🛰️ Daily AI Frontier
‹ back to 2026-08-27

TraceML: An Empirical Analysis of Human-Agent Planning in Machine Learning Development

Research LLM Agents

Ranking

Overall 79
Content 95
Popularity 42

Observed public metrics from 1 member.

Representative image for TraceML: An Empirical Analysis of Human-Agent Planning in Machine Learning Development

Merged summary

TL;DR - TraceML is a process-level benchmark comparing human and AI-agent trajectories on machine-learning competitions. It shows that agents underperform partly because they repeat narrow optimization loops rather than pivoting, revisiting abandoned ideas, and coordinating diverse development activities like experts.

  • The dataset contains 4,465 human Kaggle trajectories across 134 competitions, including 430 human and 207 agent trajectories on seven shared competitions.
  • Each code version is annotated with score, timestamp, action, intent, edit size, and score effect, enabling analysis beyond final benchmark outcomes.
  • Codex focused heavily on ensemble reweighting and submission tuning, while MLEvolve repeatedly mutated its current model; neither pivoted or reopened prior approaches as often as humans.
  • A planning prompt distilled from human behavior improved targeted behaviors and scores, but did not eliminate the agents’ broader planning-pattern gap.

Sources (1)

TraceML: An Empirical Analysis of Human-Agent Planning in Machine Learning Development

arXiv cs.LG Jiarui Yan, Weiwei Sun, Sijie Li, Wenhan Li, Yiming Yang 2026-08-26 arXiv:2608.26086
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-12 14:20:10.949499 UTC

TL;DR - TraceML is a process-level benchmark comparing human and AI-agent trajectories on machine-learning competitions. It shows that agents underperform partly because they repeat narrow optimization loops rather than pivoting, revisiting abandoned ideas, and coordinating diverse development activities like experts.

  • The dataset contains 4,465 human Kaggle trajectories across 134 competitions, including 430 human and 207 agent trajectories on seven shared competitions.
  • Each code version is annotated with score, timestamp, action, intent, edit size, and score effect, enabling analysis beyond final benchmark outcomes.
  • Codex focused heavily on ensemble reweighting and submission tuning, while MLEvolve repeatedly mutated its current model; neither pivoted or reopened prior approaches as often as humans.
  • A planning prompt distilled from human behavior improved targeted behaviors and scores, but did not eliminate the agents’ broader planning-pattern gap.
item →