ServiceNow AI Research

1

Centering Knowledge Along the Responsible LLM Supply Chain: An Empirical Study & Multi-Stakeholder Taxonomy
Centering Knowledge Along the Responsible LLM Supply Chain: An Empirical Study & Multi-Stakeholder Taxonomy
Framing LLMs as products of complex supply chains rather than monolithic entities facilitates the creation of nuanced approaches to …
Computer-Supported Cooperative Work and Social Computing (CSCW),  2026.
Cite
Privileged Information Distillation for Language Models
Privileged Information Distillation for Language Models
Training-time privileged information (PI) can enable language models to succeed on tasks they would otherwise fail, making it a …
International Conference on Machine Learning (ICML),  2026.
Paper Cite
GitChameleon: Evaluating AI Code Generation Against Python Library Version Incompatibilities
GitChameleon: Evaluating AI Code Generation Against Python Library Version Incompatibilities
The rapid evolution of software libraries presents a significant challenge for code generation models, which must adapt to frequent …
Annual Meeting of the Association for Computational Linguistics (ACL),  2026.
Paper Cite Code
Societal Alignment Frameworks Can Improve LLM Alignment
Societal Alignment Frameworks Can Improve LLM Alignment
Recent progress in large language models (LLMs) has focused on producing responses that meet human expectations and align with shared …
ACM Conference on Fairness, Accountability, and Transparency,  2026.
Paper Cite
Augmenting LLM Reasoning with Dynamic Notes Writing for Complex QA
Augmenting LLM Reasoning with Dynamic Notes Writing for Complex QA
Iterative RAG for multi-hop question answering faces challenges with lengthy contexts and the buildup of irrelevant information. This …
Language Resources and Evaluation Conference,  2026.
Paper Cite
DRBench: A Realistic Benchmark for Enterprise Deep Research
DRBench: A Realistic Benchmark for Enterprise Deep Research
We introduce DRBench, a benchmark for evaluating AI agents on complex, open-ended deep research tasks in enterprise settings. Unlike …
International Conference on Learning Representations,  2026.
Paper Cite Code
Grounding Computer Use Agents on Human Demonstrations
Grounding Computer Use Agents on Human Demonstrations
Building reliable computer-use agents requires grounding: accurately connecting natural language instructions to the correct on-screen …
International Conference on Learning Representations,  2026.
Paper Cite Code
No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms
No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms
Leading language model (LM) providers like OpenAI and Anthropic allow customers to fine-tune frontier LMs for specific use cases. To …
International Conference on Learning Representations,  2026.
Paper Cite Code
Learning a Spatial Partitioning and its Causal Relations from Temporal Data
Scientific research often seeks to understand the causal structure underlying high-level variables in a system. For example, climate …
Causal Learning and Reasoning (CLeaR),  2026.
Paper Cite
StarFlow: Generating Structured Workflow Outputs From Sketch Images
StarFlow: Generating Structured Workflow Outputs From Sketch Images
Workflows are a fundamental component of automation in enterprise platforms, enabling the orchestration of tasks, data processing, and …
European Chapter of the Association for Computational Linguistics (EACL),  2026.
Paper Cite Code Video