Slides: Task Fidelity Scaling Laws — Kobie Crawdord, Snorkel

Source Video

Task Fidelity Scaling Laws — Kobie Crawdord, Snorkel

Relationship To World's Fair 2026

These slides are extracted from a public AI Engineer YouTube video connected to World's Fair 2026. Speaker-matched clips are supporting context unless later confirmed as exact session recordings; official livestream recordings are day-level/event-level source material.

Related Scheduled Sessions

• No individual scheduled session mapping has been assigned yet; treat this as an event livestream deck.

Extracted Slides

OCR text:

Snorkel

TheFrontier AData ab

Task Fidelity Scaling Laws

Fine-tuning onhigh-quality tasks

dramatically outperforms fine-tuning on

low-quality tasks

Kobie Crawford,Developer Advocate

AlEngineer

EUROPE

OCR text:

a

SS SnorkeJ Tne Fronter Al Data Lab

rs : F .

Task Fidelity Scaling Laws

% rs

* ad

a od

Fine tuning on high quauity tasks 7

dramaticaily outcerforms fine tuning on

low-quality tasks

Koite Crawford, Developer Advocate

|

| a Google DeepMind

OCR text:

Does Task Quality Actually Matter?

e Almodel capabilities are fundamentally bounded by training data quality

— this holds regardless of model architecture, scale, or agent harness.

p ee . e For agentic benchmarks and evals, task quality is data quality

pe " — but the field currently lacks empirical evidence that curating higher-quality tasks

PY ry leads to meaningfully better training outcomes,

ad " *

Can we measure the impact of task quality on model performance?

a 2

w) | Al Engineer

[aed

J EUROPE

rs

OCR text:

S&S SnorkeJ The Frontier A! Data Lab

Thank You

* 7 [m] cA [nm] OFS 3 [m)

a op tls

- TL pe

eles la

snorkel.ai/research/ snorkel.ai/leaderboard/

_ ) . on Engineering the future of Al

Slide-Derived Subjects To Review

Subject extraction uses video title, related session titles/descriptions, transcript context, and OCR text when available. OCR is best-effort and should be reviewed against the embedded slide images.