paper-with-me

홈 › Papers

Detecting Underspecification with Local Ensembles

2019-10-21 · ICLR 2020 1 · David Madras, James Atwood, Alex D'Amour

We present local ensembles, a method for detecting underspecification -- when many possible predictors are consistent with the training data and model class -- at test time in a pre-trained model. Our method uses local second-order information to approximate the variance of predictions across an ensemble of models from the same class. We compute this approximation by estimating the norm of the component of a test point's gradient that aligns with the low-curvature directions of the Hessian, and provide a tractable method for estimating this quantity. Experimentally, we show that our method is capable of detecting when a pre-trained model is underspecified on test data, with applications to out-of-distribution detection, detecting spurious correlates, and active learning.

📄 PDF Abstract BibTeX arXiv:1910.09573

Code (2)

dmadras/local-ensembles 공식 구현 tf
StanfordASL/SCOD pytorch

Tasks

Active LearningOut-of-Distribution Detection

Methods 이 논문이 사용한 방법론

Test 설명 없음

Similar Papers 제목 키워드 기반

Who is the richest club in the championship? Detecting and Rewriting Underspecified Questions Improve QA Performance

2026-02-12 · Yunchong Huang, Gianni Barlacchi, Sandro Pezzelle arxiv

Large language models (LLMs) perform well on well-posed questions, yet standard question-answering (QA) benchmarks remain far from solved. We argue that this gap is partly due to underspecified questions - queries whose …

Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking

2023-12-14 · Jacob Eisenstein, Chirag Nagpal, Alekh Agarwal, Ahmad Beirami 외

Reward models play a key role in aligning language model applications towards human preferences. However, this setup creates an incentive for the language model to exploit errors in the reward model to achieve high estim…

Language ModelingLanguage ModellingReranking

Consistent Explanations in the Face of Model Indeterminacy via Ensembling

2023-06-09 · Dan Ley, Leonard Tang, Matthew Nazari, Hongjin Lin 외

This work addresses the challenge of providing consistent explanations for predictive models in the presence of model indeterminacy, which arises due to the existence of multiple (nearly) equally well-performing models f…

Underspecification Presents Challenges for Credibility in Modern Machine Learning

2020-11-06 · Alexander D'Amour, Katherine Heller, Dan Moldovan, Ben Adlam 외

ML models often exhibit unexpectedly poor behavior when they are deployed in real-world domains. We identify underspecification as a key reason for these failures. An ML pipeline is underspecified when it can return many…

BIG-bench Machine Learning

COCORELI: Enforcing Execution Preconditions for Reliable Collaborative Instruction Following

2025-08-29 · Swarnadeep Bhar, Omar Naim, Eleni Metheniti, Bastien Navarri 외 arxiv

Autonomous agents executing human instructions must operate reliably even when instructions are incomplete. While recent approaches improve detection of missing information, detection alone is insufficient: agents often …

Instruction Following