AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

The Truth Lies Somewhere in the Middle (of the Generated Tokens)

Sophie L. Wang, Phillip Isola, Brian Cheung

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 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
mean-pool hidden states over generated tokens instead of using any single token (ablation)Qwen3-14B generated-token embeddings · WIT vision-language alignmentCKA alignmentImproved
mean pooling versus attention or max pooling (ablation)generated-token embeddings · WIT vision-language alignmentCKA alignmentImproved
mix representations from different token segments via convex combinations (ablation)generated-token embeddings · WIT vision-language alignmentCKA alignmentImproved
mix generated-token representations on reasoning and protein tasks (ablation)Qwen3-14B · MATH-500, GPQA Diamond, UniProtCKA alignmentImproved
use generated tokens instead of prompt tokens (ablation)Qwen3-14B mean-pooled embeddings · WIT vision-language alignmentCKA alignment: 0.184 -> 0.410Improved
mean pool prompt tokens instead of using later individual prompt tokens (ablation)prompt-token embeddings · WIT vision-language alignmentCKA alignmentNo clear effect
mix representations across independent decoding seeds (ablation)generated-token embeddings · WIT vision-language alignmentCKA alignmentImproved
mix region-conditioned generations describing distinct parts of an image (ablation)generated-token embeddings · DCI vision-language alignmentCKA alignmentImproved
add isotropic noise around the mean embedding / shuffle image-text pairs as variance-reduction control (ablation)token-mean representation · WIT vision-language alignmentCKA alignmentNo clear effect
prepend the phrase 'Let me recall what I know' to a non-thinking generation (ablation)Qwen3-14B generation · WIT vision-language alignmenttokenwise alignmentImproved
embed text generated by a different model (ablation)Qwen3-14B and OLMo3-7B-Think · WIT vision-language alignmenttokenwise alignmentHurt
embed Qwen3-generated text with OLMo3 instead of OLMo3's own generations (ablation)OLMo3-7B-Think embeddings · WIT vision-language alignmentCKA alignmentImproved

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