AI Paper Atlas

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

Convergent World Representations and Divergent Tasks

Core Francisco Park

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
train on two tasks instead of onecity-coordinate transformer · layer 5 CKA across modelsCKA: 0.48 -> 0.84Improved
increase pretraining task count from 1 to 2 or 3city-coordinate transformer · layers 4-6 CKA, disjoint task setsaverage CKA: saturating at 0.85Improved
add a second task to crossingcrossing task · single-task traininglearnabilityImproved
include distance task in two-task fine-tuning mixture (ablation)7-task pretrained model · Atlantis fine-tuning, cross-task generalizationdeviation from best-teacherHurt
fine-tune on angle task alone7-task pretrained model · Atlantis integrationcross-task improvementImproved
fine-tune on distance task alone7-task pretrained model · Atlantis integrationcross-task improvementNo clear effect
fine-tune with divergent (distance) task data (ablation)7-task pretrained model · Atlantis coordinate probe reconstructionreconstruction error: nearly an order of magnitude largerHurt
include Atlantis cities from start of pretraining instead of fine-tuning (ablation)7-task model · Atlantis coordinate probe reconstructionreconstruction errorImproved
use batch size 512 instead of 128 in fine-tuning (ablation)fine-tuned models · fine-tuningtask performanceHurt

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