AI Paper Atlas

Research map / Benchmarks / CivilComments

CivilComments leaderboard

22 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on CivilComments. Best worst-group accuracy: 80.4 (CB last-layer retraining, NeurIPS 2023).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

CivilComments · worst-group accuracy

22 papers, each with its best reported number for its own method; higher is better.

#worst-group accuracy ↑MethodPaperTopic
180.4CB last-layer retraining
BERT
Towards Last-layer Retraining for Group Robustness with Fewer Annotations
NeurIPS 2023 · Table: Last-layer retraining o
RNNs and normalization
279.27LA/TTLSABeyond Invariance: Test-Time Label-Shift Adaptation for Addressing "Spurious" Correlations
NeurIPS 2023 · Table 2
Test-time adaptation
378.5Ours
JTT
A Simple Remedy for Dataset Bias via Self-Influence: A Mislabeled Sample Perspective
NeurIPS 2024 · Appendix NLP results table
Recognition and long-tail
477.83PfR
Llama-2-7B/13B
Prompting is a Double-Edged Sword: Improving Worst-Group Robustness of Foundation Models
ICML 2024 · Table 2
Zero-/few-shot and meta-learning
577.6Mixture 3:1The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
NeurIPS 2024 · Table: Mixture balancing is ro
Recognition and long-tail
675CCDB
BERT
Class-Conditional Distribution Balancing for Group Robust Classification
ICML 2026 · Table 2
Recognition and long-tail
774SCER
BERT
Spurious Correlation-Aware Embedding Regularization for Worst-Group Robustness
ICLR 2026 · Table 5
Recognition and long-tail
872.6LISA
DistilBERT-uncased
Improving Out-of-Distribution Robustness via Selective Augmentation
ICML 2022 · Table 2
Test-time adaptation
972.5GIC_Cy-MImproving Group Robustness on Spurious Correlation Requires Preciser Group Inference
ICML 2024 · Table 1
Recognition and long-tail
1072.2GroupDRO + XRMDiscovering Environments with XRM
ICML 2024 · Table: worst-group accuracies
OOD and anomaly detection
12 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.