Cactus-Compute releases Mirage vision model on Hugging Face
Trending at number 30, the repo ships mirage.safetensors and mirage.cact, downloaded on first use.
Stories per day · 30 days
Relateddocument-parsing4huggingface3vision-language3baidu2benchmark2object-detection2open-source2segmentation2cactus-compute1cer1datalab1document-ai1error-severity1evaluation-metrics1minicpm1model-release1multilingual1multimodal-models1
26 stories, newest first · one card per story
Stories tagged #ocr
Tencent Releases WeVisOne-VL Unified Vision Language Model
Tencent's WeVisOne-VL handles detection, OCR, segmentation and matting in one open model for 2B and 4B sizes.
1
Baidu open-sources Unlimited OCR 2 for long documents
The headline says the release targets hundred-page document parsing; no further specs were in the text.
Factorized vector quantization matches continuous visual tokens on understanding
The discrete model beats UniAR and Emu3-Chat by 22.8% and 59.2% on a nine-metric average.
Baidu releases Unlimited-OCR-2 for unbounded document parsing
SR-SWA keeps a fixed working set, such as 512 tokens, and the release adds the MarathonDocBench benchmark.
1
Developer’s Guide to OmniParseBench Leaderboard Benchmarks and OCR Testing
OmniParseBench leaderboard and OCR testing guide for document parsing.
WeVisOne-VL unifies object- and pixel-level visual perception
A pixel decoder behind an <OBJ> token emits boxes, masks, and mattes after curriculum SFT and mixed-task RL.
OpenBMB releases MiniCPM-OCR 1.0, a 1.3B document model
The publisher reports 93.68 on Real5-OmniDocBench and 1.2 pages per second on an RTX 5090.
OvisOCR3, a 0.8B model, covers five OCR task families
Weights are on Hugging Face; authors report 96.74 on OmniDocBench v1.6 and first place on OCRBench v2.
Datalab Releases OmniParseBench: An Open OCR Benchmark With 16,288 Pass/Fail Tests to Audit OCR Parsers
Datalab released OmniParseBench, an open OCR benchmark with 16,288 pass/fail parser tests.
2
Equal CER, Unequal Damage: Severity-Weighted Character Error Rate for OCR
semCER weights OCR character edits by judged severity, matching readers better than standard CER.
datalab-to/indra-ocr — new model trending #30 on Hugging Face
Datalab released Indra OCR, a 1M-parameter model converting documents at 49 pages per second on CPU.
When MLLMs miss what humans read: diagnosing and learning to read squint text
Squint OPSD lifts multimodal LLM reading of hard-to-see image text from 26.50 to 88.61 ANLS.
USN-8882-1: Tesseract vulnerabilities
Ubuntu’s USN-8882-1 fixes Tesseract flaws in crafted .traineddata models that can crash the engine or execute code.
PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligence
PolyOCR introduces multi-scale OCR foundation models and a corrected OCRBench, reporting competitive document and scene-text results.
PerceptFence: Content-Mediation Architecture and Deterministic Coverage for Screen-Share AI Assistants
PerceptFence mediates screen-share AI assistant capture to redact secrets, beating Presidio on OCR-surviving PII in synthetic tests.
Experts Alarmed Over Gyazo’s Breach of 490 Million Metadata Records
Attackers exploited a Gyazo upload-server vulnerability, exposing 24 million user records and 490 million metadata records including IPs, EXIF locations and OCR text.
All-in-One Multilingual Scene Text Recognition with Script-aware Mixture-of-Experts
ScriptMoE recognizes scene text across 229 languages, beating per-script models and a vision-language baseline.
HHS’ Office for Civil Rights Settles HIPAA Investigation of Ambry Genetics for Security Rule Violations
HHS Office for Civil Rights settled a HIPAA Security Rule investigation with California genetic testing firm Ambry Genetics.
Jina AI Releases jina-ocr-v1: A 3.4B MoE Document Parser With Built-In Speculative Decoding for Low-Budget GPUs
Jina AI open-sourced jina-ocr-v1, a 3.4B MoE OCR model with speculative decoding that parses documents to Markdown at 2.57 pages/second on an A100.
6
WeVisDoc: From Coverage to Capability for Robust End-to-End Document Parsing
WeVisDoc two-stage data framework yields WeVisDoc-4B parser scoring 95.38 on OmniDocBench v1.6 and ranking first across four end-to-end parsing settings.
Tables Decoded: DELTA for Structure, TARQA for Understanding
DELTA extracts tables into compact OTSL text and TARQA fine-tunes LLMs on it, beating VLM baselines on table QA.
HHS Releases Updated Security Risk Assessment Tool
HHS OCR and ONC released version 3.7 of the Security Risk Assessment Tool for healthcare organizations.
LandingAI Releases Agentic Document Extraction Gen2 with DPT-3 Pro and DPT-3 Verity
LandingAI shipped Agentic Document Extraction Gen2 with DPT-3 Pro and DPT-3 Verity parsing models, adding usage-based billing, block-tree outputs, and word-level grounding.
2