163 citations · 527 across the 28 of their papers we have counts for
28 papers
The Bias is in the Details: An Assessment of Cognitive Bias in LLMs
R. Alexander Knipper, Charles S. Knipper, Kaiqi Zhang +3
As Large Language Models (LLMs) are increasingly embedded in real-world decision-making processes, it becomes crucial to examine the extent to which they exhibit cognitive biases.…
Cost-Effective Proxy Reward Model Construction with On-Policy and Active Learning
Yifang Chen, Shuohang Wang, Ziyi Yang +6
Reinforcement learning with human feedback (RLHF), as a widely adopted approach in current large language model pipelines, is \textit{bottlenecked by the size of human preference d…
Active, anytime-valid risk controlling prediction sets
Ziyu Xu, Nikos Karampatziakis, Paul Mineiro
Rigorously establishing the safety of black-box machine learning models concerning critical risk measures is important for providing guarantees about model behavior. Recently, Bate…
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
Marah Abdin, Jyoti Aneja, Hany Awadalla +126
We introduce phi-3-mini, a 3.8 billion parameter language model trained on 3.3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal test…
LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Models
Yixiao Li, Yifan Yu, Chen Liang +4
Quantization is an indispensable technique for serving Large Language Models (LLMs) and has recently found its way into LoRA fine-tuning. In this work we focus on the scenario wher…
Meet in the Middle: A New Pre-training Paradigm
Anh Nguyen, Nikos Karampatziakis, Weizhu Chen
Most language models (LMs) are trained and applied in an autoregressive left-to-right fashion, assuming that the next token only depends on the preceding ones. However, this assump…
AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning
Qingru Zhang, Minshuo Chen, Alexander Bukharin +5
Fine-tuning large pre-trained language models on downstream tasks has become an important paradigm in NLP. However, common practice fine-tunes all of the parameters in a pre-traine…
Anytime-valid off-policy inference for contextual bandits
Ian Waudby-Smith, Lili Wu, Aaditya Ramdas +2
Contextual bandit algorithms are ubiquitous tools for active sequential experimentation in healthcare and the tech industry. They involve online learning algorithms that adaptively…
Contextual Bandit Applications in Customer Support Bot
Sandra Sajeev, Jade Huang, Nikos Karampatziakis +3
Virtual support agents have grown in popularity as a way for businesses to provide better and more accessible customer service. Some challenges in this domain include ambiguous use…
Off-policy Confidence Sequences
Nikos Karampatziakis, Paul Mineiro, Aaditya Ramdas
We develop confidence bounds that hold uniformly over time for off-policy evaluation in the contextual bandit setting. These confidence sequences are based on recent ideas from mar…
Lessons from Contextual Bandit Learning in a Customer Support Bot
Nikos Karampatziakis, Sebastian Kochman, Jade Huang +3
In this work, we describe practical lessons we have learned from successfully using contextual bandits (CBs) to improve key business metrics of the Microsoft Virtual Agent for cust…
Empirical Likelihood for Contextual Bandits
Nikos Karampatziakis, John Langford, Paul Mineiro
We propose an estimator and confidence interval for computing the value of a policy from off-policy data in the contextual bandit setting. To this end we apply empirical likelihood…
Gradient Coding
Rashish Tandon, Qi Lei, Alexandros G. Dimakis +1
We propose a novel coding theoretic framework for mitigating stragglers in distributed learning. We show how carefully replicating data blocks and coding across gradients can provi…
Log-time and Log-space Extreme Classification
Kalina Jasinska, Nikos Karampatziakis
We present LTLS, a technique for multiclass and multilabel prediction that can perform training and inference in logarithmic time and space. LTLS embeds large classification proble…
Logarithmic Time One-Against-Some
Hal Daume, Nikos Karampatziakis, John Langford +1
We create a new online reduction of multiclass classification to binary classification for which training and prediction time scale logarithmically with the number of classes. Comp…
Active Information Acquisition
He He, Paul Mineiro, Nikos Karampatziakis
We propose a general framework for sequential and dynamic acquisition of useful information in order to solve a particular task. While our goal could in principle be tackled by gen…
A Hierarchical Spectral Method for Extreme Classification
Paul Mineiro, Nikos Karampatziakis
Extreme classification problems are multiclass and multilabel classification problems where the number of outputs is so large that straightforward strategies are neither statistica…
Fast Label Embeddings for Extremely Large Output Spaces
Paul Mineiro, Nikos Karampatziakis
Many modern multiclass and multilabel problems are characterized by increasingly large output spaces. For these problems, label embeddings have been shown to be a useful primitive…
Scalable Multilabel Prediction via Randomized Methods
Nikos Karampatziakis, Paul Mineiro
Modeling the dependence between outputs is a fundamental challenge in multilabel classification. In this work we show that a generic regularized nonlinearity mapping independent pr…
Fast Label Embeddings via Randomized Linear Algebra
Paul Mineiro, Nikos Karampatziakis
Many modern multiclass and multilabel problems are characterized by increasingly large output spaces. For these problems, label embeddings have been shown to be a useful primitive…
A Randomized Algorithm for CCA
Paul Mineiro, Nikos Karampatziakis
We present RandomizedCCA, a randomized algorithm for computing canonical analysis, suitable for large datasets stored either out of core or on a distributed file system. Accurate r…
Efficient Online Bootstrapping for Large Scale Learning
Zhen Qin, Vaclav Petricek, Nikos Karampatziakis +2
Bootstrapping is a useful technique for estimating the uncertainty of a predictor, for example, confidence intervals for prediction. It is typically used on small to moderate sized…
Combining Structured and Unstructured Randomness in Large Scale PCA
Nikos Karampatziakis, Paul Mineiro
Principal Component Analysis (PCA) is a ubiquitous tool with many applications in machine learning including feature construction, subspace embedding, and outlier detection. In thi…
Least Squares Revisited: Scalable Approaches for Multi-class Prediction
Alekh Agarwal, Sham M. Kakade, Nikos Karampatziakis +2
This work provides simple algorithms for multi-class (and multi-label) prediction in settings where both the number of examples n and the data dimension d are relatively large. The…