arXiv research

Efficient exploration improves large language model performance with fewer queries.

New algorithm for clustering with faulty oracle achieves optimal queries and efficiency.

Efficiently tests two distributions with few label queries.

We investigate the problem of active learning on a given tree whose nodes are assigned binary labels in an adversarial way. Inspired by recent results by Guillory and Bilmes, we characterize (up to constant factors) the optimal placement of queries so to minimize the mistakes made on the non-queried nodes. Our query se…

sGPO optimizes RLVR training by balancing inference FLOPs and training efficiency.

Study efficient auditing of ML fairness models.

NP-Attack reduces query counts for black-box adversarial attacks.

We consider a query-based data acquisition problem for binary classification of unknown labels, which has diverse applications in communications, crowdsourcing, recommender systems and active learning. To ensure reliable recovery of unknown labels with as few number of queries as possible, we consider an effective quer…

LSDAT reduces query efficiency for decision-based adversarial attacks.

Distributions over rankings are used to model data in various settings such as preference analysis and political elections. The factorial size of the space of rankings, however, typically forces one to make structural assumptions, such as smoothness, sparsity, or probabilistic independence about these underlying distri…

Optimal query allocation improves extractive QA efficiency with LLMs.

UQE uses LLMs to analyze unstructured data efficiently.

Proposes a query-efficient blackbox attack method.

PMP extends GNNs to handle past states efficiently.

We investigate how an adversary can optimally use its query budget for targeted evasion attacks against deep neural networks in a black-box setting. We formalize the problem setting and systematically evaluate what benefits the adversary can gain by using substitute models. We show that there is an exploration-exploita…

Many machine learning tasks such as clustering, classification, and dataset search benefit from embedding data points in a space where distances reflect notions of relative similarity as perceived by humans. A common way to construct such an embedding is to request triplet similarity queries to an oracle, comparing two…

Networks are widely used to model objects with interactions and have enabled various downstream applications. However, in the real world, network mining is often done on particular query sets of objects, which does not require the construction and computation of networks including all objects in the datasets. In this w…

New methods improve black-box adversarial attacks with fewer queries.

Study learns linear utility functions from comparisons, showing learnability gaps between passive and active learning.

Efficient algorithms recover two sparse models from a mix of linear queries.

Paper introduces efficient methods for probabilistic querying of event sequences.

Marich extracts high-fidelity models from public data with minimal queries.

Efficient method for generating adversarial examples with limited query budget.

Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ distance. We …

We propose a new localized inference algorithm for answering marginalization queries in large graphical models with the correlation decay property. Given a query variable and a large graphical model, we define a much smaller model in a local region around the query variable in the target model so that the marginal dist…

Most content-based image retrieval systems consider either one single query, or multiple queries that include the same object or represent the same semantic information. In this paper we consider the content-based image retrieval problem for multiple query images corresponding to different image semantics. We propose a…

We study the most practical problem setup for evaluating adversarial robustness of a machine learning system with limited access: the hard-label black-box attack setting for generating adversarial examples, where limited model queries are allowed and only the decision is provided to a queried data input. Several algori…

A common problem in disciplines of applied Statistics research such as Astrostatistics is of estimating the posterior distribution of relevant parameters. Typically, the likelihoods for such models are computed via expensive experiments such as cosmological simulations of the universe. An urgent challenge in these rese…

The paper tackles machine unlearning by designing efficient algorithms for adaptive query classes.

Active learning optimizes correlation clustering by querying the most informative pairwise comparisons.

Paper presents efficient algorithms for reconstructing noisy pooled data.

We seek to align agent policy with human expert behavior in a reinforcement learning (RL) setting, without any prior knowledge about dynamics, reward function, and unsafe states. There is a human expert knowing the rewards and unsafe states based on his preference and objective, but querying that human expert is expens…

Paper proposes ZO-NGD for more efficient black-box attacks.

A new method for private query release using Johnson-Lindenstrauss projection.

We propose a framework for Semi-Supervised Active Clustering framework (SSAC), where the learner is allowed to interact with a domain expert, asking whether two given instances belong to the same cluster or not. We study the query and computational complexity of clustering in this framework. We consider a setting where…

New algorithms learn MNL weights efficiently for any slate size.

Improved algorithm for selecting a hypothesis locally privately with fewer queries.

Proposes efficient FOBO algorithms for global maxima of expensive functions.

Study reduces memory needs for active learning with enriched queries.

We introduce new combinatorial quantities for concept classes, and prove lower and upper bounds for learning complexity in several models of query learning in terms of various combinatorial quantities. Our approach is flexible and powerful enough to enough to give new and very short proofs of the efficient learnability…

Study efficient sequential evaluation of large language models using historical data.

This work tackles model-based RL by optimizing state-action queries to learn policies with minimal data.

The paper tests properties of trees in graphical models using covariance queries.

Oracle-efficient algorithms reduce combinatorial semi-bandit regret to logarithmic time.

Meta-algorithm for efficient reinforcement learning from human preferences.

Estimates heavy hitters in data streams with queries, balancing accuracy and efficiency.

We focus on the problem of black-box adversarial attacks, where the aim is to generate adversarial examples using information limited to loss function evaluations of input-output pairs. We use Bayesian optimization~(BO) to specifically cater to scenarios involving low query budgets to develop query efficient adversaria…

Paper proposes efficient RLHF methods for LLMs using active queries.