Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2603.19790 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915937574191104 |
|---|---|
| author | Gong, Weile Zuo, Yiping Lu, Zijian He, Xin Fan, Weibei Qi, Lianyong Jin, Shi |
| author_facet | Gong, Weile Zuo, Yiping Lu, Zijian He, Xin Fan, Weibei Qi, Lianyong Jin, Shi |
| contents | Modern vision-language models (VLMs) can act as generative OCR engines, yet open-ended decoding can expose rare but consequential failures. We identify a core deployment misalignment in generative OCR. Autoregressive decoding favors semantic plausibility, whereas OCR requires outputs that are visually grounded and geometrically verifiable. This mismatch produces severe errors, especially over-generation and unsupported substitutions, creating deployment risk even when benchmark accuracy remains high. We therefore formulate frozen VLM OCR as a selective accept/abstain problem and propose a model-agnostic Geometric Risk Controller. The controller probes multiple structured views of the same input, applies lightweight structural screening, and accepts a transcription only when cross-view consensus and stability satisfy predefined criteria, yielding a small family of operating points. Experiments on frozen VLM backbones and standard OCR benchmarks show consistent reductions in extreme-error risk and catastrophic over-generation at predictable coverage costs. Reliable deployment of generative OCR with frozen VLMs benefits from explicit system-level risk control rather than unconstrained generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_19790 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models Gong, Weile Zuo, Yiping Lu, Zijian He, Xin Fan, Weibei Qi, Lianyong Jin, Shi Computer Vision and Pattern Recognition Modern vision-language models (VLMs) can act as generative OCR engines, yet open-ended decoding can expose rare but consequential failures. We identify a core deployment misalignment in generative OCR. Autoregressive decoding favors semantic plausibility, whereas OCR requires outputs that are visually grounded and geometrically verifiable. This mismatch produces severe errors, especially over-generation and unsupported substitutions, creating deployment risk even when benchmark accuracy remains high. We therefore formulate frozen VLM OCR as a selective accept/abstain problem and propose a model-agnostic Geometric Risk Controller. The controller probes multiple structured views of the same input, applies lightweight structural screening, and accepts a transcription only when cross-view consensus and stability satisfy predefined criteria, yielding a small family of operating points. Experiments on frozen VLM backbones and standard OCR benchmarks show consistent reductions in extreme-error risk and catastrophic over-generation at predictable coverage costs. Reliable deployment of generative OCR with frozen VLMs benefits from explicit system-level risk control rather than unconstrained generation. |
| title | From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.19790 |