AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Incentivizing Generative Zero-Shot Learning via Outcome-Reward Reinforcement Learning with Visual Cues

Wenjin Hou, Xiaoxiao Sun, Hehe Fan

CVPR 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot · 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
add outcome-reward RL (ablation)RLVC · AWA2CZSL Acc and GZSL H (%): 79.4/73.9 -> 84.0/80.4Improved
add visual cues with prototype-distillation loss (ablation)RLVC · AWA2CZSL Acc (%): 74.9 -> 84.0Improved
fine-tune the visual encoder with the reward model (ablation)RLVC · CUB, SUN, AWA2GZSL H (%): H +3.7%, +1.2%, +4.1%Improved
use EMA-baseline advantage instead of raw reward (ablation)RLVC · AWA2GZSL H (%): 78.2 -> 80.4Improved
add RL and visual cues to vanilla diffusion-GAN generator (attribute vectors) (ablation)Vanilla model · CUBGZSL H (%): 75.1 -> 81.2Improved
add RL and visual cues to vanilla generator (class-name word embeddings) (ablation)Vanilla model · CUB, SUN, AWA2CZSL Acc (%): +1.0, +0.1, +1.4Improved
use prototype-distillation loss instead of KL (ablation)RLVC · CUB, SUN, AWA2CZSL Acc (%): CUB 88.9 -> 90.1; AWA2 76.4 -> 84.0Improved
use prototype-distillation loss instead of l1 (ablation)RLVC · AWA2CZSL Acc (%): 75.9 -> 84.0Improved
use prototype-distillation loss instead of KL (ablation)RLVC · SUNGZSL H (%): 58.2 -> 57.6Hurt
start RL after 30 cold-start epochs (ablation)RLVC · CUBaccuracyImproved
set visual loss coefficient to 20 (ablation)RLVC · CUBaccuracyImproved
synthesize 400 unseen samples per class (ablation)RLVC · CUBGZSL H (%)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.

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