Paper: arXiv 2602.07020

Authors: Amin Haeri, Mahdi Ghelichi, Nishant Agrawal, David Li, Catalina Gomez Sanchez

Abstract

Finding similar bonds remains challenging in fixed-income analytics, as numerical financial attributes often overshadow categorical non-financial ones such as issuer sector and domicile. This paper shows that these categorical attributes dominate the predictability of spread curves and proposes embedding models to capture their semantic similarities, outperforming one-hot and many other baselines. Evaluated via sparse-issuer augmentation, the approach improves risk modeling and curve construction.

Complexity vs Empirical Score

  • Math Complexity: 3.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper presents an applied representation learning approach to fixed-income analytics, shifting focus from pure numerical features to high-cardinality non-financial categorical attributes. While it lacks heavy mathematical derivations, it addresses a practical market problem with reasonable empirical motivation via sparse-issuer augmentation testing.

Research Flowchart

  flowchart TD
    A[Goal: Improve Bond Similarity Search for Yield/Spread Curves] --> B[Inputs: Categorical Issuer Attributes & Numerical Financial Data]
    B --> C[Methodology: Representation Learning via Embedding Models]
    C --> D[Process: Vector Embeddings vs. One-Hot & Baseline Models]
    D --> E[Evaluation: Sparse-Issuer Augmentation & Curve Construction]
    E --> F[Outcomes: Categorical Attributes Dominate Predictability & Embeddings Outperform Baselines]

Cited References (65)

  1. Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks (2020) — link
  2. Attention Is All You Need (2023) — link
  3. Quantum Models of Cognition and Decision (2012)
  4. On Robustness of Mutual Funds Categorization and Distance Metric Learning (2021) — link
  5. Quantifying Uncertainty in Random Forests via Confidence Intervals and Hypothesis Tests (2016) — link
  6. CONSISTENCY FOR A SIMPLE MODEL OF RANDOM FORESTS (2004) — link
  7. Consistency of random forests (2015) — link
  8. Exploring Simple Siamese Representation Learning (2020) — link
and 57 more
  1. Distance Metric Learning for Large Margin Nearest Neighbor Classification (2005) — link
  2. Distance Metric Learning with Application to Clustering with Side-Information (2002) — link
  3. Deep Learning (2016)
  4. Deep learning for finance: deep portfolios (2017) — link
  5. Learning to Rank for Information Retrieval and Natural Language Processing, Second Edition (2014) — link
  6. Metric Learning : A Survey By (2013) — link
  7. Representation Learning: A Review and New Perspectives (2013) — link
  8. Supervised Similarity for High-Yield Corporate Bonds with Quantum Cognition Machine Learning (2025) — link
  9. Supervised similarity learning for corporate bonds using Random Forest proximities (2022) — link
  10. Generative AI Enhanced Financial Risk Management Information Retrieval (2025) — link
  11. Applying BERT to Document Retrieval with Birch (2019) — link
  12. Passage Re-ranking with BERT (2020) — link
  13. ex BERT : Extending Pre-trained Models with Domain-specific Vocabulary Under Constrained Training Resources (2020) — link
  14. HuggingFace’s Transformers: State-of-the-art Natural Language Processing (2020) — link
  15. NV-Retriever: Improving text embedding models with effective hard-negative mining (2025) — link
  16. Dense Passage Retrieval for Open-Domain Question Answering (2020) — link
  17. Learning Discriminative Projections for Text Similarity Measures (2011) — link
  18. Efficient Natural Language Response Suggestion for Smart Reply (2017) — link
  19. Learning Dense Representations for Entity Retrieval (2019) — link
  20. Arctic-Embed: Scalable, Efficient, and Accurate Text Embedding Models (2024) — link
  21. BioBERT: a pre-trained biomedical language representation model for biomedical text mining (2019) — link
  22. RoBERTa: A Robustly Optimized BERT Pretraining Approach (2019) — link
  23. DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter (2020) — link
  24. Distributed Representations of Words and Phrases and their Compositionality (2013) — link
  25. Deep contextualized word representations (2018) — link
  26. Basel III: A global regulatory framework for more resilient banks and banking systems (2019) — link
  27. Basel III monitoring report (2017) — link
  28. Financial Institutions Management: A Risk Management Approach (2024)
  29. Financial Risk Manager Handbook (2009)
  30. Risk Management and Financial Institutions (2023)
  31. Financial Sentiment Analysis: Techniques and Applications (2024) — link
  32. Text-based sentiment analysis in finance: Synthesising the existing literature and exploring future directions (2024) — link
  33. Predicting stock market using natural language processing (2023) — link
  34. BERT-based Financial Sentiment Index and LSTM-based Stock Return Predictability (2019) — link
  35. MEMORYLLM : Towards Self-Updatable Large Language Models (2024) — link
  36. Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey (2025) — link
  37. Attention is All you Need (2017) — link
  38. A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions (2025) — link
  39. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks (2019) — link
  40. BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding (2019) — link
  41. FACTS About Building Retrieval Augmented Generation-based Chatbots (2024) — link
  42. A Comprehensive Survey of Retrieval-Augmented Generation (RAG): Evolution, Current Landscape and Future Directions (2024) — link
  43. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (2021) — link
  44. Retrieval-Augmented Generation for Large Language Models: A Survey (2024) — link
  45. A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Models (2024) — link
  46. Reducing hallucination in structured outputs via Retrieval-Augmented Generation (2024) — link
  47. Trade the Event: Corporate Events Detection for News-Based Event-Driven Trading (2021) — link
  48. FinQABench (2025) — link
  49. FinanceBench: A New Benchmark for Financial Question Answering (2023) — link
  50. TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance (2021) — link
  51. Finance Alpaca (2023) — link
  52. FIQA Personal Finance Dataset (2024) — link
  53. Stock News Sentiments (2024) — link
  54. ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering (2022) — link
  55. FinQA: A Dataset of Numerical Reasoning over Financial Data (2022) — link
  56. News Stocks (2024) — link
  57. How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection (2023) — link

0 of 65 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.