Paper

Fair and Calibrated Toxicity Detection with Robust Training and Abstention

Unreadunread

AI summary

arXiv:2605. 14074v1 Announce Type: new Abstract: Fairness in toxicity classification involves three integrated axes: ranking, calibration, and abstention.

Reading status

Priority

About this paper

Authors: Mokshit Surana
Released: 2026-05-15 (3mo ago)
Topic: Related to current project
Relation to my work: Read next because Fair and Calibrated Toxicity Detection with Robust Training and Abstention overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, under, eval, rate, compare. Source: arxiv cs.LG (Machine Learning).
arXiv: 2605.14074

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to Fair and Calibrated Toxicity Detection with Robust Training and Abstention. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.