AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Learn to Categorize or Categorize to Learn? Self-Coding for Generalized Category Discovery

Sarah Rastegar, Hazel Doughty, Cees G. M. Snoek

NeurIPS 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 10 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 code contrastive loss L_C_code (ablation)InfoSieve (L_C_in baseline) · CUB-200accuracy (All): 66.8 -> 67.7Improved
add binary code constraint L_code_cond (ablation)InfoSieve · CUB-200accuracy (All): 67.7 -> 68.5Improved
add code length loss L_length (ablation)InfoSieve · CUB-200accuracy (Novel): 64.0 -> 64.5 novel; 77.5 -> 76.1 knownMixed
add mask binary constraint L_mask_cond (ablation)InfoSieve · CUB-200accuracy (Novel): 64.5 -> 62.3 novel; 76.1 -> 78.7 knownMixed
add categorizer loss L_Cat to full set of other losses (ablation)InfoSieve · CUB-200accuracy (All): 67.8 -> 69.4Improved
add L_Cat alone to contrastive input loss (ablation)L_C_in baseline · CUB-200accuracy (Known): 78.1 -> 76.4Hurt
increase mask constraint coefficient mu from 0.01 to 2 (ablation)InfoSieve · CUB-200accuracy (All): 69.4 -> 70.6Improved
vary code constraint coefficient zeta 0.01 to 2 (ablation)InfoSieve · CUB-200accuracy (All): 69.4 -> 69.9 (zeta=1)No clear effect
vary code length coefficient delta 0.01 to 2 (ablation)InfoSieve · CUB-200accuracy (All): 69.4 -> 69.8 (delta=2)No clear effect
vary code mapping coefficient eta 0.01 to 2 (ablation)InfoSieve · CUB-200accuracy (All): 69.4 -> 68.4 (eta=0.1)No clear effect
use InfoSieve instead of best prior method on fine-grained datasetsGCD benchmarks · CUB-200accuracy (All): 63.5 -> 69.4Improved
use InfoSieve instead of best prior method on coarse-grained datasetsGCD benchmarks · CIFAR-10accuracy (All): 97.9 -> 94.8Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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