PDO optimizes LLM prompts without labels, improving performance.
problem Optimizing prompts for LLMs without access to labeled data.
method Pairwise preference feedback, dueling bandits, Thompson Sampling, mutation.
result PDO identifies stronger prompts than label-free methods.
New method improves LLM judge accuracy by accounting for dependencies in aggregated binary labels.
problem Classical label aggregation methods fail to account for dependencies among LLM judges, leading to miscalibrated predictions.
method Dependence-aware models based on Ising graphical models and latent factors.
result The proposed method outperforms classical methods on real-world datasets, reducing excess risk.
The study examines label smoothing to improve confidence calibration in fine-tuned LLMs.
problem Improving confidence calibration in fine-tuned large language models (LLMs) after instruction tuning.
method Examine various open-sourced LLMs, label smoothing, and custom kernel design.
result Label smoothing is effective in maintaining confidence calibration but faces challenges in large vocabulary LLMs.
New research shows label refinement and weak training have limitations for aligning LLMs.
problem Limitations of refinement methods for aligning large language models.
method Analyzed probabilistic assumptions and alternative approaches to label refinement and weak training.
result Label refinement and weak training suffer from irreducible error, leaving a performance gap.
LLM embeddings improve adaptation to tabular Y∣X-shifts with few labeled examples.
problem Improving robustness to Y∣X-shifts in tabular data. method Serializing tabular data to LLM embeddings and fine-tuning for adaptation.
result LLM embeddings can be adapted to target domains with minimal labeled data.
Small LLMs outperform large ones on simple tasks without extra labelling costs.
problem Performance of large commercial models in simple classification tasks.
method Logistic Regression on small LLM embeddings.
result Small LLMs equal or outperform large LLMs in 'tens-of-shot' classification tasks.
LFD method improves text classification by making features clearer and less label-leaking.
problem Creating interpretable text representations that are both predictive and understandable.
method LFD method: proposes lexical and semantic features from contrastive text pairs, screens candidates using κ, and selects features by residual gain. result LFD features achieve higher human-human and human-LLM agreement than baseline concepts and are less label-leaking.
Measures policy-violating content prevalence with ML-assisted sampling and LLM labeling.
problem Accurate measurement of content violations that are often rare and costly to label.
method Design-based measurement system using ML-assisted probability sampling and LLM labeling.
result Produces unbiased prevalence estimates with confidence intervals and dashboard drilldowns.
Study uses LLMs to categorize financial tweets, revealing useful sentiment signals.
problem Discovering meaningful sentiment signals from unstructured financial social media data.
method Leveraged LLMs to automatically label financial tweets with event categories and aligned with returns.
result Certain event labels consistently yield negative alpha, with statistically significant Sharpe ratios and information coefficients.
Framework controls uncertainty in LLMs without labels or probabilities.
problem Managing uncertainty in black-box LLMs without token-level probability or true labels.
method Integrates generative models, UCP, and conformal alignment to control uncertainty.
result Achieves close-to-nominal coverage and tighter thresholds than split UCP.
Improved financial sentiment analysis using LLMs with retrieval augmentation.
problem Limited performance of traditional NLP models in financial sentiment analysis.
method Retrieval-augmented Large Language Models (LLMs) with instruction tuning.
result Achieved 15% to 48% performance gain in accuracy and F1 score.
Novel approach trains LLMs for inductive reasoning using probabilistic programs.
problem Training LLMs for inductive reasoning with sparse, ambiguous data.
method Program-based Posterior Training (PPT) using probabilistic inference.
result Significant improvement in estimation accuracy and alignment with human judgments.
This study explores how examples influence ICL in LLMs.
problem Understanding how examples impact in-context learning in large language models.
method Theoretical study with a probabilistic model extending from Gaussian mixture model.
result The impact of pre-training knowledge and examples on ICL prediction accuracy.
Bal-PM reduces preference labeling costs for LLMs.
problem Efficiently acquiring human feedback for preference modeling in large language models.
method Bayesian Active Learning with entropy maximization in feature space.
result Bal-PM reduces the number of required preference labels by 33% to 68%.
A new framework evaluates LLMs by considering judge reliability.
problem Evaluating LLMs without ground truth labels can lead to biased results.
method Introduces judge-specific discrimination parameters and estimates model quality and judge reliability.
result Improves agreement with human preferences and produces calibrated uncertainty quantification.
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.
Work shows hallucination detection by LLMs is impossible without expert feedback.
problem Detecting hallucinations in LLMs is theoretically impossible without expert-labeled feedback.
method Investigated hallucination detection using a theoretical framework inspired by language identification.
result Automated hallucination detection is impossible for most language collections without expert-labeled feedback.
Unified framework for certifying LLM reliability without extra supervision.
problem Improving reliability of large language models without additional supervision.
method Unified framework using majority voting and Martingale Majority Certificate (MMC).
result Certifiable inference in LLMs with statistical guarantees and adaptive stopping rules.
This study evaluates zero-shot LLMs in finance, finding ChatGPT performs well but fine-tuned models are better.
problem Evaluating zero-shot LLMs in financial tasks.
method Comparison of ChatGPT and fine-tuned models on annotated data.
result Fine-tuned models generally outperform zero-shot LLMs.
AutoElicit uses LLMs to quickly create expert priors for predictive models.
problem Creating accurate priors for predictive models is time-consuming and costly.
method AutoElicit extracts knowledge from LLMs to construct priors for predictive models.
result AutoElicit yields priors that reduce error and save labelling effort.
FUSE improves verification quality without ground truth labels.
problem Verification of model outputs using imperfect judges and reward models.
method Ensembling verifiers without ground truth labels using spectral algorithms.
result FUSE matches or improves upon semi-supervised alternatives in test-time scaling experiments.
BeGIN benchmarks GNNs for instance-dependent label noise in graphs.
problem Instance-dependent label noise in graph data.
method BeGIN introduces a benchmark with various noise types and evaluates noise-handling strategies across GNN architectures.
result Challenges of instance-dependent noise, especially LLM-based corruption, and the importance of node-specific parameterization.
LLMs detect market patterns through causal reasoning, not just temporal association.
problem Detecting structural market patterns in financial data.
method Obfuscation testing using the WHO-WHOM-WHAT framework.
result LLMs achieve 71.5% detection rate of market patterns without temporal context.
Paper proposes using online text data to predict CPI with LLMs.
problem Forecasting Consumer Price Index (CPI) using low-frequency survey-based data.
method Develops an LLM-based approach combining online text time series with monthly CPI data.
result Establishes the asymptotic properties and provides prediction intervals for CPI forecasts.
BTZSC benchmarks zero-shot text classification across diverse models.
problem Systematically comparing zero-shot text classification across various models.
method Comprehensive benchmark of 22 datasets, comparing NLI cross-encoders, embedding models, rerankers, and instruction-tuned LLMs.
result Rerankers and instruction-tuned LLMs outperform NLI cross-encoders, with rerankers setting a new state-of-the-art.
CARE improves LLM aggregation by accounting for shared confounders.
problem LLM judges' correlated errors due to shared latent confounders.
method CARE explicitly models judges' scores as true quality and confounders.
result CARE reduces aggregation error by up to 26.8% across various benchmarks.
Study shows multilingual LLM calibration effects improve model confidence but not accuracy.
problem Improving multilingual language model calibration in low-resource settings.
method Analysis of two multilingual benchmarks using instruction-tuning and label smoothing.
result Model confidence increases in low-resource languages after instruction-tuning but accuracy improvements are marginal.
Study reduces human labeling in LLM-based classification systems.
problem Minimizing human intervention in training LLM-based classification systems.
method Active learning framework with Conservative Hull-based Classifier (CHC), Center-based Classifier (CC), and Generalized Hull-based Classifier (GHC).
result CHC achieves O(logdT) regret and is minimax optimal for d=1. GHC bridges the gap between different regimes. Study compares DSPy teleprompter algorithms for aligning LLM evaluations with human annotations.
problem Aligning LLM evaluation metrics with human annotations.
method Comparative analysis of five teleprompter algorithms within the DSPy framework.
result Certain teleprompters outperform others in detecting hallucinations.
Aggregates diverse zero-shot LLM outputs for better corporate disclosure classification.
problem Combining varied zero-shot LLM predictions for improved stock return prediction.
method Multi-prompt framework with three fixed zero-shot LLM classifiers, logistic meta-classifier aggregation.
result Aggregated model outperforms single classifiers and baseline models, increasing balanced accuracy from 0.566 to 0.606.
LLM forecasting benchmarks suffer from information leakage, which confounds model performance.
problem LLM forecasting benchmarks suffer from information leakage.
method A retrieval-augmented LLM forecaster observes only decision-time information.
result The full pipeline obtains a median monthly Spearman rank IC of +0.154.
A new LLM-based method enhances diversity in oversampling for imbalanced classification.
problem Limited diversity in synthetic minority samples generated by current LLM-based approaches reduces robustness and generalizability.
method Condition synthetic sample generation on minority labels and features, use permutation strategy for fine-tuning, fine-tune on minority and interpolated samples.
result Significantly outperforms eight SOTA baselines in diverse synthetic sample generation and downstream classification tasks.
SWIFT learns intrinsic rewards from LLM hidden states for efficient best-of-N sampling.
problem Efficiency and scalability of reward models for LLMs.
method SWIFT (Simple Weighted Intrinsic Feedback Technique) learns a reward function directly from LLM hidden states.
result SWIFT outperforms existing baselines by 12.7% on MATH dataset while using less than 0.005% of their parameters.
FSPO optimizes synthetic preferences for LLM personalization.
problem Personalizing large language models for diverse users.
method FSPO reframes reward modeling as a meta-learning problem, using few labeled preferences and synthetic data.
result FSPO achieves high winrates in personalized responses, both synthetic and real.
Paper fine-tunes LLMs using user edits, unifying preference, supervision, and reward feedback.
problem Adapting LLMs to user preferences and feedback types.
method Derives bounds for learning algorithms from user edits, proposes an ensembling procedure.
result Ensembling procedure outperforms individual feedback methods and robustly adapts to different user-edit distributions.
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 predicts nonfactuality in LLM responses using semantic isotropy.
problem Assessing trustworthiness of long-form LLM responses efficiently.
method Semantic isotropy of text embeddings on the unit sphere.
result Higher semantic isotropy correlates with lower factual consistency.
New approach to fairness in machine learning models using conformal prediction.
problem Fairness in machine learning models' downstream decision-making.
method Theoretical derivation and empirical evaluation of label-clustered conformal prediction.
result Label-clustered conformal prediction often provides a favorable balance between utility and substantive fairness.
Study evaluates consistency of LLMs in binary text classification, providing systematic guidance.
problem Lack of reliable methods for evaluating large language model (LLM) binary text classification.
method Adapting psychometric principles, the study determines sample size requirements, develops metrics for invalid responses, and evaluates intra- and inter-rater reliability.
result LLMs demonstrated high intra-rater consistency, achieving perfect agreement on 90-98% of examples, with smaller models outperforming larger counterparts.
New method uses model comparison signals to improve LLM evaluation accuracy.
problem Limited benchmark sizes and model stochasticity in evaluating LLMs' mathematical reasoning.
method Combines standard labeled outcomes with model comparison signals to design a statistically efficient evaluation framework.
result Semiparametric estimator achieves the semiparametric efficiency bound and substantially improves ranking accuracy.
Duel-Evolve uses LLM self-preferences for test-time optimization of discrete outputs.
problem Optimizing LLM outputs at test time with limited or unreliable scalar rewards.
method Duel-Evolve uses pairwise comparisons from the LLM to guide optimization, aggregating them via a Bayesian Bradley-Terry model.
result Achieves significant improvement over existing methods in accuracy.
Meta-Router optimizes LLM selection using gold-standard and preference-based data.
problem Training a high-quality LLM router with combined data sources is challenging due to bias and scarcity.
method Developed an integrative causal router training framework to correct bias and improve routing accuracy.
result Our approach delivers more accurate routing and improves the trade-off between cost and quality.
Active testing for large language models is made more efficient and accurate.
problem Efficient evaluation of large language models with limited labels.
method Cost-saving measures and in-context learning for constructing a surrogate model.
result Significantly more accurate evaluations of LLM performance compared to random data acquisition.
New models can't beat existing ones, so debiasing methods only slightly reduce needed labels.
problem Limiting scalability in model evaluation due to self-preferencing biases.
method Study of debiasing methods using a few high-quality labels to reduce model judgments.
result Debiasing methods can't decrease required ground truth labels by more than half when the judge is no more accurate than the model.
A new method selects models for ensemble learning to maximize mutual information, outperforming existing approaches.
problem Selecting models for ensemble learning to improve performance and reduce correlation issues.
method Formulate budgeted ensemble selection as maximizing mutual information, use Gaussian-copula to model correlated errors, propose a greedy mutual-information selection algorithm.
result Our method consistently outperforms strong baselines across multiple datasets.
Proposes RPG-RT for red-teaming T2I models without internal access.
problem Evaluating T2I models' security through red-teaming is challenging due to their closed-source nature and unknown defense mechanisms.
method Integrates LLM and rule-based preference modeling to dynamically adapt to unknown defense mechanisms.
result Demonstrates superior and practical approach for red-teaming T2I models.
DKPS provides guarantees for synthetic data from Transformer models, improving downstream tasks.
problem Lack of labeled data for building performant AI models.
method Data Kernel Perspective Space (DKPS) for mathematical analysis of synthetic data quality.
result Concrete statistical guarantees for the quality of transformer model outputs.
Training-free source selection for LLM families with shared vocabularies
problem Source selection for LLM families with shared vocabularies
method Fisher alignment at vocabulary scale
result Fisher alignment is a cosine between kernel mean embeddings in the joint activation-error space