Papers
Publications from the Samaya Research team pushing the boundaries of Financial AI research.
Highlighted Paper
Lost in the Middle: How Language Models Use Long Contexts
Language models use information best when it sits at the start or end of the input, with accuracy dropping sharply for facts buried in the middle of long contexts.
Read Paper →ACM CAIS · 2026
OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction
ARXIV · 2025
Lost in the Maze: Overcoming Context Limitations in Long-Horizon Agentic Search
ICLR · 2024
Promptriever: Instruction-Trained Retrievers Can Be Prompted Like Language Models
EMNLP · 2024
Dancing in Chains: Reconciling Instruction Following and Faithfulness in Language Models
EMNLP · 2024
RAG-QA Arena: Evaluating Domain Robustness for Long-Form Retrieval-Augmented Question Answering
ACL FINDINGS · 2024
Selective ‘Selective Prediction’: Reducing Unnecessary Abstention in Vision-Language Reasoning
NAACL · 2024
When Does Monolingual Data Help Multilingual Translation: The Role of Domain and Model Scale
TACL · 2024
Lost in the Middle: How Language Models Use Long Contexts
ACL FINDINGS · 2024
Deal, or No Deal (or Who Knows)? Forecasting Uncertainty in Conversations Using Large Language Models
NATURE MACHINE INTELLIGENCE · 2023
Improving Wikipedia Verifiability With AI
CVPR · 2023
Cross-Domain Image Captioning With Discriminative Finetuning
ICLR · 2023