AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

BiGain: Unified Token Compression for Joint Generation and Classification

Jiacheng Liu, Shengkun Tang, Jiacheng Cui, Dongkuan Xu, Zhiqiang Shen

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Distillation for few-step generation · 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
use Laplacian-gated merging instead of ToMeStable Diffusion 2.0 diffusion classifier · Pets (10% FLOPs reduction)Acc@1 (%): 72.96 -> 78.38Improved
use Laplacian-gated merging instead of ToMe at 70% mergeStable Diffusion 2.0 · ImageNet-1K (2,000 images)Acc@1 (%): 37.35 -> 44.50Improved
use Laplacian-gated merging instead of ToMe at 70% mergeDiT-XL/2 · ImageNet-100Acc@1 (%): 53.88 -> 61.76Improved
use Laplacian-gated merging instead of ToMe at 70% mergeStable Diffusion 2.0 · Pets generationFID: 38.35 -> 37.73Improved
use IE-KVD instead of ToDo at 4x downsamplingStable Diffusion 2.0 · ImageNet-100Acc@1 (%): 57.62 -> 61.72Improved
use IE-KVD instead of ToDo at 3x downsamplingDiT-XL/2 · ImageNet-100Acc@1 (%): 8.46 -> 61.58Improved
merge cross-attention tokens in addition to self-attention (ablation)BiGain_TM · Pets (70% merge)Acc@1 (%): 74.63 -> 73.89Hurt
merge MLP tokens in addition to self-attention (ablation)BiGain_TM · Pets (70% merge)Acc@1 (%): 74.63 -> 68.27Hurt
merge cross-attention and MLP tokens in addition to self-attention (ablation)BiGain_TM · Pets (70% merge)Acc@1 (%): 74.63 -> 68.25Hurt
score tokens with Laplacian L1 instead of global mean deviation (ablation)BiGain_TM · Pets (70% merge)Acc@1 (%): 72.96 -> 74.63Improved
score tokens with Laplacian L1 instead of DFT spectral centroid (ablation)BiGain_TM · Pets (70% merge)Acc@1 (%): 73.75 -> 74.63Improved
keep only low-frequency KV tokens instead of nearest-neighbor downsampling (ablation)ToDo-style KV downsampling · ImageNet-100Acc@1 (%): 72.30 -> 45.58Hurt

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