See2Think: Do Multimodal Models Really Use Intermediate Visual States?
Quick summary
arXiv:2607.26769v1 Announce Type: cross Abstract: Multimodal large language models increasingly use sketches, annotations, tools, and intermediate images during reasoning, but it remains unclear whether they truly rely on these visual states. Existing benchmarks are limited both by task collections with narrow coverage or partially text-solvable samples and by evaluations that emphasize final answers without diagnosing how intermediate visual states are generated, rendered, and used. We introduce See2Think, a unified evaluation framework comprising See2ThinkBench and Visual Action-of-Thought (
Key takeaways
- arXiv:2607.26769v1 Announce Type: cross Abstract: Multimodal large language models increasingly use sketches, annotations, tools, and intermediate images during reasoning, but it remains unclear whether they truly rely on these visual states.
- Existing benchmarks are limited both by task collections with narrow coverage or partially text-solvable samples and by evaluations that emphasize final answers without diagnosing how intermediate visual states are generated, rendered, and used.
- We introduce See2Think, a unified evaluation framework comprising See2ThinkBench and Visual Action-of-Thought (
Why it matters
“See2Think: Do Multimodal Models Really Use Intermediate Visual States?” should be evaluated beyond branding and benchmark scores. Its practical importance will emerge in task accuracy, latency, unit cost, safety and integration with real workflows.