paper-with-me

Papers

Evaluating Superhuman Models with Consistency Checks

2023-06-16 · Lukas Fluri, Daniel Paleka, Florian Tramèr

If machine learning models were to achieve superhuman abilities at various reasoning or decision-making tasks, how would we go about evaluating such models, given that humans would necessarily be poor proxies for ground truth? In this paper, we propose a framework for evaluating superhuman models via consistency checks. Our premise is that while the correctness of superhuman decisions may be impossible to evaluate, we can still surface mistakes if the model's decisions fail to satisfy certain logical, human-interpretable rules. We instantiate our framework on three tasks where correctness of decisions is hard to evaluate due to either superhuman model abilities, or to otherwise missing ground truth: evaluating chess positions, forecasting future events, and making legal judgments. We show that regardless of a model's (possibly superhuman) performance on these tasks, we can discover logical inconsistencies in decision making. For example: a chess engine assigning opposing valuations to semantically identical boards; GPT-4 forecasting that sports records will evolve non-monotonically over time; or an AI judge assigning bail to a defendant only after we add a felony to their criminal record.

📄 PDF Abstract BibTeX arXiv:2306.09983

Code (2)

ethz-privsec/superhuman-ai-consistency 공식 구현
ethz-spylab/superhuman-ai-consistency 공식 구현

Tasks

Decision Making

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
fail 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

What's the Meaning of Superhuman Performance in Today's NLU?

2023-05-15 · Simone Tedeschi, Johan Bos, Thierry Declerck, Jan Hajic 외

In the last five years, there has been a significant focus in Natural Language Processing (NLP) on developing larger Pretrained Language Models (PLMs) and introducing benchmarks such as SuperGLUE and SQuAD to measure the…

PositionReading Comprehension

Humanly Certifying Superhuman Classifiers

2021-09-16 · Qiongkai Xu, Christian Walder, Chenchen Xu

Estimating the performance of a machine learning system is a longstanding challenge in artificial intelligence research. Today, this challenge is especially relevant given the emergence of systems which appear to increas…

Deepchecks: Evaluating Retrieval-Augmented Generation (RAG)

2026-05-14 · Assaf Gerner, Netta Madvil, Nadav Barak, Alex Zaikman 외 arxiv

Large Language Models (LLMs) augmented with Retrieval-Augmented Generation (RAG) techniques are revolutionizing applications across multiple domains, such as healthcare, finance, and customer service. Despite their poten…

ADC: Adversarial attacks against object Detection that evade Context consistency checks

2021-10-24 · Mingjun Yin, Shasha Li, Chengyu Song, M. Salman Asif 외

Deep Neural Networks (DNNs) have been shown to be vulnerable to adversarial examples, which are slightly perturbed input images which lead DNNs to make wrong predictions. To protect from such examples, various defense st…

Objectobject-detectionObject Detection

SST-EM: Advanced Metrics for Evaluating Semantic, Spatial and Temporal Aspects in Video Editing

2025-01-13 · Varun Biyyala, Bharat Chanderprakash Kathuria, Jialu Li, Youshan Zhang

Video editing models have advanced significantly, but evaluating their performance remains challenging. Traditional metrics, such as CLIP text and image scores, often fall short: text scores are limited by inadequate tra…

Objectobject-detectionObject DetectionObject Tracking+1