paper-with-me

Papers

Conditional probing: measuring usable information beyond a baseline

2021-09-19 · EMNLP 2021 11 · John Hewitt, Kawin Ethayarajh, Percy Liang, Christopher D. Manning

Probing experiments investigate the extent to which neural representations make properties -- like part-of-speech -- predictable. One suggests that a representation encodes a property if probing that representation produces higher accuracy than probing a baseline representation like non-contextual word embeddings. Instead of using baselines as a point of comparison, we're interested in measuring information that is contained in the representation but not in the baseline. For example, current methods can detect when a representation is more useful than the word identity (a baseline) for predicting part-of-speech; however, they cannot detect when the representation is predictive of just the aspects of part-of-speech not explainable by the word identity. In this work, we extend a theory of usable information called $\mathcal{V}$-information and propose conditional probing, which explicitly conditions on the information in the baseline. In a case study, we find that after conditioning on non-contextual word embeddings, properties like part-of-speech are accessible at deeper layers of a network than previously thought.

📄 PDF Abstract BibTeX arXiv:2109.09234

Code (1)

john-hewitt/conditional-probing 공식 구현 pytorch

Tasks

Word Embeddings

Similar Papers 제목 키워드 기반

Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck

2026-05-08 · Zihan Huang, Junda Wu, Tong Yu, Qianqi Yan 외 arxiv

While LLM-based agents excel at planning and executing long action sequences, their execution often remains inconsistent across trials, limiting reliability. Consolidating agent consistency requires distilling trial-erro…

Measuring Memorization Effect in Word-Level Neural Networks Probing

2020-06-29 · Rudolf Rosa, Tomáš Musil, David Mareček

Multiple studies have probed representations emerging in neural networks trained for end-to-end NLP tasks and examined what word-level linguistic information may be encoded in the representations. In classical probing, a…

Machine TranslationMemorizationTranslation

Understanding Probe Behaviors through Variational Bounds of Mutual Information

2023-12-15 · Kwanghee Choi, Jee-weon Jung, Shinji Watanabe

With the success of self-supervised representations, researchers seek a better understanding of the information encapsulated within a representation. Among various interpretability methods, we focus on classification-bas…

Beyond Binary Instrument QA: Probing Instrument Grounding in Music Audio-Language Models

2026-06-30 · Yujun Lee, Joonhyeok Shin, Hyoeun Kim, Kyuhong Shim arxiv

Recent music audio-language models achieve high accuracy on instrument question-answering benchmarks, but it remains unclear whether this reflects robust audio grounding or benchmark-specific shortcuts. In this paper, we…

Amnesic Probing: Behavioral Explanation with Amnesic Counterfactuals

2020-06-01 · Yanai Elazar, Shauli Ravfogel, Alon Jacovi, Yoav Goldberg

A growing body of work makes use of probing to investigate the working of neural models, often considered black boxes. Recently, an ongoing debate emerged surrounding the limitations of the probing paradigm. In this work…