DP Tabular Synthesis

A revised arXiv paper challenges the widely held view that statistical methods outperform neural networks when generating differentially private tabular data. The authors argue that this conclusion glosses over cases involving densely correlated data, where neural approaches may be more effective. The submission is an updated version of previously posted research.

differential-privacy/tabular-data/synthetic-data/research

papersSEP 12 04:00 UTC

arXiv Paper Models Collapse When Multiple AI Systems Train on Each Other's Output

A new arXiv preprint examines how recursive training on AI-generated text leads to model collapse, extending prior work from a single model to settings where many models exchange and train on one another's outputs. The authors analyze how the dynamics play out across a multi-model ecosystem, where each participant learns from a shared pool of synthetic data. The work is a preprint and has not yet been peer reviewed.

model-collapse/synthetic-data/research/training-data

papersSEP 12 04:00 UTC

Study examines catastrophic forgetting in skill retrieval for LLM agents

A new arXiv paper studies how synthetic data affects the ability of LLM agents to select the right external skill from large repositories. The authors describe a deployed skill router covering 34,396 skills and run a large-scale evaluation of retrieval under limited data conditions. The findings point to catastrophic forgetting as a risk when synthetic data is used for training these routers.

agents/synthetic-data/retrieval/llm-agents

papersSEP 12 04:00 UTC

Paper Proposes Generator for Multi-System Enterprise Data Without Real Datasets

A new arXiv paper describes a synthetic data generator that produces relational business data without any real dataset at either end, requiring only inputs such as industry and company size. It also introduces a reference-free way to evaluate quality, avoiding the usual comparison against real data. The authors present the method as an alternative for creating consistent multi-system enterprise datasets.

synthetic-data/enterprise/evaluation/arxiv