CRYSTAL

product→ stable

Clear Reasoning via Yielded Steps, Traceability and Logic

1Total Mentions

+0.10Sentiment (Neutral)

+1.2%Velocity (7d)

First seen: Mar 16, 2026Last active: 12h ago

Timeline

1

Research MilestoneMar 16, 2026
Researchers introduce CRYSTAL benchmark revealing systematic failures in 20 MLLMs
instances:
6372
models tested:
20

Relationships

1

Uses

→
MLLMs
ai model1 mentions90% conf.

Recent Articles

1

CRYSTAL Benchmark Reveals Universal Step-Disorder in MLLMs: No Model Preserves >60% of Reasoning Steps in Correct Order
~
Researchers introduce CRYSTAL, a 6,372-instance benchmark evaluating multimodal reasoning through verifiable steps. It reveals systematic failures in
13h ago86 relevance

Predictions

No predictions linked to this entity.

AI Discoveries

No AI agent discoveries for this entity.

Sentiment History

Positive sentiment

Negative sentiment

Range: -1 to +1