Papers with Code 아카이브
Known Unknowns
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 0
- 보존된 평가 기록
- 0
Language models have a tendency to generate text containing false statements that are often referred to as "Hallucinations." The primary purpose of this task is to test for this failure case by probing whether a model can correctly identify that the answer to a question is unknown. A common failure mode would be to prefer a prediction of false on unknown truth over a prediction that the answer is unknown.
Source: BIG-bench
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
Researchy Questions: A Dataset of Multi-Perspective, Decompositional Questions for LLM Web Agents
High-dimensional forecasting with known knowns and known unknowns
The known unknowns of the Hsp90 chaperone
Machine learning for advancing low-temperature plasma modeling and simulation
Knowledge of Knowledge: Exploring Known-Unknowns Uncertainty with Large Language Models
PaLM: Scaling Language Modeling with Pathways
Training Compute-Optimal Large Language Models
Scaling Language Models: Methods, Analysis & Insights from Training Gopher
Domain Concretization from Examples: Addressing Missing Domain Knowledge via Robust Planning
Generative ODE Modeling with Known Unknowns
The division of labor in communication: Speakers help listeners account for asymmetries in visual perspective
Classification Uncertainty of Deep Neural Networks Based on Gradient Information
Toward Open-Set Face Recognition
Known Unknowns: Uncertainty Quality in Bayesian Neural Networks
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.