Tiny benchmarks reduce LLM evaluation costs by using fewer examples.
problem Expensive evaluation of LLMs with tens of thousands of examples.
method Developed evaluation tools and tiny versions of popular benchmarks.
result Accurately estimate LLM performance with just 100 curated examples.
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.
MultiwayPAM clusters LLM-as-a-Judge scores to reveal evaluator bias.
problem Revealing bias in LLM evaluations of text quality.
method Tensor clustering with MultiwayPAM method.
result Medoids reveal cluster membership of questions, answerers, and evaluators.
Bayesian approach quantifies uncertainty in LLM evaluations.
problem Statistical uncertainty in evaluating LLM behavior.
method Bayesian evaluation of LLM behavior using probabilistic text generation strategies.
result Bayesian approach provides useful uncertainty quantification about LLM behavior.
This paper proposes a method to use LLMs as auxiliary evaluators in place of human judges.
problem The need for cost-effective and scalable evaluation of AI systems.
method Formulates a two-stage sampling design with LLM evaluations and human ratings, using a doubly robust estimator.
result Proposes a method to determine optimal sample sizes for human and LLM ratings.
LLM evaluation suffers from systematic biases and lacks reliable positive judgments.
problem LLM evaluation suffers from systematic biases and lacks reliable positive judgments.
method Formulate LLM evaluation as a positive-unlabelled learning problem and propose a geometric auditing framework based on Partial Optimal Transport.
result Improved alignment with human preferences, increased robustness to presentation biases, and interpretable confidence estimates.
This paper uses LLMs and cycle consistency for better machine translation evaluation.
problem Evaluating translation quality and LLM capabilities without ground truth.
method Generate translation candidates, back-translate, and evaluate cycle consistency.
result Larger LLMs or more inference passes improve cycle consistency.
PromptEval estimates LLM performance across many prompts, improving reproducibility.
problem Limited prompt templates affect LLM benchmark reproducibility.
method Estimates performance distribution across many prompts using borrowed strength.
result PromptEval accurately estimates performance quantiles with practical budget.
Framework assesses autograders' reliability and biases.
problem Mixed reliability and biases in autograders for LLM evaluation.
method Bayesian GLMs to model evaluation outcomes.
result Explicit quantification of scoring differences and biases.
Look-Ahead-Bench evaluates financial LLMs for lookahead bias, revealing significant differences in model performance.
problem Measuring and mitigating lookahead bias in financial LLMs.
method Standardized benchmark evaluating model behavior in practical financial scenarios, analyzing performance decay across market regimes.
result Standard LLMs exhibit significant lookahead bias, while Pitinf models show improved generalization and reasoning abilities.
Language model benchmarks often misrepresent true understanding, revealing vulnerabilities in evaluation methods.
problem Language model benchmarks fail to accurately reflect true language understanding and adaptability.
method Systematic analysis of NLP evaluation frameworks, identifying vulnerabilities in static benchmarks, human evaluation protocols, and LLM-as-judge frameworks.
result Current evaluation methods are unreliable and need improvement to accurately assess LLM performance.
Financial LLMs need explicit bias consideration to avoid invalid results.
problem Finance-specific biases inflate performance and contaminate backtests.
method Identified five recurring biases and proposed a Structural Validity Framework.
result Explicit bias consideration is necessary for valid deployment claims.
Proposes a method to choose thresholds for LLM evaluation metrics.
problem Ensuring reliable large language models (LLMs) with correct threshold selection.
method Identify risks, stakeholders' risk tolerance, and use ground-truth data to determine thresholds.
result Demonstrates a concrete example with the Faithfulness metric and HaluBench dataset.
New method for evaluating LLMs reduces bias in open-ended evaluations.
problem Bias in Elo-based ratings of LLMs due to data redundancies.
method Proposes evaluation as a 3-player game and introduces novel solution concepts.
result Novel method leads to more robust and intuitive ratings.
Cer-Eval saves LLM evaluation costs while maintaining accuracy.
problem Challenges in evaluating large language models due to large dataset requirements.
method Adapts to different evaluation objectives, uses test sample complexity, and develops a partition-based algorithm.
result Cer-Eval can save 20-40% test points with comparable accuracy and 95% confidence guarantee.
No-knowledge alarms detect misaligned LLM judges without trusting them.
problem Monitoring LLM judges for consistency when ground truth is unknown.
method Formal logic and Linear Programming to detect disagreements.
result No false positives in detecting misaligned judges.
Benchmark evaluates LLM trading agents by masking identifiers to prevent memory leaks.
problem Evaluate LLM trading agents without relying on market memory or noise.
method Data-side masking protocol, Barra-style performance attribution framework.
result LLM agents' returns are largely explained by market and style exposure, not stock selection.
Unified framework to bridge human and LLM judgments.
problem Systematic discrepancies between human and LLM evaluations.
method Latent human preference score and linear transformations of covariates.
result Higher agreement with human ratings and exposure of systematic gaps.
Study evaluates if LLMs have company-specific biases in financial sentiment analysis.
problem Evaluating if large language models exhibit company-specific biases in financial sentiment analysis.
method Comparing sentiment scores with and without company names, constructing economic models, and empirical analysis.
result LLMs show company-specific biases in sentiment analysis, impacting investor behavior and stock prices.
Study evaluates five LLMs for financial report analysis, revealing performance differences and variability.
problem Lack of understanding in reliability, consistency, and transparency of LLMs in financial analysis.
method Human evaluation, automated similarity metrics, and behavioral diagnostics applied to five transformer-based LLMs over U.S. 10-K filings.
result No single LLM consistently dominates across all evaluation perspectives, highlighting variability and need for interpretability.
LLMs struggle to outperform markets over long periods and diverse stocks.
problem Overstated effectiveness of LLM-based investing strategies due to biases.
method FINSABER framework for systematic backtests over two decades and 100+ symbols.
result Previously reported LLM advantages deteriorate significantly under broader evaluation.
Novel method diagnoses large language models' reasoning abilities.
problem Fine-grained evaluation of large language models' reasoning abilities.
method Adapting cognitive diagnosis models to LLMs, estimating mastery profiles and Q-matrix, incorporating textual information.
result Accurate parameter recovery and insights into LLMs' capabilities.
New benchmarks show LLMs struggle with causal discovery.
problem Leveraging LLMs for causal discovery is unreliable due to dataset leakage.
method Developing science-grounded benchmarks and hybrid methods combining LLM predictions with statistical analysis.
result LLMs perform poorly on novel, real-world scientific studies compared to classical methods.
Paper proposes FinAR-Bench to evaluate LLMs in financial analysis tasks.
problem Inaccurate financial analysis by LLMs leading to investment and regulatory issues.
method Proposes FinAR-Bench, a benchmark dataset with three steps: key info extraction, financial indicator calculation, and logical reasoning.
result LLMs perform better in key info extraction and indicator calculation but struggle with logical reasoning.
InvestorBench benchmarks LLM agents in financial tasks.
problem Lack of a comprehensive benchmark for LLM-based financial agents.
method Developed a benchmark with diverse financial tasks and datasets.
result Evaluated LLM agents' performance across various financial products and market environments.
PredictaBoard benchmarks LLM score predictors to assess their ability to anticipate errors.
problem Inconsistent performance of LLMs in common sense reasoning tasks.
method Collaborative benchmarking framework evaluating pairs of LLMs and assessors using rejection rate at different tolerance errors.
result Highlights the need to evaluate predictability alongside performance for safer AI systems.
Fairness in LLMs is impossible due to inherent technical challenges.
problem Ensuring fairness in large language models (LLMs) with rigorous definitions.
method Analysis of various technical fairness frameworks.
result No feasible technical fairness frameworks for LLMs due to large amounts of unstructured data and many potential combinations.
This paper evaluates LLMs on large graph property estimation tasks.
problem Limited context length of LLMs limits their evaluation on large graphs.
method Developed EstGraph dataset and introduced four tasks for LLMs to estimate large graph properties.
result LLMs perform better on graph property estimation tasks when provided with context-rich prompts based on random walks.
R-AutoEval+ improves model evaluation efficiency and reliability using adaptive synthetic data.
problem Accurate model selection from AI candidates using real-world data is costly and impractical at scale.
method R-AutoEval+ uses adaptive prediction-powered inference to correct bias in autoevaluators while maintaining or improving sample efficiency.
result R-AutoEval+ provides finite-sample reliability guarantees and enhanced sample efficiency compared to conventional methods.
Telescope detects LLM generated text by measuring token repetition probability.
problem Distinguishing LLM generated text from human writing.
method Telescope Perplexity, evaluating token repetition probability.
result Telescope Perplexity enables effective zero-shot LLM detection.
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.
LaRT models LLMs' response accuracy and CoT length to evaluate reasoning ability and speed.
problem Valid evaluation of Large Language Models (LLMs) via response accuracy and chain-of-thought length.
method Introduces Latency-Response Theory (LaRT) to jointly model response accuracy and CoT length using latent ability and latent speed.
result LaRT yields higher estimation accuracy and shorter confidence intervals for latent traits compared to IRT.
PriceSeer benchmarks LLMs in real-time stock prediction.
problem Evaluating LLMs' stock prediction accuracy and robustness.
method Real-time benchmark with 110 U.S. stocks, internal and external information expansion.
result LLMs perform suboptimally in long-term predictions due to fake news and specific industries.
This paper uses CausalGANs and RL with LLM to predict bond yields.
problem Challenges in financial bond yield forecasting due to data scarcity and market conditions.
method Proposes a novel framework combining CausalGANs, RL, and LLM for synthetic data generation and trading signals.
result Improves forecasting performance over existing methods with low Mean Absolute Error.
DUET optimizes LLM training data mixtures using feedback from unseen tasks.
problem Optimizing LLM performance on unseen tasks without prior data knowledge.
method Interleaves influence function with Bayesian optimization for data selection.
result DUET converges to optimal training data mixture for unseen tasks.
AURA: Adaptive Uncertainty-aware Refinement for LLM-as-a-Judge Auditing
problem Auditing LLM-as-a-Judge decisions
method Adaptive uncertainty-aware refinement
result Human-consistent signal learning and evidence propagation
AlphaForgeBench evaluates LLMs as quantitative researchers, not trading agents, to address instability in financial decision-making.
problem Behavioral instability of LLMs in sequential decision-making under financial uncertainty.
method Proposes AlphaForgeBench, a framework that requires LLMs to generate executable alpha factors and compose factor-based trading strategies.
result Eliminates execution-induced instability and provides a rigorous benchmark for evaluating financial reasoning.
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.
Efficiently allocate budgets for LLM-assisted virtual screening to reduce costs.
problem Reducing the cost of evaluating alternatives in large-scale screening tasks.
method Propose a top-m greedy evaluation mechanism and the EFG-m algorithm for efficient budget allocation. result Prove that EFG-m is both sample-optimal and consistent in large-scale virtual screening. 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.
Study evaluates how much knowledge LLMs have by comparing their prediction accuracy to flexible models.
problem Evaluating the predictive power of LLMs without access to their training data.
method Equivalent sample size measure, comparing LLM's prediction error to flexible models trained on varying amounts of domain-specific data.
result LLMs encode varying amounts of predictive information across different economic variables.
Study evaluates LLMs like ChatGPT and GPT-4 on financial analysis tasks.
problem Assessing financial reasoning capabilities of large language models.
method Mock CFA exam questions, Zero-Shot, Chain-of-Thought, Few-Shot scenarios.
result LLMs perform well on financial analysis tasks, but have limitations.
Framework evaluates AI proposals for drug discovery, finds no LLM advantage.
problem No principled framework exists for evaluating AI-guided scientific selection under budget constraints.
method Formally verified metric (BSDS/DQS) penalizes false discoveries and excessive abstention.
result LLMs provide no marginal value over existing classifiers in drug discovery.
A simple method reduces bias in LLM auto-evaluators by controlling output length.
problem Bias in LLM auto-annotators, particularly length bias.
method Simple regression analysis to control for length difference and other mediators.
result Improved robustness and increased correlation with human preferences.
LLMs fail to match statistical ground truth despite stable run-to-run performance.
problem LLMs lack validation against statistical ground truth in automated scientific workflows.
method Introduced a behavioral evaluation framework for LLMs, separating four decision-making dimensions.
result LLMs can exhibit near-perfect stability but diverge from statistical ground truth.
UCFE benchmarks LLMs in financial tasks with human feedback.
problem Evaluating LLMs' financial task performance and user satisfaction.
method Hybrid approach combining human expert evaluations and dynamic interactions.
result Significant alignment between benchmark scores and human preferences (Pearson correlation coefficient of 0.78).
LLM-based trading systems vary in execution realism and reproducibility.
problem LLM-based trading systems vary in execution realism and reproducibility.
method Reproducibility audit of 30 trade-relevant primary studies.
result LLM-based trading systems vary in execution realism and reproducibility.
A new framework evaluates LLM calibration in open-ended QA.
problem Evaluating LLM calibration in open-ended QA settings.
method Sem-ECE framework: sampling answers, grouping by semantic classes, and using frequencies as confidence.
result Sem-ECE estimators are unbiased and Sem2 achieves smaller calibration error.