Document Intelligence · research & products
Document AI that reads — and remembers.
I build systems that understand visually rich documents, ground every answer in evidence on the page, and carry what they read forward as persistent, provenance-linked state, so the amendment that arrives tomorrow is read in the light of the contract signed today.
- Now
- IRISA · ESIR · Université de Rennes
- Before
- Associate Professor, L3i, La Rochelle Université (2023–2025)
- PhD
- La Rochelle Université, 2022 (European Doctorate label)
- Company
- Co-founder, SynaptYon SARL
Contract · MSA-2026-014 · 2026-03-02
MASTER SERVICES AGREEMENT
1. Parties. This Agreement is made between Halden Freight SAS (the “Supplier”) and Orvelle Analytics SA (the “Client”).
2. Term. Twenty-four (24) months from 2 March 2026.
4.2 Payment terms. Invoices are payable within thirty (30) days of receipt.
| Service | Unit | Rate |
|---|---|---|
| Standard pallet | pallet | €42.00 |
| Express pallet | pallet | €68.00 |
| Storage | pallet-week | €3.10 |
7.1 Liability. The Supplier’s aggregate liability is capped at €250,000 per contract year.
9. Governing law. This Agreement is governed by French law; courts of Rennes.
What a stateful system notes when the amendment and the invoice arrive
- Supersede Payment terms: 30 days → 45 days from 2026-11-01 (AMD-1 §1)
- Supersede Express pallet rate: €68.00 → €71.50 from 2026-11-01 (AMD-1 §2)
- Support Liability cap: unchanged, €250,000 / year stands (AMD-1 §3)
- Retain conflict Express pallet rate on the invoice: €68.00 vs €71.50 in force (INV-2026-0117)
- Escalate for review Payment terms on the invoice: 30 days vs 45 days in force (INV-2026-0117)
§ 01 The problem
Today’s document AI has amnesia, by design.
Models now read contracts, extract fields and point to the page region behind each answer — then forget all of it. My 2026 survey with Ayoub Merimi names this the statelessness bottleneck and specifies what it takes to remove it.
“Storage is not retrieval, and retrieval is not consolidation.”
St = (Et, Kt, Γt, 𝒱t, Pt, At)
§ 02 From papers to products
One stack, from the first read to proof that memory helps.
The memory layer of the SynaptYon chain, as the 2026 survey specifies it. Each layer implements a part of the paper, builds on published work, and maps to one of the memory products. Pick a layer to see the research behind it.
-
Acquire Read the whole document
Extract the multimodal observation: text, layout coordinates, table structure and visual elements, for questions not yet asked.
- Paper
- Observe · Dt = (V, T, B, G, H, M)
- Product
- Mnemos In development
- Research
-
- GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification WACV 2025
- VLCDoC: Vision-Language Contrastive Pre-Training Model for Cross-Modal Document Classification Pattern Recognition 2023
- EAML: Ensemble Self-Attention-Based Mutual Learning Network for Document Image Classification IJDAR 2021
- Multimodal Adaptive Inference for Document Image Classification with Anytime Early Exiting ICDAR 2024
- WildKhmerST: A Comprehensive Dataset and Benchmark for Khmer Scene Text Detection and Recognition in the Wild ICDAR 2025
- BHCP-DOC: A Real-World Bilingual Dataset for Handwritten Clinical Prescription Detection and Recognition ICPR Workshops 2026
-
Validate Gate before writing
Corroborate each observation against the provenance ledger and reject it when confidence is below threshold, leaving state unchanged.
- Paper
- Pre-write gate · Theorem III.4
- Product
- Argus In development
- Research
-
Consolidate Write and abstract
Append admitted evidence with its timestamp and provenance, then abstract recurring families into reusable schemas.
- Paper
- Et, Kt, Γt · max J(α) − λR(α)
- Product
- Mnemos In development
- Research
-
Reconcile Know what is in force
Resolve conflicts with one of five outcomes and close superseded claims at their effective time, keeping history for audit.
- Paper
- Temporal supersession · 𝒱t
- Product
- Chronos In development
- Research
-
Trace Cite the box
Keep the chain from every conclusion back to document, region and version, through consolidation, supersession and repair.
- Paper
- Provenance invariant · Pt
- Product
- Ariadne In development
- Research
-
Forget Erase, and prove it
Remove a target across every store, repair the schemas that depended on it, and certify what can no longer be recovered.
- Paper
- Five-level unlearning · Algorithm 2
- Product
- Lethe In development
- Research
-
Measure Show that memory helps
Compare warm, retrieval-only and cold arms on the same future documents and report benefit, cost, risk and governability together.
- Paper
- Longitudinal harness · five metrics
- Product
- Anamnesis In development
- Research
§ 03 Products
The whole chain, from ingestion to answers.
SynaptYon SARL, which I co-founded, builds Document Intelligence for business, financial, historical and healthcare documents: four products that read, extract, answer and remember. None is generally available yet.
-
Grammata
Read. Every page, read
-
Kleis
Extract. The data you need, from any document
-
Sibyl
Ask. Answers you can check
-
Mnemos
Remember. Better with every document
Mnemos is the memory layer specified in my 2026 survey; the products page shows them in research detail, with demos.
§ 04 Research trajectory
From fusing text and image to remembering documents.
Since 2019: cross-modal document understanding, then retrieval and grounded answers, low-resource scripts, and now persistent state. Filter by theme.
2019 1 publication
- Face Detection in Camera Captured Images of Identity Documents Under Challenging Conditions ICDAR Workshops 2019
- PhD researcher, L3i PhD Researcher and Lecturer, L3i-Lab, La Rochelle Université
2020 2 publications
2021 1 publication
2022 1 publication
- PhD Ph.D. in Computer Science and Applications, La Rochelle Université, with the European Doctorate Label
- Multimodal Document Understanding with Unified Vision and Language Cross-Modal Learning PhD thesis 2022
2023 2 publications
- Postdoc, L3i Postdoctoral Researcher and Lecturer, L3i-Lab, La Rochelle Université
- TransferDoc: A Self-Supervised Transferable Document Representation Learning Model Unifying Vision and Language SSRN 2023
- Featured: VLCDoC: Vision-Language Contrastive Pre-Training Model for Cross-Modal Document Classification Pattern Recognition 2023
- Associate Professor, L3i Associate Professor (Maître de Conférences), L3i-Lab, La Rochelle Université
2024 4 publications
- State-of-the-Art Khmer Text Recognition Using Deep Learning Models ACET 2024
- LLMChain: Blockchain-Based Reputation System for Sharing and Evaluating Large Language Models COMPSAC 2024
- Featured: Multimodal Adaptive Inference for Document Image Classification with Anytime Early Exiting ICDAR 2024
- Featured: KhmerST: A Low-Resource Khmer Scene Text Detection and Recognition Benchmark ACCV 2024
2025 10 publications
- Featured: GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification WACV 2025
- DocSum: Domain-Adaptive Pre-training for Document Abstractive Summarization WACV Workshops 2025
- IDTrust: Deep Identity Document Quality Detection with Bandpass Filtering WACV Workshops 2025
- Fusion of GNN and GBDT Models for Graph and Node Classification GbRPR 2025
- Evaluating the Impact of Khmer Font Types on Text Recognition arXiv 2025
- Confidence-Based Knowledge Distillation to Reduce Training Costs and Carbon Footprint for Low-Resource Neural Machine Translation Applied Sciences 2025
- IRISA, Université de Rennes Enseignant-Chercheur Contractuel, IRISA-Lab, ESIR, Université de Rennes (SHADOC team)
- Visual Text Generation in Khmer Language: Challenges and Trends with Diffusion Models ICDAR Workshops 2025
- Cross-Lingual Learning for Low-Resource Khmer Scene Text Detection and Recognition ICDAR Workshops 2025
- Featured: WildKhmerST: A Comprehensive Dataset and Benchmark for Khmer Scene Text Detection and Recognition in the Wild ICDAR 2025
- Featured: MTRACE: Multilingual Retrieval-Augmented Generation for Temporally Diverse Text Corpora arXiv 2025
2026 8 publications
- Cross-Lingual Transfer for Complex Scripts: A Benchmark on End-to-End Khmer Scene Text Spotting WACV Workshops 2026
- Investigating Model Compression for Neural Machine Translation in the Biomedical Domain AICS 2025
- Terminology-Aware Retrieval-Augmented Knowledge Distillation for Biomedical Neural Machine Translation EAMT 2026
- Featured: Amnesia by Design, Memory By Necessity: Persistent State for Document Intelligence arXiv 2026
- HSB-GDL: A Heterogeneous Subgraph Benchmark for Classification and Query Spotting IJDAR 2026
- BHCP-DOC: A Real-World Bilingual Dataset for Handwritten Clinical Prescription Detection and Recognition ICPR Workshops 2026
- Featured: Fresh at Decision, Stale at Action: Routing New Information to LLM Decisions Already Made arXiv 2026
- An Empirical Study of Recent Large Language Models for Knowledge Graph to Text Generation on Standard Benchmarks arXiv 2026
§ 05 Selected publications
Selected work.
-
arXiv 2026Preprint
Fresh at Decision, Stale at Action: Routing New Information to LLM Decisions Already Made
A timestamp theory of LLM decision validity and a decision ledger that realises it. Six timestamps define a decision, a freshness ceiling bounds what up-to-date evidence can guarantee at action time, and accuracy factorises into routing new information to affected decisions and judging it for the time of action. Controlled decision streams confirm four predictions across five LLMs.
-
arXiv 2026Preprint
Amnesia by Design, Memory By Necessity: Persistent State for Document Intelligence
A survey that names the statelessness bottleneck of Document AI, formalises persistent evidence-grounded document state, audits ten benchmarks against eight statefulness criteria, and derives a longitudinal harness with five counterfactual metrics: Experience Gain, Cost Efficiency, Memory Harm, Forgetting Fidelity and Coverage Retention.
-
arXiv 2025Preprint
MTRACE: Multilingual Retrieval-Augmented Generation for Temporally Diverse Text Corpora
A modular multilingual RAG pipeline evaluated on the French and English subsets of MIRACL: semantic query expansion with reciprocal rank fusion, a generation prompt that enforces strict grounding and explicit abstention, and ablations on entity extraction and retriever choice. It gives faithful answers for well-supported queries and abstains correctly on unanswerable ones.
-
ICDAR 2025Conference
WildKhmerST: A Comprehensive Dataset and Benchmark for Khmer Scene Text Detection and Recognition in the Wild
A large-scale dataset of Khmer scene text captured in real-world environments, with 29,601 annotated text lines from 10,000 images, and a benchmark for detection and recognition.
-
WACV 2025Conference
GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification
A cross-modal transformer pre-trained with three self-supervised pretext tasks that unify language and vision, reducing reliance on OCR positional information. It introduces two industrial-style tasks: few-shot document image classification and content-based document image retrieval.
-
ACCV 2024Conference
KhmerST: A Low-Resource Khmer Scene Text Detection and Recognition Benchmark
A benchmark for detecting and recognising scene text in Khmer, a low-resource script with complex character structure.
§ 06 Work together
Research collaboration, or a product pilot.
Academic
Co-supervision, joint projects, reviewing
I work with groups across Europe and Asia on document understanding, low-resource scripts and grounded retrieval, and co-supervise PhD students.
SynaptYon SARL
Document Intelligence for your documents
If your team handles business, financial, historical or healthcare documents, from invoices and contracts to registers and clinical reports, we would like to hear about it. None of the products is generally available yet; early pilots shape them. Write to sbakkali.sirajai@gmail.com.