AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

What are the best Systems? New Perspectives on NLP Benchmarking

Pierre Colombo, Nathan Noiry, Ekhine Irurozki, Stephan CLEMENCON

NeurIPS 2022 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 20 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use two-level Borda ranking aggregation instead of mean aggregationsystem ranking · synthetic corrupted taskscorrupted tasks needed to reach error 0.75: 2 -> 10Improved
use two-level instead of one-level Borda aggregation of instance-level rankings (ablation)Kemeny/Borda ranking · synthetic corrupted taskscorrupted tasks needed to reach error 0.75: 5 -> 10Improved
use ranking-based aggregation instead of mean when one task is rescaledsystem ranking · synthetic rescalingranking error: error >90% at factor 2 for mean; unaffected for rankingImproved
use Borda rank aggregation instead of mean on task-level benchmarkssystem ranking · GLUE/SGLUE/XTREMEranking agreement with mean (Kendall tau): tau 0.82 to 0.92Mixed
use Borda rank aggregation instead of mean when randomly dropping taskssystem ranking · GLUE task subsamplingKendall correlation to full-task rankingImproved
use two-level or one-level Borda instead of mean aggregation on NLG instance-level data when removing metricssystem ranking · NLG datasets (TAC, FLICKR, etc.)Kendall correlation to full rankingImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy