Decimal Point Analytics Launches Bindu FRS 1.0 for Financial Statement Analysis with 98.79% Accuracy

PNN
Mumbai (Maharashtra) [India], May 26: Decimal Point Analytics (DPA), a global provider of AI-powered analytics, digital transformation, and consulting solutions, today announced the launch of Arka-Bindu (AKA Bindu FRS 1.0) -- a proprietary, distilled GenAI model built and trained entirely in-house for the analysis and standardisation of company financial statements. Bindu FRS 1.0 is India's first GenAI model purpose-built for financial statement analysis and standardisation - a category distinct from horizontal Indian foundation models, BFSI workflow copilots, and payments-operations models, none of which is engineered for the specific task of converting unstructured financial disclosures into standardised, machine-readable data.
In controlled testing across 100+ company financial filings - comprising annual reports, quarterly reports, and half-yearly reports from UK and Canadian issuers - Bindu FRS 1.0 achieved 98.79% first-pass accuracy on financial statement mapping, a 77% reduction in first-pass errors versus a frontier commercial LLM baseline on the same standardization task, with materially reduced hallucination incidence on newly reported data points.
Bindu FRS 1.0 is purpose-built for one domain: converting unstructured company financial disclosures - annual reports, interim filings, and prospectuses - into standardized, machine-readable data. It is operational today across DPA's managed data services for financial data aggregators, rating agencies, and research providers.
DPA's Bindu FRS 1.0 offers three structural advantages over frontier commercial LLMs for this task. First, its domain specialization enables it to be trained exclusively on financial statement data, allowing the model to avoid the pattern-matching failures that general-purpose LLMs often exhibit when mapping novel line items in company reports. Second, it features continuous self-improvement, meaning that when a mapping error is identified and corrected by a DPA analyst, the model learns from that correction. Commercial API-based LLMs do not provide an equivalent mechanism, as their learning processes remain outside the user's control. Third, Bindu FRS 1.0 incurs no per-query API or licensing fees. Since it is deployed on DPA infrastructure, it operates without usage-based pricing, creating a structural cost advantage for high-volume financial data workflows.
Shailesh Dhuri, CEO, Decimal Point Analytics says:
"India has built strong credibility in horizontal AI. The next frontier is vertical AI - models built deep into a single domain, with proprietary data as the moat and continuous improvement as the engine. Bindu FRS 1.0 is our entry into that frontier for the specific task of financial statement analysis and standardisation. Frontier commercial LLMs are extraordinary general-purpose tools, but for high-volume, high-accuracy, closed-domain work - like mapping the financial disclosures of tens of thousands of companies - general-purpose is the wrong architecture. This is the first of what we expect to be a family of Arka-Bindu models across specialised financial domains."
Shyam Pardeshi, Chief Solutions Officer, Decimal Point Analytics, says:
"Commercial LLMs hallucinate most on the data points that matter most - the newly reported, non-standard line items in real company filings. Bindu FRS 1.0 removes that failure mode by narrowing the model's world to a single well-bounded task and teaching it from corrections, not from scale. The accuracy curve keeps improving; the unit cost stays flat. That is what makes this model commercially different, not just technically different."
Financial data aggregators, rating agencies, equity and credit research providers, and buy-side research teams process tens of thousands of financial statements annually. Error rates compound into downstream data quality issues, regulatory exposure, and manual review costs. Bindu FRS 1.0 targets that workflow directly - operational in production today.
About Decimal Point Analytics
Decimal Point Analytics is an AI-native digital transformation company headquartered in Mumbai, serving financial services, healthcare, CPG, and manufacturing clients across North America, India, the GCC, the UK, and Europe. DPA's capabilities span data & AI, digital modernization, cloud migration, workflow automation, and managed data operations. The company is a recipient of the Aegis Graham Bell Award for Innovation in GenAI - BFSI.
Please visit https://decimalpointanalytics.com/contact-us for more information.
(ADVERTORIAL DISCLAIMER: The above press release has been provided by PNN. ANI will not be responsible in any way for the content of the same.)