Chain-of-Thought reasoning in LLMs is often unfaithful, with models relying on it variably by task and less so as models scale larger.
URL https: //www.science.org/doi/abs/10.1126/sc irobotics.aay7120
4 Pith papers cite this work, alongside 1,903 external citations. Polarity classification is still indexing.
4
Pith papers citing it
1,903
external citations · OpenAlex
representative citing papers
The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job loss and environmental costs.
The paper establishes a reproducible retrospective benchmark for ranking daily active-fire detections in Cerrado conservation units by comparing atmospheric, surface, static spatial, and short-term memory covariates with standard ML models under time-series cross-validation and held-out AOI tests.
citing papers explorer
-
Measuring Faithfulness in Chain-of-Thought Reasoning
Chain-of-Thought reasoning in LLMs is often unfaithful, with models relying on it variably by task and less so as models scale larger.
-
Ethical and social risks of harm from Language Models
The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job loss and environmental costs.
-
A Retrospective Benchmark of Spatiotemporal Covariates for Daily Active-Fire Detection in Cerrado Conservation Units
The paper establishes a reproducible retrospective benchmark for ranking daily active-fire detections in Cerrado conservation units by comparing atmospheric, surface, static spatial, and short-term memory covariates with standard ML models under time-series cross-validation and held-out AOI tests.
- Enhancing AI Interpretability with Localised Architectures