AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

Intrinsic Gradient Suppression for Label-Noise Prompt Tuning in Vision–Language Models

Jia-yu Li, Jiaxin Qi, Sheng Zhou, Jianqiang Huang, Xian-Sheng Hua

ICML 2026 · Label-efficient and robust learning › Noisy labels · approach: Loss correction and noise robustness theory · 0 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
replace cross-entropy with double-softmax cross-entropyCoOp prompt tuning · Average of 9 datasets, 80% sym noiseaccuracy (%): 57.77 -> 74.22Improved
replace cross-entropy with double-softmax cross-entropyCoOp prompt tuning · FGVCAircraft, 80% sym noiseaccuracy (%): 12.27 -> 29.12Improved
use double-softmax instead of NLPrompt sample selection plus MAEprompt tuning with noisy labels · Flowers102, 40% sym noiseaccuracy (%): 93.29 -> 91.12Hurt
use double-softmax instead of NLPromptprompt tuning with noisy labels · Caltech101, 90% sym noiseaccuracy (%): 87.80 -> 86.68Hurt
use double-softmax instead of NLPromptprompt tuning with noisy labels · DTD, 80% pair noiseaccuracy (%): 30.93 -> 33.66Improved
use double-softmax instead of naive consistent-sample selection (ablation)prompt tuning with noisy labels · DTD, 80% pair noiseaccuracy (%): 5.44 -> 33.66Improved
use double-softmax instead of LogitNorm (ablation)prompt tuning with noisy labels · Caltech101, 60% sym noiseaccuracy (%): 82.14 -> 95.31Improved
use double-softmax instead of NCE loss (ablation)prompt tuning with noisy labels · OxfordPets, 30% pair noiseaccuracy (%): 71.91 -> 92.83Improved
use double-softmax instead of square logit normalization (ablation)prompt tuning with noisy labels · Caltech101, 30% pair noiseaccuracy (%): 81.59 -> 95.85Improved
use double-softmax instead of LogitClip with a fixed threshold (ablation)prompt tuning with noisy labels · Caltech101, 60% sym noiseaccuracy (%): 49.31 -> 95.31Improved
add double-softmax to suppress gradients of mislabeled samplesCoOp prompt tuning · Caltech101, 60% sym noisegradient magnitude on logitsImproved
tune LogitClip threshold tau (ablation)LogitClip · DTD sym noiseaccuracy (%): more than 15% dropHurt

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