surface↓classification↓narrative_llm

Classificationdimension · returns varchar

NARRATIVE_LLM

LLM-backed narrative-frame identification (escape hatch for NARRATIVE)

Per-row classifier — stable across GROUP BY.

classificationllmllm-escape-hatchtext

Arguments

name	type	description
text	VARCHAR	—
context	VARCHAR	—

About

LLM-backed alternative to NARRATIVE for identifying which story structure or rhetorical frame an author is using. Reach for this when the base NLI operator misses subtle framings — ironic framings, unstated comparisons, or tropes that need multi-sentence context to recognize. Useful for media-analysis work where the narrative frame is intentionally veiled (advocacy journalism, satire, editorials). The LLM can follow the rhetorical setup across paragraphs; the NLI path works from a short hypothesis template and can miss the shape. For high-volume routine tagging against a standard frame set, use NARRATIVE — much cheaper per row.

Nearby rabbit holes

same domain

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext

Climb back to The Looking Glass

surface↓classification↓narrative_llm

Classificationdimension · returns varchar

NARRATIVE_LLM

LLM-backed narrative-frame identification (escape hatch for NARRATIVE)

Per-row classifier — stable across GROUP BY.

classificationllmllm-escape-hatchtext

Arguments

name	type	description
text	VARCHAR	—
context	VARCHAR	—

About

LLM-backed alternative to NARRATIVE for identifying which story structure or rhetorical frame an author is using. Reach for this when the base NLI operator misses subtle framings — ironic framings, unstated comparisons, or tropes that need multi-sentence context to recognize. Useful for media-analysis work where the narrative frame is intentionally veiled (advocacy journalism, satire, editorials). The LLM can follow the rhetorical setup across paragraphs; the NLI path works from a short hypothesis template and can miss the shape. For high-volume routine tagging against a standard frame set, use NARRATIVE — much cheaper per row.

Nearby rabbit holes

same domain

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext

Climb back to The Looking Glass