AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Uncertainty-Aware Knowledge Distillation for Multimodal Large Language Models

Jingchen Sun, Shaobo Han, Deep Patel, Wataru Kohno, Can Jin, Changyou Chen

CVPR 2026 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 1 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
add task-level Beta-KD weighting over manual weights (ablation)CE + FKL · ScienceQA testVQA accuracy: 48.6 -> 49.3Improved
add instance-level Beta-KD weighting (ablation)CE + FKL · ScienceQA testVQA accuracy: 48.2 -> 51.8Improved
instance-level Beta-KD on three-loss objectives (ablation)CE + KD + FD objectives · ScienceQA testVQA accuracy: +5.2 (FKL+FD) to -2.8 (TVD+FD)Mixed
task-level Beta-KD on RKL+FD (ablation)CE + RKL + FD · ScienceQA testVQA accuracy: 49.8 -> 49.5No clear effect
use instance-level rather than task-level weighting (ablation)Beta-KD · ScienceQA testVQA accuracy: e.g. 49.3 -> 51.8 (FKL); 53.1 -> 51.0 (SFKL)Mixed
match teacher in probability space with cosine distance instead of logit space (ablation)CE distillation · ScienceQA testVQA accuracy: 4.0 -> 47.2Improved
use Cosine-Probs instead of FKL as distillation loss (ablation)CE distillation · ScienceQA testVQA accuracy: 45.7 -> 47.2Improved
add Beta-KD (Instance) to Align-KDAlign-KD · 6-benchmark averageaverage: 63.4 -> 64.6Improved
add Beta-KD (Instance) to Cosine KDCosine KD · 6-benchmark averageaverage: 63.4 -> 65.5Improved
add Beta-KD (Instance) to LLaVA-KDLLaVA-KD Qwen2.5-0.5B · TextVQAaccuracy: 52.0 -> 54.9Improved
add Beta-KD on POPEAlign-KD · POPEaccuracy: 85.4 -> 85.4No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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