AI Paper Atlas

Research map / Large language models / Agents, code and math

LLM as Dataset Analyst: Subpopulation Structure Discovery with Large Language Model

Yulin Luo, Ruichuan An, Bocheng Zou, Yiming Tang, Jiaming Liu, Shanghang Zhang

ECCV 2024 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 25 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 Criteria Refinement via self-consistency (ablation)SSD-LLM · Stanford 40 Action (Action)clustering accuracy (%): 75.6 -> 81.7Improved
reduce dimension suggestion rounds from 10 to 5 (ablation)SSD-LLM · Stanford 40 Action (Action)clustering accuracy (%): 75.6 -> 65.2Hurt
increase dimension suggestion rounds from 10 to 20 (ablation)SSD-LLM · Stanford 40 Action (Action)clustering accuracy (%): 75.6 -> 75.9No clear effect
use more captions per attribute-summarization round (40 vs 20) (ablation)SSD-LLM · Stanford 40 Action (Action)clustering accuracy (%): 75.6 -> 70.3Hurt
use more rounds with fewer captions per round (40x10) for attributes (ablation)SSD-LLM · Stanford 40 Action (Action)clustering accuracy (%): 75.6 -> 80.2Improved
use SSD-LLM-discovered attributes in generation prompts instead of unknown attributes (ablation)diffusion augmentation · Waterbirdsworst group accuracy: 71.3 -> 76.2Improved
sample underrepresented subpopulations with weights instead of randomly (ablation)diffusion augmentation · Waterbirdsworst group accuracy: 76.2 -> 77.4Improved
use LLM-generated sentences as diffusion prompts instead of direct attribute prompts (ablation)diffusion augmentation · Waterbirdsworst group accuracy: 77.4 -> 78.3Improved
add LLM-suggested extra dimensions and attributes (ablation)diffusion augmentation · Waterbirdsworst group accuracy: 78.3 -> 79.1Improved
double the number of generated images (ablation)diffusion augmentation · Waterbirdsworst group accuracy: 78.3 -> 79.2 (worst group), 90.1 -> 91.1 (average)No clear effect
use SSD-LLM for subpopulation shift instead of best prior methodimage classification · Waterbirds, Metashift, Nico++worst group accuracy (average): 58.7 -> 62.0Improved
use SSD-LLM for slice discovery instead of DominoImageNet-1K · ImageNet-1K slice discoverytopic error rate (%): 43.44 -> 47.39Improved

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