Unified evaluation for both quality and diversity in NLP.
problem Measuring both quality and diversity in NLP models.
method Proposes HUSE, a metric combining human and statistical evaluation.
result HUSE detects both quality and diversity defects in NLP models.
New framework predicts diverse, contextually plausible 3D human motions.
problem Predicting multiple plausible future 3D poses given observed poses.
method Developed a new variational framework that conditions latent variable on past observation to encourage relevant information.
result Our approach generates motions of higher quality and preserves contextual information.
Paper withdrawn; AI game behavior needs diversity.
problem Creating varied human-like playing styles in games.
method Evolutionary multi-objective deep reinforcement learning.
result Generated diverse AI behaviors for games.
Study tackles RLHF with diverse human feedback, showing limitations and proposing a meta-learning approach.
problem Traditional RLHF fails to balance diverse human preferences.
method Integrates meta-learning and multiple social welfare functions to optimize diverse preferences.
result Establishes sample complexity bounds for optimizing diverse social welfare functions.
Paper introduces a new method for generating diverse human motion predictions.
problem Stochastic human motion prediction with limited flexibility.
method Stochastically combines root variations with previous pose information in a recurrent network.
result Model generates more diverse motion sequences than existing techniques.
This paper introduces f-DPO, a generalized approach to Direct Preference Optimization using diverse divergence constraints.
problem Aligning large language models with human preferences while mitigating safety risks.
method Incorporates diverse divergence constraints to simplify the relationship between reward and optimal policy, eliminating the need for estimating the normalizing constant.
result Optimizes LLMs to align with human preferences more efficiently and under a broader set of divergence constraints.
CrowdLLM uses LLMs and generative models to create diverse digital populations.
problem Lack of diversity and accuracy in digital populations created by LLMs.
method Integrates pretrained LLMs and generative models to enhance diversity and fidelity.
result CrowdLLM achieves promising performance in accuracy and distributional fidelity.
Improves generative Visual Dialog by asking diverse questions.
problem Generative Visual Dialog models degrade after a few rounds of interaction.
method Introduce a simple auxiliary objective to incentivize Qbot to ask diverse questions.
result Better dialog diversity, consistency, fluency, and detail with improved image relevance.
Quriosity analyzes curiosity-driven questions from diverse sources.
problem Understanding and analyzing human curiosity-driven questions.
method Collection of 13.5K questions from various sources, development of an iterative prompt improvement framework.
result 42% of questions are causal, revealing unique linguistic properties.
New model improves diversity in question generation over text.
problem Inadequate evaluation metrics for diverse questions.
method Proposed a variational encoder-decoder model and evaluation scheme.
result Improved diversity in generated questions without loss of quality.
LoCo-RLHF models diverse human feedback with contextual information.
problem Heterogeneous human feedback from diverse contexts and preferences.
method Low-rank contextual preference model, PRS policy.
result LoCo-RLHF achieves tighter sub-optimality gap than existing methods.
We propose a method to generate multiple diverse and valid human pose hypotheses in 3D all consistent with the 2D detection of joints in a monocular RGB image. We use a novel generative model uniform (unbiased) in the space of anatomically plausible 3D poses. Our model is compositional (produces a pose by combining par…
Paper explores limits and possibilities of aligning LLMs with human preferences.
problem Aligning LLMs with diverse human preferences to ensure fairness and informed outcomes.
method Analysis of probabilistic representation of human preferences and preservation of diverse preferences.
result LLMs can't fully align with human preferences using reward-based approaches due to Condorcet cycles, but mixed strategies are statistically possible.
Adversarial examples are perturbed inputs designed to fool machine learning models. Most recent works on adversarial examples for image classification focus on directly modifying pixels with minor perturbations. A common requirement in all these works is that the malicious perturbations should be small enough (measured…
Experiment shows cognitive biases impact human-AI collaboration, highlighting the need for diverse evaluator samples.
problem Cognitive biases affect human-AI collaboration, leading to suboptimal outcomes.
method Randomized experiment with 2,784 participants, manipulating AI suggestion quality, task burden, and financial incentives.
result Individual attitudes toward AI are the strongest predictor of performance, influencing accuracy and overcorrection.
Bayesian framework infers personalized embeddings for diverse human demonstrations.
problem Lack of personalized models for diverse human behaviors.
method Bayesian LfD framework inferring human-specific embeddings.
result Outperforms state-of-the-art techniques on synthetic and real-world data.
DreamerV3 learns diverse tasks with a single configuration.
problem Generalizing reinforcement learning across multiple domains.
method World model and farsighted strategies.
result Outperforms specialized methods across 150 diverse tasks.
New simulation model predicts financial market dynamics with high accuracy.
problem Extreme difficulty in financial market projections due to human behavioural complexity.
method Agent-based modeling with a hierarchical knowledge architecture to simulate diverse human groups.
result Simulator achieves 13.29% deviation in crisis scenarios and lower mean square error under normal conditions.
Introduces Motion Programs for better video analysis of human motion.
problem Current video analysis focuses on raw pixels or keypoints, missing higher-level motion primitives.
method Introduces Motion Programs as a neuro-symbolic representation of motions as a composition of high-level primitives.
result Motion Programs accurately describe diverse human motions and improve downstream tasks.
In recent years, reinforcement learning (RL) methods have been applied to model gameplay with great success, achieving super-human performance in various environments, such as Atari, Go, and Poker. However, those studies mostly focus on winning the game and have largely ignored the rich and complex human motivations, w…
SharedRep-RLHF learns shared traits for diverse groups, improving fairness and performance.
problem Uniform-reward RLHF fails to capture diverse preferences, leading to unfairness.
method SharedRep-RLHF learns shared traits among various groups, improving fairness and performance.
result SharedRep-RLHF outperforms MaxMin-RLHF by up to 20% in win rate.
Study flaws in generative model evaluation metrics, especially for diffusion models.
problem Flaws in existing metrics for evaluating generative models, particularly for diffusion models.
method Systematic study of generative models, human perception experiments, and analysis of feature extractors.
result State-of-the-art perceptual realism of diffusion models is not reflected in commonly reported metrics.
Despite significant advances in the field of deep Reinforcement Learning (RL), today's algorithms still fail to learn human-level policies consistently over a set of diverse tasks such as Atari 2600 games. We identify three key challenges that any algorithm needs to master in order to perform well on all games: process…
Paper tackles unobserved confounding in human-AI collaborations.
problem Unobserved confounding undermines human-AI collaboration effectiveness.
method Combines sensitivity analysis from causal inference with AI-driven statistical modeling.
result Enhances robustness and reliability of collaborative outcomes.
Proposes a method to increase diversity without sacrificing meritocracy.
problem Systemic bias in datasets affecting diversity and meritocracy.
method Optimally flipping outcome labels and training classification models simultaneously.
result The price of diversity is low and sometimes negative, enhancing diversity without significantly affecting meritocracy.
Study models human investors' sub-rational behavior in financial markets.
problem Lack of a comprehensive model for human sub-rationality in financial markets.
method Flexible reinforcement learning model incorporating five human sub-rational aspects.
result Model accurately reproduces human behavior and reveals insights into market dynamics.
VHGM-MAE generates synthetic humans from healthcare data.
problem Handling high-dimensional, sparse healthcare data with missing values.
method Masked autoencoder (MAE) tailored for healthcare data, addressing heterogeneity, missingness, and high-dimensionality.
result VHGM-MAE outperforms existing methods in missing value imputation and synthetic data generation.
New benchmark for causal reasoning from human video descriptions.
problem Lack of diversity in event types and natural language descriptions, and differences from human judgments.
method Iterative event cloze task and data augmentation techniques.
result Improved data collection efficiency and diverse causal judgments.
Generative Augmented Inference improves AI-generated data for causal inference.
problem Challenges in using AI-generated annotations for reliable causal inference.
method Generative Augmented Inference (GAI) treats AI outputs as informative features for learning true labels, flexibly modeling the relationship using nonparametric methods.
result GAI significantly reduces estimation error and improves confidence interval quality compared to human-only and PPI-based methods.
Dual-decoder model generates responses with targeted sentiment.
problem Generating human-like responses with specific sentiment.
method Simple dual-decoder model with two sentiment decoders connected to one encoder.
result Significant performance gain in sentiment accuracy and word diversity.
Study evaluates human vs. machine review generation, finds human assessments correlate better with lexical overlaps.
problem Evaluating natural language generation models for online reviews is challenging and inconsistent.
method Compared human evaluators with various automated evaluation methods, including discriminative and word overlap metrics.
result Human evaluators do not correlate well with discriminative evaluators, but correlate better with lexical overlaps.
New metric for video generation models, addressing quality, coherence, and diversity.
problem Lack of metrics for evaluating generative models of video.
method Proposed Fréchet Video Distance (FVD) and StarCraft 2 Videos (SCV) benchmark.
result FVD correlates well with human judgment of generated videos.
XPO enhances RLHF by encouraging diverse responses, offering improved sample efficiency.
problem Limited exploration in RLHF leads to suboptimal models and bottlenecks.
method XPO is a simple one-line change to DPO, introducing a novel exploration bonus.
result XPO achieves strong theoretical guarantees and promising empirical performance.
Automated discovery of diverse self-organized patterns in complex systems.
problem Automated identification of interesting spatially localized patterns in self-organizing systems.
method Intrinsically motivated machine learning algorithms (POP-IMGEPs) combined with deep auto-encoders and CPPN primitives.
result Efficiency and effectiveness of the proposed method in discovering diverse patterns compared to baselines.
This paper solves the multiple reference model problem in RLHF with exact solutions and sample complexity guarantees.
problem Limitations of single reference models in aligning LLMs with human feedback.
method Integrates multiple reference models into RLHF frameworks, addressing theoretical challenges with exact solutions and sample complexity guarantees.
result First exact solution to the multiple reference model problem in reverse KL-regularized RLHF.
New framework formalizes RLHF trilemma: improving safety, fairness, and robustness is computationally infeasible.
problem Aligning large language models with diverse human values while maintaining computational feasibility and robustness.
method Complexity-theoretic analysis integrating statistical learning theory and robust optimization.
result Achieving both representativeness (epsilon <= 0.01) and robustness (delta <= 0.001) for global-scale populations requires super-polynomial operations.
Decoding strategies often exclude human-like tokens, creating a detectable gap in generated text.
problem Decoding strategies exclude contextually appropriate but statistically rare tokens, creating a detectable gap in generated text.
method Analysis of 1.8 million texts across 8 language models, 5 decoding strategies, and 53 hyperparameter configurations.
result 8-18% of human-selected tokens fall outside typical truncation boundaries, indicating a detectable gap.
Proposes GM Score to evaluate GANs considering diversity, disentanglement, and discriminability.
problem Evaluation of GANs for sample quality and diversity.
method Integrates various factors including intra-class and inter-class diversity, disentanglement, and discriminability metrics.
result Demonstrates improved evaluation of GANs on MNIST dataset.
The paper explores game-theoretic alignment of LLMs with human preferences, finding limitations and conditions.
problem Aligning LLMs with human preferences using game theory.
method Systematic study of payoff choices in a two-player zero-sum game for desirable alignment properties.
result Impossibility of preference matching in game-theoretic LLM alignment under standard assumptions.
Subject Cross Validation improves Human Activity Recognition performance by up to 16%.
problem Overestimation of Human Activity Recognition performance using k-fold cross validation.
method Investigated Subject Cross Validation vs. k-fold cross validation for Human Activity Recognition.
result Subject Cross Validation increases performance by up to 16%.
Crowdsourcing can improve scientific investigation by enabling reproducibility and transparency.
problem Current research methods lack reproducibility and transparency, leading to unreliable decisions.
method Next-generation investigative approach leveraging human diversity, micro-specialized crowds, and computer-assisted control methods.
result The Theory of Enablers provides specific cognitive and non-cognitive enablers for crowd-based scientific investigation.
Large-scale eye-tracking dataset for Atari games.
problem Improving AI decision-making through human eye-tracking data.
method Recorded 117 hours of gameplay with eye movements from 20 Atari games.
result Human gameplay decisions and scores comparable to human records.
Contrary to conventional economic growth theory, which reduces a country's output to one aggregate variable (GDP), product diversity is central to economic development, as recent 'economic complexity' research suggests. A country's product diversity reflects its diversity of knowhow or 'capabilities'. Researchers propo…
MallowsPO enhances LLM fine-tuning with a dispersion index of human preferences.
problem Lack of diversity in human preferences in DPO.
method Developed a dispersion index based on Mallows' theory to characterize preference diversity.
result Demonstrated improved performance in various tasks using the dispersion index.
This work explores how XAI methods can visualize the diversity of feature representations in Bayesian Neural Networks.
problem Explaining the diversity of feature representations learned by Bayesian Neural Networks.
method Application of global XAI methods to visualize and quantify the diversity of feature representations.
result The diversity of learned feature representations correlates with uncertainty estimates and network width.
Face verification remains a challenging problem in very complex conditions with large variations such as pose, illumination, expression, and occlusions. This problem is exacerbated when we rely unrealistically on a single training data source, which is often insufficient to cover the intrinsically complex face variatio…
Machine learning workflow development is anecdotally regarded to be an iterative process of trial-and-error with humans-in-the-loop. However, we are not aware of quantitative evidence corroborating this popular belief. A quantitative characterization of iteration can serve as a benchmark for machine learning workflow d…
Mixed likelihood GPs improve model performance in human-in-the-loop experiments.
problem Lack of auxiliary information in traditional GPs for human responses.
method Propose mixed likelihood variational GPs to leverage auxiliary information.
result Modeling performance improvements across diverse applications.