paper-with-me

홈 › Papers

View From Above: A Framework for Evaluating Distribution Shifts in Model Behavior

2024-07-01 · Tanush Chopra, Michael Li, Jacob Haimes

When large language models (LLMs) are asked to perform certain tasks, how can we be sure that their learned representations align with reality? We propose a domain-agnostic framework for systematically evaluating distribution shifts in LLMs decision-making processes, where they are given control of mechanisms governed by pre-defined rules. While individual LLM actions may appear consistent with expected behavior, across a large number of trials, statistically significant distribution shifts can emerge. To test this, we construct a well-defined environment with known outcome logic: blackjack. In more than 1,000 trials, we uncover statistically significant evidence suggesting behavioral misalignment in the learned representations of LLM.

📄 PDF Abstract BibTeX arXiv:2407.00948

Code (1)

Bluefin-Tuna/ApartResearch 공식 구현

Tasks

Decision Making

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Non-adversarial Robustness of Deep Learning Methods for Computer Vision

2023-05-24 · Gorana Gojić, Vladimir Vincan, Ognjen Kundačina, Dragiša Mišković 외

Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced with distribution shifts caused by natural variations in data.…

Adversarial RobustnessDeep LearningNon-Adversarial Robustness

A Fine-Grained Analysis on Distribution Shift

2021-10-21 · ICLR 2022 4 · Olivia Wiles, Sven Gowal, Florian Stimberg, Sylvestre Alvise-Rebuffi 외

Robustness to distribution shifts is critical for deploying machine learning models in the real world. Despite this necessity, there has been little work in defining the underlying mechanisms that cause these shifts and …

Evaluating Model Robustness and Stability to Dataset Shift

2020-10-28 · Adarsh Subbaswamy, Roy Adams, Suchi Saria

As the use of machine learning in high impact domains becomes widespread, the importance of evaluating safety has increased. An important aspect of this is evaluating how robust a model is to changes in setting or popula…

BIG-bench Machine Learningmodel

Evaluating Predictive Uncertainty under Distributional Shift on Dialogue Dataset

2021-09-01 · Nyoungwoo Lee, ChaeHun Park, Ho-Jin Choi

In open-domain dialogues, predictive uncertainties are mainly evaluated in a domain shift setting to cope with out-of-distribution inputs. However, in real-world conversations, there could be more extensive distributiona…

CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization

2021-11-01 · EMNLP 2021 11 · Arjun Akula, Soravit Changpinyo, Boqing Gong, Piyush Sharma 외

One challenge in evaluating visual question answering (VQA) models in the cross-dataset adaptation setting is that the distribution shifts are multi-modal, making it difficult to identify if it is the shifts in visual or…

Answer GenerationQuestion-Answer-GenerationQuestion AnsweringVisual Question Answering+1