Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning To Hallucinate Examples From Extrinsic and Intrinsic Supervision
Liangke Gui, Adrien Bardes, Ruslan Salakhutdinov, Alexander Hauptmann, Martial Hebert, Yu-Xiong Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add intrinsic contrastive supervision (DMAS w/ in) to a cosine classifier with plain features (ablation) | Cosine Classifier · ImageNet1K 311-way, 1-shot | top-5 accuracy: 37.8 -> 43.4 | Improved |
| add extrinsic mentor guidance (DMAS w/ ex) to the cosine classifier (ablation) | Cosine Classifier · ImageNet1K 311-way, 1-shot | top-5 accuracy: 37.8 -> 44.5 | Improved |
| add mentor guidance to the plain hallucinator in place of classification loss only (ablation) | Cosine Classifier · ImageNet1K 311-way, 1-shot | top-5 accuracy: 42.6 -> 44.5 | Improved |
| add progressive guidance by weakening the student on top of mentor guidance (ablation) | Cosine Classifier DMAS w/ ex · ImageNet1K 311-way, 1-shot | top-5 accuracy: 44.5 -> 45.4 | Improved |
| add progressive guidance by strengthening the mentor on top of mentor guidance (ablation) | Cosine Classifier DMAS w/ ex · ImageNet1K 311-way, 1-shot | top-5 accuracy: 44.5 -> 44.3 | No clear effect |
| weaken the student progressively instead of strengthening the mentor (ablation) | Cosine Classifier DMAS · ImageNet1K 311-way, 1-shot | top-5 accuracy: 44.3 -> 45.4 | Improved |
| add intrinsic contrastive loss to mentor-guided progressive training (full DMAS) (ablation) | Cosine Classifier DMAS w/ ex↓ · ImageNet1K 311-way, 1-shot | top-5 accuracy: 45.4 -> 46.5 | Improved |
| use DMAS hallucinator instead of plain hallucinator with ProtoNet (ablation) | ProtoNet · ImageNet1K 311-way, 1-shot | top-5 accuracy: 45.0 -> 45.9 | Improved |
| pretrain feature backbone with cosine classifier instead of standard cross-entropy, then apply DMAS (ablation) | Cosine Classifier DMAS (full) · ImageNet1K 311-way, 1-shot | top-5 accuracy: 46.5 -> 47.9 | Improved |
| learned hallucination instead of hand-crafted data augmentation (ablation) | ProtoNet · ImageNet1K 311-way, 1-shot | top-5 accuracy: 40.2 -> 45.0 | Improved |
| add plain hallucinator Gen to DeepEMD | DeepEMD · miniImageNet 5-way 1-shot | accuracy: 65.91 -> 64.73 | Hurt |
| add DMAS hallucinator to MetaOptNet instead of plain Gen | MetaOptNet · miniImageNet 5-way 1-shot | accuracy: 62.64 -> 64.55 | Improved |
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.
- DeepEMD — DeepEMD: Few-Shot Image Classification With Differentiable Earth Mover's Distance and Structured Classifiers (CVPR 2020) · better on 6 of 6 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
- Beyond Token Probes: Hallucination Detection via Activation Tensors with ACT-ViT (NeurIPS 2025 · similarity 0.88)
- No Free Lunch: Fundamental Limits of Learning Non-Hallucinating Generative Models (ICLR 2025 · similarity 0.87)
- HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection (NeurIPS 2024 · similarity 0.87)
- Hallucination-aware Intermediate Representation Editing in Large Vision-Lanugage Models (ICLR 2026 · similarity 0.87)
- Robust Hallucination Detection in LLMs via Adaptive Token Selection (NeurIPS 2025 · similarity 0.86)
- Steer LLM Latents for Hallucination Detection (ICML 2025 · similarity 0.86)
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.
