Papersby GlyphOfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning research paper by Databricks, 2026
Databricks · Agents and evaluation · Mar 9, 2026 · 48 citations · 3 upvotes · unverified
What it shows
OfficeQA Pro evaluates AI agents on multi-document reasoning across historical financial documents, revealing persistent challenges in grounded reasoning despite advanced model capabilities.
By Krista Opsahl-Ong, Arnav Singhvi, Jasmine Collins and 10 more · arXiv 2603.08655 · PDF · Code
UnverifiedHugging Face's summary; not yet checked by hand.
More from Databricks
All 3KARL: Knowledge Agents via Reinforcement LearningA reinforcement learning system for enterprise search agents achieves superior performance through diverse training data generation and multi-task learning approaches.
Databricks2 citations7 upvotesRead →LoRA Learns Less and Forgets LessLoRA learns less than full fine-tuning on code and math, but forgets less of what the model already knew.Databricks486 citations91 upvotesRead →
Other agents and evaluation papers
TopicGemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic CapabilitiesGemini 2.X model family, including Gemini 2.5 Pro and Flash, offers superior coding, reasoning, and multimodal understanding capabilities across a range of computational efficiencies.
Google4,594 citations67 upvotesRead →DeepSeek-V3.2: Pushing the Frontier of Open Large Language ModelsDeepSeek-V3.2 introduces DeepSeek Sparse Attention and a scalable reinforcement learning framework, achieving superior reasoning and performance compared to GPT-5 and Gemini-3.0-Pro in complex reasoning tasks.
DeepSeek887 citations275 upvotesRead →WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?WorkArena and BrowserGym evaluate large language model-based agents' ability to perform enterprise software tasks, revealing gaps in current agent capabilities and differences between open and closed-source LLMs.
ServiceNow443 citations2 upvotesRead →Aya Model: An Instruction Finetuned Open-Access Multilingual Language ModelAya, a multilingual generative language model supporting over 50% lower-resourced languages, excels in both generative and discriminative tasks across 99 languages, and provides extensive evaluations and open-source resources.
Cohere438 citations48 upvotesRead →Rubrics as Rewards: Reinforcement Learning Beyond Verifiable DomainsRubrics as Rewards (RaR) framework uses structured rubrics as interpretable reward signals for on-policy training, improving performance in real-world reinforcement learning tasks with subjective criteria.
Scale AI431 citations6 upvotesRead →SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?SWE-Bench Pro is a challenging benchmark for coding models, featuring complex, enterprise-level problems that require substantial code modifications, with performance evaluations showing significant limitations in current models.Scale AI367 citations21 upvotesRead →
About this paper
- Venue
- arXiv.org
- Citations
- 48, 6 influential · Semantic Scholar
- Upvotes
- 3 · Hugging Face
- Lab
- Databricks · on Companies · on Acquisitions · on Rounds · on Paydays · on Releases · on TechConf
Changes
Influential citations rose from 5 to 6.
Citations rose from 43 to 48.
A first count of 5 influential citations.
A first count of 43 citations.
Found by the weekly Hugging Face scan; not yet checked by hand.