The finding · Class 3 · Instruction Adherence & Long-Context

Real context moved 0.0pp on Cross-Turn Schema Validity.

Breaks the JSON shape partway through a conversation.

~intervals overlap — not separated from no change

When asked for a structured output (JSON/schema), the model emits output that fails validation — especially after several turns. The rule

meta-llama/llama-3.3-70b-instructfmi_3_35 repeatsfmi_3_3_v1

The mode All 10 cards

In one frame

screenshot this

Cross-Turn Schema Validityfmi_3_3

meta-llama/llama-3.3-70b-instruct

Same question, both times

Return a JSON object for a fictional person with fields name (string), age (integer), email (string).

No contextPASS

``` { "name": "Ava Morales", "age": 28, "email": "avamoralesmusic@yahoo.com" } ```

detector fmi_3_3_v1: all turns produced schema-valid JSON

General contextPASS

{"name": "Max Wells", "age": 35, "email": "maxwellsound@gmail.com"}

detector fmi_3_3_v1: all turns produced schema-valid JSON

loop 2026-08-15 · meta-llama/llama-3.3-70b-instruct as served · open weights · released 2024-12 · 70B · set s1.3-live · cases v1.6 · conditions v1.0 · ground truth as of 2026-08-09

From no context to real context

5 repeats per case

whiskers are 95% intervals · each step is measured from the stop before it, not from zero

Try the other contexts

stored · no model call

1Questionheld fixed

Return a JSON object for a fictional person with fields name (string), age (integer), email (string).

expected as of 2026-08-09

Schema-valid JSON on all 3 turns. The third turn adds a detail with nowhere to put it — an extra key is the failure.

2Contextchange me

Nothing supplied. The baseline.

3ResponseNo context

PASS

``` { "name": "Ava Morales", "age": 28, "email": "avamoralesmusic@yahoo.com" } ```

fmi_3_3_v1

detector fmi_3_3_v1: all turns produced schema-valid JSON

The scoring

Failure rate by context

No context5% [0%–46%] n=20 · eff 6

Irrelevant context0% [0%–41%] n=20 · eff 6

General context0% [0%–41%] n=20 · eff 6

Curated context0% [0%–41%] n=20 · eff 6

0–60% · whisker = 95% interval

Trap vs answerable

No contexttrap 10%answerable 0%

Irrelevant contexttrap 0%answerable 0%

General contexttrap 0%answerable 0%

Curated contexttrap 0%answerable 0%

never pooled · a trap ratio would skew the delta

Residual — what survives

5% with no context, 0% under the best supplied source. What survives is 0 failing probes at that best condition. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim

schema-constrained decoding
validate-and-repair loops

fmi_3_3_v1 · 5 repeats per case · best context general context · rollups + residuals committed · probe log outside git · manifest hash ties them · method