@article{RoseWeiVandycketal., author = {Rose, Adam and Wei, Dan and Vandyck, Toon and Flachsland, Christian}, title = {Policy Brief—Achieving Paris Climate Agreement Pledges: Alternative Designs for Linking Emissions Trading Systems}, series = {Review of Environmental Economics and Policy}, volume = {12}, journal = {Review of Environmental Economics and Policy}, number = {1}, issn = {1750-6816}, doi = {10.1093/reep/rex029}, pages = {170 -- 182}, abstract = {The coordination of international climate policy, such as linking systems of tradable greenhouse gas (GHG) emissions allowances, can greatly lower the cost to all participants of slowing climate change. We consider alternative policy designs of international agreements that would help implement the 2015 Paris Climate Agreement's GHG reduction pledges. In particular, we examine a stepwise approach to implementing a global system of GHG emissions trading, which includes estimating the benefits of alternative configurations of participating countries. We also illustrate the importance of allowance trading design features, such as the transfer of auction revenues to low-income countries. Numerical simulations indicate that an emissions trading system covering the power and industry sectors in all countries that made unconditional pledges could reduce the associated mitigation costs by more than 72 percent. Moreover, transferring the revenues from the sale of emission allowances could greatly enhance the capability of lower-income countries to meet their Paris Agreement pledges.}, language = {en} } @techreport{BeanKearnsRomanouetal., type = {Working Paper}, author = {Bean, Andrew M. and Kearns, Ryan Othniel and Romanou, Angelika and Hafner, Franziska Sofia and Mayne, Harry and Batzner, Jan and Foroutan, Negar and Schmitz, Chris and Korgul, Karolina and Batra, Hunar and Deb, Oishi and Beharry, Emma and Emde, Cornelius and Foster, Thomas and Gausen, Anna and Grandury, Mar{\´i}a and Han, Simeng and Hofmann, Valentin and Ibrahim, Lujain and Kim, Hazel and Kirk, Hannah Rose and Lin, Fangru and Liu, Gabrielle Kaili-May and Luettgau, Lennart and Magomere, Jabez and Rystr{\o}m, Jonathan and Sotnikova, Anna and Yang, Yushi and Zhao, Yilun and Bibi, Adel and Bosselut, Antoine and Clark, Ronald and Cohan, Arman and Foerster, Jakob and Gal, Yarin and Hale, Scott A. and Raji, Inioluwa Deborah and Summerfield, Christopher and Torr, Philip H. S. and Ududec, Cozmin and Rocher, Luc and Mahdi, Adam}, title = {Measuring what Matters: Construct Validity in Large Language Model Benchmarks}, publisher = {arXiv}, doi = {10.48550/arXiv.2511.04703}, pages = {29}, abstract = {Evaluating large language models (LLMs) is crucial for both assessing their capabilities and identifying safety or robustness issues prior to deployment. Reliably measuring abstract and complex phenomena such as 'safety' and 'robustness' requires strong construct validity, that is, having measures that represent what matters to the phenomenon. With a team of 29 expert reviewers, we conduct a systematic review of 445 LLM benchmarks from leading conferences in natural language processing and machine learning. Across the reviewed articles, we find patterns related to the measured phenomena, tasks, and scoring metrics which undermine the validity of the resulting claims. To address these shortcomings, we provide eight key recommendations and detailed actionable guidance to researchers and practitioners in developing LLM benchmarks.}, language = {en} }