A model self-improves on math reasoning by majority-voting its own rollouts, using the winning answer as a pseudo-solution, and distilling itself on the disagreeing rollouts, without any labeled data.
Vision Language Models Cannot Reason About Physical Transformation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Understanding physical transformations is fundamental for reasoning in dynamic environments. While Vision Language Models (VLMs) show promise in embodied applications, whether they genuinely understand physical transformations remains unclear. We introduce ConservationBench evaluating conservation -- whether physical quantities remain invariant under transformations. Spanning four properties with paired conserving/non-conserving scenarios, we generate and evaluate 23,040 questions across 112 VLMs. Results reveal systematic failure: performance remains near chance with improvements on conservation tasks accompanied by drops on controls. Control experiments show strong textual priors favoring invariance, yet models perform worse with actual visual content when performance is balanced across conserving and non-conserving scenarios. Neither temporal resolution, prompting, nor curated sampling helps. These findings show that current VLMs fail to maintain transformation-invariant representations of physical properties across dynamic scenes.
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On-Policy Self-Distillation without Any Supervision
A model self-improves on math reasoning by majority-voting its own rollouts, using the winning answer as a pseudo-solution, and distilling itself on the disagreeing rollouts, without any labeled data.